Source code
Revision control
Copy as Markdown
Other Tools
diff --git a/src/llama-arch.cpp b/src/llama-arch.cpp
index 18dcc6ddfe..335eff7976 100644
--- a/src/llama-arch.cpp
+++ b/src/llama-arch.cpp
@@ -3,2255 +3,2368 @@
#include "llama-impl.h"
#include <map>
-
-static const std::map<llm_arch, const char *> LLM_ARCH_NAMES = {
- { LLM_ARCH_LLAMA, "llama" },
- { LLM_ARCH_LLAMA4, "llama4" },
- { LLM_ARCH_DECI, "deci" },
- { LLM_ARCH_FALCON, "falcon" },
- { LLM_ARCH_GROK, "grok" },
- { LLM_ARCH_GPT2, "gpt2" },
- { LLM_ARCH_GPTJ, "gptj" },
- { LLM_ARCH_GPTNEOX, "gptneox" },
- { LLM_ARCH_MPT, "mpt" },
- { LLM_ARCH_BAICHUAN, "baichuan" },
- { LLM_ARCH_STARCODER, "starcoder" },
- { LLM_ARCH_REFACT, "refact" },
- { LLM_ARCH_BERT, "bert" },
- { LLM_ARCH_NOMIC_BERT, "nomic-bert" },
- { LLM_ARCH_NOMIC_BERT_MOE, "nomic-bert-moe" },
- { LLM_ARCH_NEO_BERT, "neo-bert" },
- { LLM_ARCH_JINA_BERT_V2, "jina-bert-v2" },
- { LLM_ARCH_BLOOM, "bloom" },
- { LLM_ARCH_STABLELM, "stablelm" },
- { LLM_ARCH_QWEN, "qwen" },
- { LLM_ARCH_QWEN2, "qwen2" },
- { LLM_ARCH_QWEN2MOE, "qwen2moe" },
- { LLM_ARCH_QWEN2VL, "qwen2vl" },
- { LLM_ARCH_QWEN3, "qwen3" },
- { LLM_ARCH_QWEN3MOE, "qwen3moe" },
- { LLM_ARCH_PHI2, "phi2" },
- { LLM_ARCH_PHI3, "phi3" },
- { LLM_ARCH_PHIMOE, "phimoe" },
- { LLM_ARCH_PLAMO, "plamo" },
- { LLM_ARCH_PLAMO2, "plamo2" },
- { LLM_ARCH_CODESHELL, "codeshell" },
- { LLM_ARCH_ORION, "orion" },
- { LLM_ARCH_INTERNLM2, "internlm2" },
- { LLM_ARCH_MINICPM, "minicpm" },
- { LLM_ARCH_MINICPM3, "minicpm3" },
- { LLM_ARCH_GEMMA, "gemma" },
- { LLM_ARCH_GEMMA2, "gemma2" },
- { LLM_ARCH_GEMMA3, "gemma3" },
- { LLM_ARCH_GEMMA3N, "gemma3n" },
- { LLM_ARCH_STARCODER2, "starcoder2" },
- { LLM_ARCH_MAMBA, "mamba" },
- { LLM_ARCH_MAMBA2, "mamba2" },
- { LLM_ARCH_JAMBA, "jamba" },
- { LLM_ARCH_FALCON_H1, "falcon-h1" },
- { LLM_ARCH_XVERSE, "xverse" },
- { LLM_ARCH_COMMAND_R, "command-r" },
- { LLM_ARCH_COHERE2, "cohere2" },
- { LLM_ARCH_DBRX, "dbrx" },
- { LLM_ARCH_OLMO, "olmo" },
- { LLM_ARCH_OLMO2, "olmo2" },
- { LLM_ARCH_OLMOE, "olmoe" },
- { LLM_ARCH_OPENELM, "openelm" },
- { LLM_ARCH_ARCTIC, "arctic" },
- { LLM_ARCH_DEEPSEEK, "deepseek" },
- { LLM_ARCH_DEEPSEEK2, "deepseek2" },
- { LLM_ARCH_CHATGLM, "chatglm" },
- { LLM_ARCH_GLM4, "glm4" },
- { LLM_ARCH_GLM4_MOE, "glm4moe" },
- { LLM_ARCH_BITNET, "bitnet" },
- { LLM_ARCH_T5, "t5" },
- { LLM_ARCH_T5ENCODER, "t5encoder" },
- { LLM_ARCH_JAIS, "jais" },
- { LLM_ARCH_NEMOTRON, "nemotron" },
- { LLM_ARCH_EXAONE, "exaone" },
- { LLM_ARCH_EXAONE4, "exaone4" },
- { LLM_ARCH_RWKV6, "rwkv6" },
- { LLM_ARCH_RWKV6QWEN2, "rwkv6qwen2" },
- { LLM_ARCH_RWKV7, "rwkv7" },
- { LLM_ARCH_ARWKV7, "arwkv7" },
- { LLM_ARCH_GRANITE, "granite" },
- { LLM_ARCH_GRANITE_MOE, "granitemoe" },
- { LLM_ARCH_GRANITE_HYBRID, "granitehybrid" },
- { LLM_ARCH_CHAMELEON, "chameleon" },
- { LLM_ARCH_WAVTOKENIZER_DEC, "wavtokenizer-dec" },
- { LLM_ARCH_PLM, "plm" },
- { LLM_ARCH_BAILINGMOE, "bailingmoe" },
- { LLM_ARCH_DOTS1, "dots1" },
- { LLM_ARCH_ARCEE, "arcee" },
- { LLM_ARCH_ERNIE4_5, "ernie4_5" },
- { LLM_ARCH_ERNIE4_5_MOE, "ernie4_5-moe" },
- { LLM_ARCH_HUNYUAN_MOE, "hunyuan-moe" },
- { LLM_ARCH_HUNYUAN_DENSE, "hunyuan-dense" },
- { LLM_ARCH_SMOLLM3, "smollm3" },
- { LLM_ARCH_OPENAI_MOE, "gpt-oss" },
- { LLM_ARCH_LFM2, "lfm2" },
- { LLM_ARCH_DREAM, "dream" },
- { LLM_ARCH_SMALLTHINKER, "smallthinker" },
- { LLM_ARCH_LLADA, "llada" },
- { LLM_ARCH_UNKNOWN, "(unknown)" },
-};
-
-static const std::map<llm_kv, const char *> LLM_KV_NAMES = {
- { LLM_KV_GENERAL_TYPE, "general.type" },
- { LLM_KV_GENERAL_ARCHITECTURE, "general.architecture" },
- { LLM_KV_GENERAL_QUANTIZATION_VERSION, "general.quantization_version" },
- { LLM_KV_GENERAL_ALIGNMENT, "general.alignment" },
- { LLM_KV_GENERAL_FILE_TYPE, "general.file_type" },
- { LLM_KV_GENERAL_NAME, "general.name" },
- { LLM_KV_GENERAL_AUTHOR, "general.author" },
- { LLM_KV_GENERAL_VERSION, "general.version" },
- { LLM_KV_GENERAL_URL, "general.url" },
- { LLM_KV_GENERAL_DESCRIPTION, "general.description" },
- { LLM_KV_GENERAL_LICENSE, "general.license" },
- { LLM_KV_GENERAL_SOURCE_URL, "general.source.url" },
- { LLM_KV_GENERAL_SOURCE_HF_REPO, "general.source.huggingface.repository" },
-
- { LLM_KV_VOCAB_SIZE, "%s.vocab_size" },
- { LLM_KV_CONTEXT_LENGTH, "%s.context_length" },
- { LLM_KV_EMBEDDING_LENGTH, "%s.embedding_length" },
- { LLM_KV_FEATURES_LENGTH, "%s.features_length" },
- { LLM_KV_BLOCK_COUNT, "%s.block_count" },
- { LLM_KV_LEADING_DENSE_BLOCK_COUNT, "%s.leading_dense_block_count" },
- { LLM_KV_FEED_FORWARD_LENGTH, "%s.feed_forward_length" },
- { LLM_KV_EXPERT_FEED_FORWARD_LENGTH, "%s.expert_feed_forward_length" },
- { LLM_KV_EXPERT_SHARED_FEED_FORWARD_LENGTH, "%s.expert_shared_feed_forward_length" },
- { LLM_KV_USE_PARALLEL_RESIDUAL, "%s.use_parallel_residual" },
- { LLM_KV_TENSOR_DATA_LAYOUT, "%s.tensor_data_layout" },
- { LLM_KV_EXPERT_COUNT, "%s.expert_count" },
- { LLM_KV_EXPERT_USED_COUNT, "%s.expert_used_count" },
- { LLM_KV_EXPERT_SHARED_COUNT, "%s.expert_shared_count" },
- { LLM_KV_EXPERT_WEIGHTS_SCALE, "%s.expert_weights_scale" },
- { LLM_KV_EXPERT_WEIGHTS_NORM, "%s.expert_weights_norm" },
- { LLM_KV_EXPERT_GATING_FUNC, "%s.expert_gating_func" },
- { LLM_KV_MOE_EVERY_N_LAYERS, "%s.moe_every_n_layers" },
- { LLM_KV_NEXTN_PREDICT_LAYERS, "%s.nextn_predict_layers" },
- { LLM_KV_POOLING_TYPE, "%s.pooling_type" },
- { LLM_KV_LOGIT_SCALE, "%s.logit_scale" },
- { LLM_KV_DECODER_START_TOKEN_ID, "%s.decoder_start_token_id" },
- { LLM_KV_ATTN_LOGIT_SOFTCAPPING, "%s.attn_logit_softcapping" },
- { LLM_KV_FINAL_LOGIT_SOFTCAPPING, "%s.final_logit_softcapping" },
- { LLM_KV_SWIN_NORM, "%s.swin_norm" },
- { LLM_KV_RESCALE_EVERY_N_LAYERS, "%s.rescale_every_n_layers" },
- { LLM_KV_TIME_MIX_EXTRA_DIM, "%s.time_mix_extra_dim" },
- { LLM_KV_TIME_DECAY_EXTRA_DIM, "%s.time_decay_extra_dim" },
- { LLM_KV_RESIDUAL_SCALE, "%s.residual_scale" },
- { LLM_KV_EMBEDDING_SCALE, "%s.embedding_scale" },
- { LLM_KV_TOKEN_SHIFT_COUNT, "%s.token_shift_count" },
- { LLM_KV_INTERLEAVE_MOE_LAYER_STEP, "%s.interleave_moe_layer_step" },
-
- { LLM_KV_ATTENTION_HEAD_COUNT, "%s.attention.head_count" },
- { LLM_KV_ATTENTION_HEAD_COUNT_KV, "%s.attention.head_count_kv" },
- { LLM_KV_ATTENTION_MAX_ALIBI_BIAS, "%s.attention.max_alibi_bias" },
- { LLM_KV_ATTENTION_CLAMP_KQV, "%s.attention.clamp_kqv" },
- { LLM_KV_ATTENTION_KEY_LENGTH, "%s.attention.key_length" },
- { LLM_KV_ATTENTION_VALUE_LENGTH, "%s.attention.value_length" },
- { LLM_KV_ATTENTION_LAYERNORM_EPS, "%s.attention.layer_norm_epsilon" },
- { LLM_KV_ATTENTION_LAYERNORM_RMS_EPS, "%s.attention.layer_norm_rms_epsilon" },
- { LLM_KV_ATTENTION_GROUPNORM_EPS, "%s.attention.group_norm_epsilon" },
- { LLM_KV_ATTENTION_GROUPNORM_GROUPS, "%s.attention.group_norm_groups" },
- { LLM_KV_ATTENTION_CAUSAL, "%s.attention.causal" },
- { LLM_KV_ATTENTION_Q_LORA_RANK, "%s.attention.q_lora_rank" },
- { LLM_KV_ATTENTION_KV_LORA_RANK, "%s.attention.kv_lora_rank" },
- { LLM_KV_ATTENTION_DECAY_LORA_RANK, "%s.attention.decay_lora_rank" },
- { LLM_KV_ATTENTION_ICLR_LORA_RANK, "%s.attention.iclr_lora_rank" },
- { LLM_KV_ATTENTION_VALUE_RESIDUAL_MIX_LORA_RANK, "%s.attention.value_residual_mix_lora_rank" },
- { LLM_KV_ATTENTION_GATE_LORA_RANK, "%s.attention.gate_lora_rank" },
- { LLM_KV_ATTENTION_RELATIVE_BUCKETS_COUNT, "%s.attention.relative_buckets_count" },
- { LLM_KV_ATTENTION_SLIDING_WINDOW, "%s.attention.sliding_window" },
- { LLM_KV_ATTENTION_SCALE, "%s.attention.scale" },
- { LLM_KV_ATTENTION_KEY_LENGTH_MLA, "%s.attention.key_length_mla" },
- { LLM_KV_ATTENTION_VALUE_LENGTH_MLA, "%s.attention.value_length_mla" },
-
- { LLM_KV_ROPE_DIMENSION_COUNT, "%s.rope.dimension_count" },
- { LLM_KV_ROPE_DIMENSION_SECTIONS, "%s.rope.dimension_sections" },
- { LLM_KV_ROPE_FREQ_BASE, "%s.rope.freq_base" },
- { LLM_KV_ROPE_SCALE_LINEAR, "%s.rope.scale_linear" },
- { LLM_KV_ROPE_SCALING_TYPE, "%s.rope.scaling.type" },
- { LLM_KV_ROPE_SCALING_FACTOR, "%s.rope.scaling.factor" },
- { LLM_KV_ROPE_SCALING_ATTN_FACTOR, "%s.rope.scaling.attn_factor" },
- { LLM_KV_ROPE_SCALING_ORIG_CTX_LEN, "%s.rope.scaling.original_context_length" },
- { LLM_KV_ROPE_SCALING_FINETUNED, "%s.rope.scaling.finetuned" },
- { LLM_KV_ROPE_SCALING_YARN_LOG_MUL, "%s.rope.scaling.yarn_log_multiplier" },
-
- { LLM_KV_SPLIT_NO, "split.no" },
- { LLM_KV_SPLIT_COUNT, "split.count" },
- { LLM_KV_SPLIT_TENSORS_COUNT, "split.tensors.count" },
-
- { LLM_KV_SSM_CONV_KERNEL, "%s.ssm.conv_kernel" },
- { LLM_KV_SSM_INNER_SIZE, "%s.ssm.inner_size" },
- { LLM_KV_SSM_STATE_SIZE, "%s.ssm.state_size" },
- { LLM_KV_SSM_TIME_STEP_RANK, "%s.ssm.time_step_rank" },
- { LLM_KV_SSM_GROUP_COUNT, "%s.ssm.group_count" },
- { LLM_KV_SSM_DT_B_C_RMS, "%s.ssm.dt_b_c_rms" },
-
- { LLM_KV_WKV_HEAD_SIZE, "%s.wkv.head_size" },
-
- { LLM_KV_POSNET_EMBEDDING_LENGTH, "%s.posnet.embedding_length" },
- { LLM_KV_POSNET_BLOCK_COUNT, "%s.posnet.block_count" },
-
- { LLM_KV_CONVNEXT_EMBEDDING_LENGTH, "%s.convnext.embedding_length" },
- { LLM_KV_CONVNEXT_BLOCK_COUNT, "%s.convnext.block_count" },
-
- { LLM_KV_CLASSIFIER_OUTPUT_LABELS, "%s.classifier.output_labels" },
-
- { LLM_KV_SHORTCONV_L_CACHE, "%s.shortconv.l_cache" },
-
- { LLM_KV_TOKENIZER_MODEL, "tokenizer.ggml.model" },
- { LLM_KV_TOKENIZER_PRE, "tokenizer.ggml.pre" },
- { LLM_KV_TOKENIZER_LIST, "tokenizer.ggml.tokens" },
- { LLM_KV_TOKENIZER_TOKEN_TYPE, "tokenizer.ggml.token_type" },
- { LLM_KV_TOKENIZER_TOKEN_TYPE_COUNT, "tokenizer.ggml.token_type_count" },
- { LLM_KV_TOKENIZER_SCORES, "tokenizer.ggml.scores" },
- { LLM_KV_TOKENIZER_MERGES, "tokenizer.ggml.merges" },
- { LLM_KV_TOKENIZER_BOS_ID, "tokenizer.ggml.bos_token_id" },
- { LLM_KV_TOKENIZER_EOS_ID, "tokenizer.ggml.eos_token_id" },
- { LLM_KV_TOKENIZER_EOT_ID, "tokenizer.ggml.eot_token_id" },
- { LLM_KV_TOKENIZER_EOM_ID, "tokenizer.ggml.eom_token_id" },
- { LLM_KV_TOKENIZER_UNK_ID, "tokenizer.ggml.unknown_token_id" },
- { LLM_KV_TOKENIZER_SEP_ID, "tokenizer.ggml.seperator_token_id" },
- { LLM_KV_TOKENIZER_PAD_ID, "tokenizer.ggml.padding_token_id" },
- { LLM_KV_TOKENIZER_CLS_ID, "tokenizer.ggml.cls_token_id" },
- { LLM_KV_TOKENIZER_MASK_ID, "tokenizer.ggml.mask_token_id" },
- { LLM_KV_TOKENIZER_ADD_BOS, "tokenizer.ggml.add_bos_token" },
- { LLM_KV_TOKENIZER_ADD_EOS, "tokenizer.ggml.add_eos_token" },
- { LLM_KV_TOKENIZER_ADD_SEP, "tokenizer.ggml.add_sep_token" },
- { LLM_KV_TOKENIZER_ADD_PREFIX, "tokenizer.ggml.add_space_prefix" },
- { LLM_KV_TOKENIZER_REMOVE_EXTRA_WS, "tokenizer.ggml.remove_extra_whitespaces" },
- { LLM_KV_TOKENIZER_PRECOMPILED_CHARSMAP, "tokenizer.ggml.precompiled_charsmap" },
- { LLM_KV_TOKENIZER_HF_JSON, "tokenizer.huggingface.json" },
- { LLM_KV_TOKENIZER_RWKV, "tokenizer.rwkv.world" },
- { LLM_KV_TOKENIZER_CHAT_TEMPLATE, "tokenizer.chat_template" },
- { LLM_KV_TOKENIZER_FIM_PRE_ID, "tokenizer.ggml.fim_pre_token_id" },
- { LLM_KV_TOKENIZER_FIM_SUF_ID, "tokenizer.ggml.fim_suf_token_id" },
- { LLM_KV_TOKENIZER_FIM_MID_ID, "tokenizer.ggml.fim_mid_token_id" },
- { LLM_KV_TOKENIZER_FIM_PAD_ID, "tokenizer.ggml.fim_pad_token_id" },
- { LLM_KV_TOKENIZER_FIM_REP_ID, "tokenizer.ggml.fim_rep_token_id" },
- { LLM_KV_TOKENIZER_FIM_SEP_ID, "tokenizer.ggml.fim_sep_token_id" },
-
- { LLM_KV_ADAPTER_TYPE, "adapter.type" },
- { LLM_KV_ADAPTER_LORA_ALPHA, "adapter.lora.alpha" },
-
- // deprecated
- { LLM_KV_TOKENIZER_PREFIX_ID, "tokenizer.ggml.prefix_token_id" },
- { LLM_KV_TOKENIZER_SUFFIX_ID, "tokenizer.ggml.suffix_token_id" },
- { LLM_KV_TOKENIZER_MIDDLE_ID, "tokenizer.ggml.middle_token_id" },
-};
-
-static const std::map<llm_arch, std::map<llm_tensor, const char *>> LLM_TENSOR_NAMES = {
- {
- LLM_ARCH_LLAMA,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_GATE_EXP, "blk.%d.ffn_gate.%d" },
- { LLM_TENSOR_FFN_DOWN_EXP, "blk.%d.ffn_down.%d" },
- { LLM_TENSOR_FFN_UP_EXP, "blk.%d.ffn_up.%d" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- },
- },
- {
- LLM_ARCH_ARCEE,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_LLAMA4,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_GATE_EXP, "blk.%d.ffn_gate.%d" },
- { LLM_TENSOR_FFN_DOWN_EXP, "blk.%d.ffn_down.%d" },
- { LLM_TENSOR_FFN_UP_EXP, "blk.%d.ffn_up.%d" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
- { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
- { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
- },
- },
- {
- LLM_ARCH_DECI,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_GATE_EXP, "blk.%d.ffn_gate.%d" },
- { LLM_TENSOR_FFN_DOWN_EXP, "blk.%d.ffn_down.%d" },
- { LLM_TENSOR_FFN_UP_EXP, "blk.%d.ffn_up.%d" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- },
- },
- {
- LLM_ARCH_BAICHUAN,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_FALCON,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_NORM_2, "blk.%d.attn_norm_2" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_GROK,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE_EXP, "blk.%d.ffn_gate.%d" },
- { LLM_TENSOR_FFN_DOWN_EXP, "blk.%d.ffn_down.%d" },
- { LLM_TENSOR_FFN_UP_EXP, "blk.%d.ffn_up.%d" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- { LLM_TENSOR_LAYER_OUT_NORM, "blk.%d.layer_output_norm" },
- { LLM_TENSOR_ATTN_OUT_NORM, "blk.%d.attn_output_norm" },
- },
- },
- {
- LLM_ARCH_GPT2,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_POS_EMBD, "position_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- },
- },
- {
- LLM_ARCH_GPTJ,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- },
- },
- {
- LLM_ARCH_GPTNEOX,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_MPT,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output"},
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_ACT, "blk.%d.ffn.act" },
- { LLM_TENSOR_POS_EMBD, "position_embd" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm"},
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm"},
- },
- },
- {
- LLM_ARCH_STARCODER,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_POS_EMBD, "position_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- },
- },
- {
- LLM_ARCH_REFACT,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_BERT,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
- { LLM_TENSOR_TOKEN_TYPES, "token_types" },
- { LLM_TENSOR_POS_EMBD, "position_embd" },
- { LLM_TENSOR_ATTN_OUT_NORM, "blk.%d.attn_output_norm" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_LAYER_OUT_NORM, "blk.%d.layer_output_norm" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_CLS, "cls" },
- { LLM_TENSOR_CLS_OUT, "cls.output" },
- },
- },
- {
- LLM_ARCH_NOMIC_BERT,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
- { LLM_TENSOR_TOKEN_TYPES, "token_types" },
- { LLM_TENSOR_ATTN_OUT_NORM, "blk.%d.attn_output_norm" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_LAYER_OUT_NORM, "blk.%d.layer_output_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_NOMIC_BERT_MOE,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
- { LLM_TENSOR_TOKEN_TYPES, "token_types" },
- { LLM_TENSOR_ATTN_OUT_NORM, "blk.%d.attn_output_norm" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_LAYER_OUT_NORM, "blk.%d.layer_output_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- },
- },
- {
- LLM_ARCH_NEO_BERT,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_ENC_OUTPUT_NORM, "enc.output_norm" },
- { LLM_TENSOR_CLS, "cls" },
- { LLM_TENSOR_CLS_OUT, "cls.output" },
- },
- },
- {
- LLM_ARCH_JINA_BERT_V2,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
- { LLM_TENSOR_TOKEN_TYPES, "token_types" },
- { LLM_TENSOR_ATTN_NORM_2, "blk.%d.attn_norm_2" },
- { LLM_TENSOR_ATTN_OUT_NORM, "blk.%d.attn_output_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_LAYER_OUT_NORM, "blk.%d.layer_output_norm" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_CLS, "cls" },
- },
- },
- {
- LLM_ARCH_BLOOM,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- },
- },
- {
- LLM_ARCH_STABLELM,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
- },
- },
- {
- LLM_ARCH_QWEN,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_QWEN2,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_QWEN2VL,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_QWEN2MOE,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- { LLM_TENSOR_FFN_GATE_INP_SHEXP, "blk.%d.ffn_gate_inp_shexp" },
- { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
- { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
- { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
- },
- },
- {
- LLM_ARCH_QWEN3,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_QWEN3MOE,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- },
- },
- {
- LLM_ARCH_PHI2,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_PHI3,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FACTORS_LONG, "rope_factors_long" },
- { LLM_TENSOR_ROPE_FACTORS_SHORT, "rope_factors_short" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_PHIMOE,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FACTORS_LONG, "rope_factors_long" },
- { LLM_TENSOR_ROPE_FACTORS_SHORT, "rope_factors_short" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- },
- },
- {
- LLM_ARCH_PLAMO,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_PLAMO2,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_SSM_IN, "blk.%d.ssm_in" },
- { LLM_TENSOR_SSM_CONV1D, "blk.%d.ssm_conv1d" },
- { LLM_TENSOR_SSM_X, "blk.%d.ssm_x" },
- { LLM_TENSOR_SSM_DT, "blk.%d.ssm_dt" },
- { LLM_TENSOR_SSM_A, "blk.%d.ssm_a" },
- { LLM_TENSOR_SSM_D, "blk.%d.ssm_d" },
- { LLM_TENSOR_SSM_OUT, "blk.%d.ssm_out" },
- { LLM_TENSOR_SSM_DT_NORM, "blk.%d.ssm_dt_norm" },
- { LLM_TENSOR_SSM_B_NORM, "blk.%d.ssm_b_norm" },
- { LLM_TENSOR_SSM_C_NORM, "blk.%d.ssm_c_norm" },
- { LLM_TENSOR_ATTN_POST_NORM, "blk.%d.post_attention_norm" },
- { LLM_TENSOR_FFN_POST_NORM, "blk.%d.post_ffw_norm" },
- },
- },
- {
- LLM_ARCH_CODESHELL,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_ORION,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_INTERNLM2,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_MINICPM,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ROPE_FACTORS_LONG, "rope_factors_long" },
- { LLM_TENSOR_ROPE_FACTORS_SHORT, "rope_factors_short" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_GATE_EXP, "blk.%d.ffn_gate.%d" },
- { LLM_TENSOR_FFN_DOWN_EXP, "blk.%d.ffn_down.%d" },
- { LLM_TENSOR_FFN_UP_EXP, "blk.%d.ffn_up.%d" },
- },
- },
- {
- LLM_ARCH_MINICPM3,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FACTORS_LONG, "rope_factors_long" },
- { LLM_TENSOR_ROPE_FACTORS_SHORT, "rope_factors_short" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q_A_NORM, "blk.%d.attn_q_a_norm" },
- { LLM_TENSOR_ATTN_KV_A_NORM, "blk.%d.attn_kv_a_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_Q_A, "blk.%d.attn_q_a" },
- { LLM_TENSOR_ATTN_Q_B, "blk.%d.attn_q_b" },
- { LLM_TENSOR_ATTN_KV_A_MQA, "blk.%d.attn_kv_a_mqa" },
- { LLM_TENSOR_ATTN_KV_B, "blk.%d.attn_kv_b" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- },
- },
- {
- LLM_ARCH_GEMMA,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_GEMMA2,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_POST_NORM, "blk.%d.post_attention_norm" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_POST_NORM, "blk.%d.post_ffw_norm" },
- },
- },
- {
- LLM_ARCH_GEMMA3,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_POST_NORM, "blk.%d.post_attention_norm" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_POST_NORM, "blk.%d.post_ffw_norm" },
- },
- },
- {
- LLM_ARCH_GEMMA3N,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_POST_NORM, "blk.%d.post_attention_norm" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_POST_NORM, "blk.%d.post_ffw_norm" },
- { LLM_TENSOR_PER_LAYER_TOKEN_EMBD, "per_layer_token_embd" },
- { LLM_TENSOR_PER_LAYER_MODEL_PROJ, "per_layer_model_proj" },
- { LLM_TENSOR_PER_LAYER_PROJ_NORM, "per_layer_proj_norm" },
- { LLM_TENSOR_ALTUP_UNEMBD_PROJ, "altup_unembd_proj" },
- { LLM_TENSOR_ALTUP_PROJ, "altup_proj" },
- { LLM_TENSOR_PER_LAYER_INP_GATE, "blk.%d.inp_gate" },
- { LLM_TENSOR_PER_LAYER_PROJ, "blk.%d.proj" },
- { LLM_TENSOR_PER_LAYER_POST_NORM, "blk.%d.post_norm" },
- { LLM_TENSOR_ALTUP_CORRECT_COEF, "blk.%d.altup_correct_coef" },
- { LLM_TENSOR_ALTUP_CORRECT_SCALE, "blk.%d.altup_correct_scale" },
- { LLM_TENSOR_ALTUP_PREDICT_COEF, "blk.%d.altup_predict_coef" },
- { LLM_TENSOR_ALTUP_ROUTER, "blk.%d.altup_router" },
- { LLM_TENSOR_ALTUP_ROUTER_NORM, "blk.%d.altup_router_norm" },
- { LLM_TENSOR_LAUREL_L, "blk.%d.laurel_l" },
- { LLM_TENSOR_LAUREL_R, "blk.%d.laurel_r" },
- { LLM_TENSOR_LAUREL_POST_NORM, "blk.%d.laurel_post_norm" },
- },
- },
- {
- LLM_ARCH_STARCODER2,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_MAMBA,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_SSM_IN, "blk.%d.ssm_in" },
- { LLM_TENSOR_SSM_CONV1D, "blk.%d.ssm_conv1d" },
- { LLM_TENSOR_SSM_X, "blk.%d.ssm_x" },
- { LLM_TENSOR_SSM_DT, "blk.%d.ssm_dt" },
- { LLM_TENSOR_SSM_A, "blk.%d.ssm_a" },
- { LLM_TENSOR_SSM_D, "blk.%d.ssm_d" },
- { LLM_TENSOR_SSM_OUT, "blk.%d.ssm_out" },
- },
- },
- {
- LLM_ARCH_MAMBA2,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_SSM_IN, "blk.%d.ssm_in" },
- { LLM_TENSOR_SSM_CONV1D, "blk.%d.ssm_conv1d" },
- { LLM_TENSOR_SSM_DT, "blk.%d.ssm_dt" },
- { LLM_TENSOR_SSM_A, "blk.%d.ssm_a" },
- { LLM_TENSOR_SSM_D, "blk.%d.ssm_d" },
- { LLM_TENSOR_SSM_NORM, "blk.%d.ssm_norm" },
- { LLM_TENSOR_SSM_OUT, "blk.%d.ssm_out" },
- },
- },
- {
- LLM_ARCH_JAMBA,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_SSM_IN, "blk.%d.ssm_in" },
- { LLM_TENSOR_SSM_CONV1D, "blk.%d.ssm_conv1d" },
- { LLM_TENSOR_SSM_X, "blk.%d.ssm_x" },
- { LLM_TENSOR_SSM_DT, "blk.%d.ssm_dt" },
- { LLM_TENSOR_SSM_DT_NORM, "blk.%d.ssm_dt_norm" },
- { LLM_TENSOR_SSM_A, "blk.%d.ssm_a" },
- { LLM_TENSOR_SSM_B_NORM, "blk.%d.ssm_b_norm" },
- { LLM_TENSOR_SSM_C_NORM, "blk.%d.ssm_c_norm" },
- { LLM_TENSOR_SSM_D, "blk.%d.ssm_d" },
- { LLM_TENSOR_SSM_OUT, "blk.%d.ssm_out" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- },
- },
- {
- LLM_ARCH_FALCON_H1,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_SSM_IN, "blk.%d.ssm_in" },
- { LLM_TENSOR_SSM_CONV1D, "blk.%d.ssm_conv1d" },
- { LLM_TENSOR_SSM_DT, "blk.%d.ssm_dt" },
- { LLM_TENSOR_SSM_A, "blk.%d.ssm_a" },
- { LLM_TENSOR_SSM_D, "blk.%d.ssm_d" },
- { LLM_TENSOR_SSM_NORM, "blk.%d.ssm_norm" },
- { LLM_TENSOR_SSM_OUT, "blk.%d.ssm_out" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_XVERSE,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_COMMAND_R,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
- },
- },
- {
- LLM_ARCH_COHERE2,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_DBRX,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_OUT_NORM, "blk.%d.attn_output_norm" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- },
- },
- {
- LLM_ARCH_OLMO,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_OLMO2,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_POST_NORM, "blk.%d.post_attention_norm" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
- { LLM_TENSOR_FFN_POST_NORM, "blk.%d.post_ffw_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_OLMOE,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- },
- },
- {
- LLM_ARCH_OPENELM,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_ARCTIC,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_NORM_EXPS, "blk.%d.ffn_norm_exps" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- },
- },
- {
- LLM_ARCH_DEEPSEEK,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- { LLM_TENSOR_FFN_GATE_INP_SHEXP, "blk.%d.ffn_gate_inp_shexp" },
- { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
- { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
- { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
- },
- },
- {
- LLM_ARCH_DEEPSEEK2,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q_A_NORM, "blk.%d.attn_q_a_norm" },
- { LLM_TENSOR_ATTN_KV_A_NORM, "blk.%d.attn_kv_a_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_Q_A, "blk.%d.attn_q_a" },
- { LLM_TENSOR_ATTN_Q_B, "blk.%d.attn_q_b" },
- { LLM_TENSOR_ATTN_KV_A_MQA, "blk.%d.attn_kv_a_mqa" },
- { LLM_TENSOR_ATTN_KV_B, "blk.%d.attn_kv_b" },
- { LLM_TENSOR_ATTN_K_B, "blk.%d.attn_k_b" },
- { LLM_TENSOR_ATTN_V_B, "blk.%d.attn_v_b" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- { LLM_TENSOR_FFN_GATE_INP_SHEXP, "blk.%d.ffn_gate_inp_shexp" },
- { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
- { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
- { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
- { LLM_TENSOR_FFN_EXP_PROBS_B, "blk.%d.exp_probs_b" },
- },
- },
- {
- LLM_ARCH_PLM,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_KV_A_MQA, "blk.%d.attn_kv_a_mqa" },
- { LLM_TENSOR_ATTN_KV_A_NORM, "blk.%d.attn_kv_a_norm" },
- { LLM_TENSOR_ATTN_KV_B, "blk.%d.attn_kv_b" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_CHATGLM,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- },
- },
- {
- LLM_ARCH_GLM4,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_ATTN_POST_NORM, "blk.%d.post_attention_norm" },
- { LLM_TENSOR_FFN_POST_NORM, "blk.%d.post_ffw_norm" },
- },
- },
- {
- LLM_ARCH_GLM4_MOE,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_POST_NORM, "blk.%d.post_attention_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
- { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
- { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
- { LLM_TENSOR_FFN_EXP_PROBS_B, "blk.%d.exp_probs_b" },
- // NextN/MTP tensors - preserved but unused (in final layer, dynamic layer number)
- { LLM_TENSOR_NEXTN_EH_PROJ, "blk.%d.nextn.eh_proj" },
- { LLM_TENSOR_NEXTN_EMBED_TOKENS, "blk.%d.nextn.embed_tokens" },
- { LLM_TENSOR_NEXTN_ENORM, "blk.%d.nextn.enorm" },
- { LLM_TENSOR_NEXTN_HNORM, "blk.%d.nextn.hnorm" },
- { LLM_TENSOR_NEXTN_SHARED_HEAD_HEAD, "blk.%d.nextn.shared_head_head" },
- { LLM_TENSOR_NEXTN_SHARED_HEAD_NORM, "blk.%d.nextn.shared_head_norm" },
- },
- },
- {
- LLM_ARCH_BITNET,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_SUB_NORM, "blk.%d.attn_sub_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_SUB_NORM, "blk.%d.ffn_sub_norm" },
- },
- },
- {
- LLM_ARCH_T5,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_DEC_OUTPUT_NORM, "dec.output_norm" },
- { LLM_TENSOR_DEC_ATTN_NORM, "dec.blk.%d.attn_norm" },
- { LLM_TENSOR_DEC_ATTN_Q, "dec.blk.%d.attn_q" },
- { LLM_TENSOR_DEC_ATTN_K, "dec.blk.%d.attn_k" },
- { LLM_TENSOR_DEC_ATTN_V, "dec.blk.%d.attn_v" },
- { LLM_TENSOR_DEC_ATTN_OUT, "dec.blk.%d.attn_o" },
- { LLM_TENSOR_DEC_ATTN_REL_B, "dec.blk.%d.attn_rel_b" },
- { LLM_TENSOR_DEC_CROSS_ATTN_NORM, "dec.blk.%d.cross_attn_norm" },
- { LLM_TENSOR_DEC_CROSS_ATTN_Q, "dec.blk.%d.cross_attn_q" },
- { LLM_TENSOR_DEC_CROSS_ATTN_K, "dec.blk.%d.cross_attn_k" },
- { LLM_TENSOR_DEC_CROSS_ATTN_V, "dec.blk.%d.cross_attn_v" },
- { LLM_TENSOR_DEC_CROSS_ATTN_OUT, "dec.blk.%d.cross_attn_o" },
- { LLM_TENSOR_DEC_CROSS_ATTN_REL_B, "dec.blk.%d.cross_attn_rel_b" },
- { LLM_TENSOR_DEC_FFN_NORM, "dec.blk.%d.ffn_norm" },
- { LLM_TENSOR_DEC_FFN_GATE, "dec.blk.%d.ffn_gate" },
- { LLM_TENSOR_DEC_FFN_DOWN, "dec.blk.%d.ffn_down" },
- { LLM_TENSOR_DEC_FFN_UP, "dec.blk.%d.ffn_up" },
- { LLM_TENSOR_ENC_OUTPUT_NORM, "enc.output_norm" },
- { LLM_TENSOR_ENC_ATTN_NORM, "enc.blk.%d.attn_norm" },
- { LLM_TENSOR_ENC_ATTN_Q, "enc.blk.%d.attn_q" },
- { LLM_TENSOR_ENC_ATTN_K, "enc.blk.%d.attn_k" },
- { LLM_TENSOR_ENC_ATTN_V, "enc.blk.%d.attn_v" },
- { LLM_TENSOR_ENC_ATTN_OUT, "enc.blk.%d.attn_o" },
- { LLM_TENSOR_ENC_ATTN_REL_B, "enc.blk.%d.attn_rel_b" },
- { LLM_TENSOR_ENC_FFN_NORM, "enc.blk.%d.ffn_norm" },
- { LLM_TENSOR_ENC_FFN_GATE, "enc.blk.%d.ffn_gate" },
- { LLM_TENSOR_ENC_FFN_DOWN, "enc.blk.%d.ffn_down" },
- { LLM_TENSOR_ENC_FFN_UP, "enc.blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_T5ENCODER,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ENC_OUTPUT_NORM, "enc.output_norm" },
- { LLM_TENSOR_ENC_ATTN_NORM, "enc.blk.%d.attn_norm" },
- { LLM_TENSOR_ENC_ATTN_Q, "enc.blk.%d.attn_q" },
- { LLM_TENSOR_ENC_ATTN_K, "enc.blk.%d.attn_k" },
- { LLM_TENSOR_ENC_ATTN_V, "enc.blk.%d.attn_v" },
- { LLM_TENSOR_ENC_ATTN_OUT, "enc.blk.%d.attn_o" },
- { LLM_TENSOR_ENC_ATTN_REL_B, "enc.blk.%d.attn_rel_b" },
- { LLM_TENSOR_ENC_FFN_NORM, "enc.blk.%d.ffn_norm" },
- { LLM_TENSOR_ENC_FFN_GATE, "enc.blk.%d.ffn_gate" },
- { LLM_TENSOR_ENC_FFN_DOWN, "enc.blk.%d.ffn_down" },
- { LLM_TENSOR_ENC_FFN_UP, "enc.blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_JAIS,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- },
- },
- {
- LLM_ARCH_NEMOTRON,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_EXAONE,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_EXAONE4,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_POST_NORM, "blk.%d.post_attention_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_POST_NORM, "blk.%d.post_ffw_norm" },
- }
- },
- {
- LLM_ARCH_RWKV6,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_NORM_2, "blk.%d.attn_norm_2" },
- { LLM_TENSOR_TIME_MIX_W1, "blk.%d.time_mix_w1" },
- { LLM_TENSOR_TIME_MIX_W2, "blk.%d.time_mix_w2" },
- { LLM_TENSOR_TIME_MIX_LERP_X, "blk.%d.time_mix_lerp_x" },
- { LLM_TENSOR_TIME_MIX_LERP_W, "blk.%d.time_mix_lerp_w" },
- { LLM_TENSOR_TIME_MIX_LERP_K, "blk.%d.time_mix_lerp_k" },
- { LLM_TENSOR_TIME_MIX_LERP_V, "blk.%d.time_mix_lerp_v" },
- { LLM_TENSOR_TIME_MIX_LERP_R, "blk.%d.time_mix_lerp_r" },
- { LLM_TENSOR_TIME_MIX_LERP_G, "blk.%d.time_mix_lerp_g" },
- { LLM_TENSOR_TIME_MIX_LERP_FUSED, "blk.%d.time_mix_lerp_fused" },
- { LLM_TENSOR_TIME_MIX_FIRST, "blk.%d.time_mix_first" },
- { LLM_TENSOR_TIME_MIX_DECAY, "blk.%d.time_mix_decay" },
- { LLM_TENSOR_TIME_MIX_DECAY_W1, "blk.%d.time_mix_decay_w1" },
- { LLM_TENSOR_TIME_MIX_DECAY_W2, "blk.%d.time_mix_decay_w2" },
- { LLM_TENSOR_TIME_MIX_KEY, "blk.%d.time_mix_key" },
- { LLM_TENSOR_TIME_MIX_VALUE, "blk.%d.time_mix_value" },
- { LLM_TENSOR_TIME_MIX_RECEPTANCE, "blk.%d.time_mix_receptance" },
- { LLM_TENSOR_TIME_MIX_GATE, "blk.%d.time_mix_gate" },
- { LLM_TENSOR_TIME_MIX_LN, "blk.%d.time_mix_ln" },
- { LLM_TENSOR_TIME_MIX_OUTPUT, "blk.%d.time_mix_output" },
- { LLM_TENSOR_CHANNEL_MIX_LERP_K, "blk.%d.channel_mix_lerp_k" },
- { LLM_TENSOR_CHANNEL_MIX_LERP_R, "blk.%d.channel_mix_lerp_r" },
- { LLM_TENSOR_CHANNEL_MIX_KEY, "blk.%d.channel_mix_key" },
- { LLM_TENSOR_CHANNEL_MIX_VALUE, "blk.%d.channel_mix_value" },
- { LLM_TENSOR_CHANNEL_MIX_RECEPTANCE, "blk.%d.channel_mix_receptance" },
- },
- },
- {
- LLM_ARCH_RWKV6QWEN2,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_TIME_MIX_W1, "blk.%d.time_mix_w1" },
- { LLM_TENSOR_TIME_MIX_W2, "blk.%d.time_mix_w2" },
- { LLM_TENSOR_TIME_MIX_LERP_X, "blk.%d.time_mix_lerp_x" },
- { LLM_TENSOR_TIME_MIX_LERP_FUSED, "blk.%d.time_mix_lerp_fused" },
- { LLM_TENSOR_TIME_MIX_FIRST, "blk.%d.time_mix_first" },
- { LLM_TENSOR_TIME_MIX_DECAY, "blk.%d.time_mix_decay" },
- { LLM_TENSOR_TIME_MIX_DECAY_W1, "blk.%d.time_mix_decay_w1" },
- { LLM_TENSOR_TIME_MIX_DECAY_W2, "blk.%d.time_mix_decay_w2" },
- { LLM_TENSOR_TIME_MIX_KEY, "blk.%d.time_mix_key" },
- { LLM_TENSOR_TIME_MIX_VALUE, "blk.%d.time_mix_value" },
- { LLM_TENSOR_TIME_MIX_RECEPTANCE, "blk.%d.time_mix_receptance" },
- { LLM_TENSOR_TIME_MIX_GATE, "blk.%d.time_mix_gate" },
- { LLM_TENSOR_TIME_MIX_OUTPUT, "blk.%d.time_mix_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_RWKV7,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_NORM_2, "blk.%d.attn_norm_2" },
- { LLM_TENSOR_TIME_MIX_W0, "blk.%d.time_mix_w0" },
- { LLM_TENSOR_TIME_MIX_W1, "blk.%d.time_mix_w1" },
- { LLM_TENSOR_TIME_MIX_W2, "blk.%d.time_mix_w2" },
- { LLM_TENSOR_TIME_MIX_A0, "blk.%d.time_mix_a0" },
- { LLM_TENSOR_TIME_MIX_A1, "blk.%d.time_mix_a1" },
- { LLM_TENSOR_TIME_MIX_A2, "blk.%d.time_mix_a2" },
- { LLM_TENSOR_TIME_MIX_V0, "blk.%d.time_mix_v0" },
- { LLM_TENSOR_TIME_MIX_V1, "blk.%d.time_mix_v1" },
- { LLM_TENSOR_TIME_MIX_V2, "blk.%d.time_mix_v2" },
- { LLM_TENSOR_TIME_MIX_G1, "blk.%d.time_mix_g1" },
- { LLM_TENSOR_TIME_MIX_G2, "blk.%d.time_mix_g2" },
- { LLM_TENSOR_TIME_MIX_K_K, "blk.%d.time_mix_k_k" },
- { LLM_TENSOR_TIME_MIX_K_A, "blk.%d.time_mix_k_a" },
- { LLM_TENSOR_TIME_MIX_R_K, "blk.%d.time_mix_r_k" },
- { LLM_TENSOR_TIME_MIX_LERP_FUSED, "blk.%d.time_mix_lerp_fused" },
- { LLM_TENSOR_TIME_MIX_KEY, "blk.%d.time_mix_key" },
- { LLM_TENSOR_TIME_MIX_VALUE, "blk.%d.time_mix_value" },
- { LLM_TENSOR_TIME_MIX_RECEPTANCE, "blk.%d.time_mix_receptance" },
- { LLM_TENSOR_TIME_MIX_LN, "blk.%d.time_mix_ln" },
- { LLM_TENSOR_TIME_MIX_OUTPUT, "blk.%d.time_mix_output" },
- { LLM_TENSOR_CHANNEL_MIX_LERP_K, "blk.%d.channel_mix_lerp_k" },
- { LLM_TENSOR_CHANNEL_MIX_KEY, "blk.%d.channel_mix_key" },
- { LLM_TENSOR_CHANNEL_MIX_VALUE, "blk.%d.channel_mix_value" },
- },
- },
- {
- LLM_ARCH_ARWKV7,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_TIME_MIX_W0, "blk.%d.time_mix_w0" },
- { LLM_TENSOR_TIME_MIX_W1, "blk.%d.time_mix_w1" },
- { LLM_TENSOR_TIME_MIX_W2, "blk.%d.time_mix_w2" },
- { LLM_TENSOR_TIME_MIX_A0, "blk.%d.time_mix_a0" },
- { LLM_TENSOR_TIME_MIX_A1, "blk.%d.time_mix_a1" },
- { LLM_TENSOR_TIME_MIX_A2, "blk.%d.time_mix_a2" },
- { LLM_TENSOR_TIME_MIX_V0, "blk.%d.time_mix_v0" },
- { LLM_TENSOR_TIME_MIX_V1, "blk.%d.time_mix_v1" },
- { LLM_TENSOR_TIME_MIX_V2, "blk.%d.time_mix_v2" },
- { LLM_TENSOR_TIME_MIX_G1, "blk.%d.time_mix_g1" },
- { LLM_TENSOR_TIME_MIX_G2, "blk.%d.time_mix_g2" },
- { LLM_TENSOR_TIME_MIX_K_K, "blk.%d.time_mix_k_k" },
- { LLM_TENSOR_TIME_MIX_K_A, "blk.%d.time_mix_k_a" },
- { LLM_TENSOR_TIME_MIX_R_K, "blk.%d.time_mix_r_k" },
- { LLM_TENSOR_TIME_MIX_LERP_FUSED, "blk.%d.time_mix_lerp_fused" },
- { LLM_TENSOR_TIME_MIX_KEY, "blk.%d.time_mix_key" },
- { LLM_TENSOR_TIME_MIX_VALUE, "blk.%d.time_mix_value" },
- { LLM_TENSOR_TIME_MIX_RECEPTANCE, "blk.%d.time_mix_receptance" },
- { LLM_TENSOR_TIME_MIX_LN, "blk.%d.time_mix_ln" },
- { LLM_TENSOR_TIME_MIX_OUTPUT, "blk.%d.time_mix_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_GRANITE,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_GRANITE_MOE,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
- { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
- { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
- },
- },
- {
- LLM_ARCH_GRANITE_HYBRID,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- // mamba(2) ssm layers
- { LLM_TENSOR_SSM_IN, "blk.%d.ssm_in" },
- { LLM_TENSOR_SSM_CONV1D, "blk.%d.ssm_conv1d" },
- { LLM_TENSOR_SSM_DT, "blk.%d.ssm_dt" },
- { LLM_TENSOR_SSM_A, "blk.%d.ssm_a" },
- { LLM_TENSOR_SSM_D, "blk.%d.ssm_d" },
- { LLM_TENSOR_SSM_NORM, "blk.%d.ssm_norm" },
- { LLM_TENSOR_SSM_OUT, "blk.%d.ssm_out" },
- // attention layers
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- // dense FFN
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- // moe FFN
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- // shared expert
- { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
- { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
- { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
- },
- },
- {
- LLM_ARCH_CHAMELEON,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
- },
- },
- {
- LLM_ARCH_WAVTOKENIZER_DEC,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
- { LLM_TENSOR_CONV1D, "conv1d" },
- { LLM_TENSOR_CONVNEXT_DW, "convnext.%d.dw" },
- { LLM_TENSOR_CONVNEXT_NORM, "convnext.%d.norm" },
- { LLM_TENSOR_CONVNEXT_PW1, "convnext.%d.pw1" },
- { LLM_TENSOR_CONVNEXT_PW2, "convnext.%d.pw2" },
- { LLM_TENSOR_CONVNEXT_GAMMA, "convnext.%d.gamma" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_POS_NET_CONV1, "posnet.%d.conv1" },
- { LLM_TENSOR_POS_NET_CONV2, "posnet.%d.conv2" },
- { LLM_TENSOR_POS_NET_NORM, "posnet.%d.norm" },
- { LLM_TENSOR_POS_NET_NORM1, "posnet.%d.norm1" },
- { LLM_TENSOR_POS_NET_NORM2, "posnet.%d.norm2" },
- { LLM_TENSOR_POS_NET_ATTN_NORM, "posnet.%d.attn_norm" },
- { LLM_TENSOR_POS_NET_ATTN_Q, "posnet.%d.attn_q" },
- { LLM_TENSOR_POS_NET_ATTN_K, "posnet.%d.attn_k" },
- { LLM_TENSOR_POS_NET_ATTN_V, "posnet.%d.attn_v" },
- { LLM_TENSOR_POS_NET_ATTN_OUT, "posnet.%d.attn_output" },
- },
- },
- {
- LLM_ARCH_BAILINGMOE,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- { LLM_TENSOR_FFN_GATE_INP_SHEXP, "blk.%d.ffn_gate_inp_shexp" },
- { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
- { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
- { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
- },
- },
- {
- LLM_ARCH_DOTS1,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- { LLM_TENSOR_FFN_GATE_INP_SHEXP, "blk.%d.ffn_gate_inp_shexp" },
- { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
- { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
- { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
- { LLM_TENSOR_FFN_EXP_PROBS_B, "blk.%d.exp_probs_b" },
- }
- },
- {
- LLM_ARCH_ERNIE4_5,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_ERNIE4_5_MOE,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
- { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
- { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- { LLM_TENSOR_FFN_EXP_PROBS_B, "blk.%d.exp_probs_b" },
- },
- },
- {
- LLM_ARCH_HUNYUAN_MOE,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
- { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
- { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- },
- },
- {
- LLM_ARCH_HUNYUAN_DENSE,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
-
- },
- },
- {
- LLM_ARCH_SMOLLM3,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_OPENAI_MOE,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_POST_NORM, "blk.%d.post_attention_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_SINKS, "blk.%d.attn_sinks" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
- },
- },
- {
- LLM_ARCH_LFM2,
- {
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_SHORTCONV_CONV, "blk.%d.shortconv.conv" },
- { LLM_TENSOR_SHORTCONV_INPROJ, "blk.%d.shortconv.in_proj" },
- { LLM_TENSOR_SHORTCONV_OUTPROJ, "blk.%d.shortconv.out_proj" },
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
- }
- },
- {
- LLM_ARCH_SMALLTHINKER,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
- { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
- { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
- { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" }
- },
- },
- {
- LLM_ARCH_DREAM,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_LLADA,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
- { LLM_TENSOR_OUTPUT, "output" },
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
- { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
- { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
- { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
- { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
- },
- },
- {
- LLM_ARCH_UNKNOWN,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
- },
- },
-};
-
-static const std::map<llm_tensor, llm_tensor_info> LLM_TENSOR_INFOS = {
- {LLM_TENSOR_TOKEN_EMBD, {LLM_TENSOR_LAYER_INPUT, GGML_OP_GET_ROWS}},
- {LLM_TENSOR_POS_EMBD, {LLM_TENSOR_LAYER_INPUT, GGML_OP_GET_ROWS}},
- {LLM_TENSOR_TOKEN_EMBD_NORM, {LLM_TENSOR_LAYER_INPUT, GGML_OP_GET_ROWS}},
- {LLM_TENSOR_TOKEN_TYPES, {LLM_TENSOR_LAYER_INPUT, GGML_OP_GET_ROWS}},
- {LLM_TENSOR_OUTPUT, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_CLS, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_CLS_OUT, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_OUTPUT_NORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
- {LLM_TENSOR_DEC_OUTPUT_NORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
- {LLM_TENSOR_ENC_OUTPUT_NORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
- {LLM_TENSOR_ROPE_FREQS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ROPE}},
- {LLM_TENSOR_ROPE_FACTORS_LONG, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ROPE}},
- {LLM_TENSOR_ROPE_FACTORS_SHORT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ROPE}},
- {LLM_TENSOR_ATTN_Q, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ATTN_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ATTN_V, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ATTN_QKV, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ATTN_OUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_FFN_GATE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_FFN_DOWN, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_FFN_UP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_FFN_DOWN_SHEXP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_FFN_GATE_SHEXP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_FFN_UP_SHEXP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ATTN_Q_A, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ATTN_Q_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ATTN_KV_A_MQA, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ATTN_KV_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ATTN_K_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ATTN_V_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ATTN_SINKS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_SCALE}},
- {LLM_TENSOR_DEC_ATTN_Q, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_DEC_ATTN_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_DEC_ATTN_V, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_DEC_ATTN_OUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_DEC_CROSS_ATTN_Q, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_DEC_CROSS_ATTN_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_DEC_CROSS_ATTN_V, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_DEC_CROSS_ATTN_OUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_DEC_FFN_GATE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_DEC_FFN_DOWN, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_DEC_FFN_UP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ENC_ATTN_Q, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ENC_ATTN_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ENC_ATTN_V, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ENC_ATTN_OUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ENC_FFN_GATE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ENC_FFN_DOWN, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ENC_FFN_UP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_FFN_GATE_INP_SHEXP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_FFN_GATE_INP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_SSM_IN, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_SSM_X, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_SSM_DT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_SSM_OUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_TIME_MIX_W1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_TIME_MIX_W2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_TIME_MIX_A1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_TIME_MIX_A2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_TIME_MIX_V1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_TIME_MIX_V2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_TIME_MIX_G1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_TIME_MIX_G2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_TIME_MIX_DECAY_W1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_TIME_MIX_DECAY_W2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_TIME_MIX_KEY, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_TIME_MIX_VALUE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_TIME_MIX_RECEPTANCE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_TIME_MIX_GATE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_TIME_MIX_OUTPUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_CHANNEL_MIX_KEY, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_CHANNEL_MIX_RECEPTANCE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_CHANNEL_MIX_VALUE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_FFN_ACT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_DIV}},
- {LLM_TENSOR_SSM_CONV1D, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_SSM_CONV}},
- {LLM_TENSOR_SSM_A, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_SSM_SCAN}},
- {LLM_TENSOR_SSM_DT_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_SSM_B_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_SSM_C_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_SSM_D, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_SSM_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_TIME_MIX_LERP_X, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_TIME_MIX_LN, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_CHANNEL_MIX_LERP_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_CHANNEL_MIX_LERP_R, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_TIME_MIX_K_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_TIME_MIX_K_A, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_TIME_MIX_R_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_TIME_MIX_LERP_W, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
- {LLM_TENSOR_TIME_MIX_LERP_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
- {LLM_TENSOR_TIME_MIX_LERP_V, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
- {LLM_TENSOR_TIME_MIX_LERP_R, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
- {LLM_TENSOR_TIME_MIX_LERP_G, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
- {LLM_TENSOR_TIME_MIX_LERP_FUSED, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
- {LLM_TENSOR_TIME_MIX_DECAY, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
- {LLM_TENSOR_TIME_MIX_W0, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
- {LLM_TENSOR_TIME_MIX_A0, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
- {LLM_TENSOR_TIME_MIX_V0, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
- {LLM_TENSOR_TIME_MIX_FIRST, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_RWKV_WKV6}},
- {LLM_TENSOR_ATTN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_ATTN_NORM_2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_ATTN_OUT_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_ATTN_POST_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_FFN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_FFN_POST_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_FFN_NORM_EXPS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_ATTN_Q_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_ATTN_K_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_LAYER_OUT_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_ATTN_Q_A_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_ATTN_KV_A_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_ATTN_SUB_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_FFN_SUB_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_DEC_ATTN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_DEC_CROSS_ATTN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_DEC_FFN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_ENC_ATTN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_ENC_FFN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_DEC_ATTN_REL_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_GET_ROWS}},
- {LLM_TENSOR_ENC_ATTN_REL_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_GET_ROWS}},
- {LLM_TENSOR_FFN_DOWN_EXPS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT_ID}},
- {LLM_TENSOR_FFN_GATE_EXPS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT_ID}},
- {LLM_TENSOR_FFN_UP_EXPS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT_ID}},
- {LLM_TENSOR_FFN_EXP_PROBS_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
- // altup / laurel (gemma 3n)
- {LLM_TENSOR_PER_LAYER_TOKEN_EMBD, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_GET_ROWS}},
- {LLM_TENSOR_PER_LAYER_MODEL_PROJ, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_PER_LAYER_PROJ_NORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
- {LLM_TENSOR_ALTUP_PROJ, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ALTUP_UNEMBD_PROJ, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_PER_LAYER_INP_GATE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_PER_LAYER_PROJ, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_PER_LAYER_POST_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_ALTUP_CORRECT_COEF, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ALTUP_CORRECT_SCALE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_ALTUP_PREDICT_COEF, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ALTUP_ROUTER, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_ALTUP_ROUTER_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_LAUREL_L, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_LAUREL_R, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_LAUREL_POST_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- // this tensor is loaded for T5, but never used
- {LLM_TENSOR_DEC_CROSS_ATTN_REL_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_NONE}},
- {LLM_TENSOR_CONV1D, {LLM_TENSOR_LAYER_INPUT, GGML_OP_IM2COL}},
- {LLM_TENSOR_POS_NET_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_POS_NET_NORM1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_POS_NET_NORM2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_POS_NET_CONV1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_IM2COL}},
- {LLM_TENSOR_POS_NET_CONV2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_IM2COL}},
- {LLM_TENSOR_POS_NET_ATTN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_POS_NET_ATTN_Q, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_POS_NET_ATTN_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_POS_NET_ATTN_V, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_POS_NET_ATTN_OUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_CONVNEXT_DW, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_IM2COL}},
- {LLM_TENSOR_CONVNEXT_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_CONVNEXT_PW1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_CONVNEXT_PW2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_CONVNEXT_GAMMA, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
- {LLM_TENSOR_SHORTCONV_CONV, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_SSM_CONV}},
- {LLM_TENSOR_SHORTCONV_INPROJ, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_SHORTCONV_OUTPROJ, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- // NextN/MTP tensors are currently ignored (reserved for future MTP support)
- // These tensors only exist in the last layer(s) and are treated as output tensors
- {LLM_TENSOR_NEXTN_EH_PROJ, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_NEXTN_EMBED_TOKENS, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_GET_ROWS}},
- {LLM_TENSOR_NEXTN_ENORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_GET_ROWS}},
- {LLM_TENSOR_NEXTN_HNORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
- {LLM_TENSOR_NEXTN_SHARED_HEAD_HEAD, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
- {LLM_TENSOR_NEXTN_SHARED_HEAD_NORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
-};
+#include <mutex>
+
+// Forward declarations for accessor functions
+static const std::map<llm_arch, const char *>& get_llm_arch_names();
+static const std::map<llm_kv, const char *>& get_llm_kv_names();
+static const std::map<llm_tensor, llm_tensor_info>& get_llm_tensor_infos();
+
+const char* llama_arch_name(llm_arch arch) {
+ const auto& names = get_llm_arch_names();
+ auto it = names.find(arch);
+ if (it != names.end()) {
+ return it->second;
+ }
+ return "unknown";
+}
+
+const char* llama_get_kv_name(llm_kv kv) {
+ const auto& names = get_llm_kv_names();
+ auto it = names.find(kv);
+ if (it != names.end()) {
+ return it->second;
+ }
+ return "unknown";
+}
+
+const char* tensor_name(llm_arch arch, llm_tensor tensor) {
+ // This function will be updated to use an accessor, but for now use a
+ // basic approach since the map is huge and complex
+ static std::once_flag flag;
+ static std::map<llm_arch, std::map<llm_tensor, const char *>>* LLM_TENSOR_NAMES = nullptr;
+
+ std::call_once(flag, []() {
+ LLM_TENSOR_NAMES = new std::map<llm_arch, std::map<llm_tensor, const char *>>{
+ {
+ LLM_ARCH_LLAMA,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_GATE_EXP, "blk.%d.ffn_gate.%d" },
+ { LLM_TENSOR_FFN_DOWN_EXP, "blk.%d.ffn_down.%d" },
+ { LLM_TENSOR_FFN_UP_EXP, "blk.%d.ffn_up.%d" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ },
+ },
+ {
+ LLM_ARCH_ARCEE,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_LLAMA4,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_GATE_EXP, "blk.%d.ffn_gate.%d" },
+ { LLM_TENSOR_FFN_DOWN_EXP, "blk.%d.ffn_down.%d" },
+ { LLM_TENSOR_FFN_UP_EXP, "blk.%d.ffn_up.%d" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
+ { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
+ { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
+ },
+ },
+ {
+ LLM_ARCH_DECI,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_GATE_EXP, "blk.%d.ffn_gate.%d" },
+ { LLM_TENSOR_FFN_DOWN_EXP, "blk.%d.ffn_down.%d" },
+ { LLM_TENSOR_FFN_UP_EXP, "blk.%d.ffn_up.%d" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ },
+ },
+ {
+ LLM_ARCH_BAICHUAN,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_FALCON,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_NORM_2, "blk.%d.attn_norm_2" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_GROK,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE_EXP, "blk.%d.ffn_gate.%d" },
+ { LLM_TENSOR_FFN_DOWN_EXP, "blk.%d.ffn_down.%d" },
+ { LLM_TENSOR_FFN_UP_EXP, "blk.%d.ffn_up.%d" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ { LLM_TENSOR_LAYER_OUT_NORM, "blk.%d.layer_output_norm" },
+ { LLM_TENSOR_ATTN_OUT_NORM, "blk.%d.attn_output_norm" },
+ },
+ },
+ {
+ LLM_ARCH_GPT2,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_POS_EMBD, "position_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ },
+ },
+ {
+ LLM_ARCH_GPTJ,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ },
+ },
+ {
+ LLM_ARCH_GPTNEOX,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_MPT,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output"},
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_ACT, "blk.%d.ffn.act" },
+ { LLM_TENSOR_POS_EMBD, "position_embd" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm"},
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm"},
+ },
+ },
+ {
+ LLM_ARCH_STARCODER,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_POS_EMBD, "position_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ },
+ },
+ {
+ LLM_ARCH_REFACT,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_BERT,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
+ { LLM_TENSOR_TOKEN_TYPES, "token_types" },
+ { LLM_TENSOR_POS_EMBD, "position_embd" },
+ { LLM_TENSOR_ATTN_OUT_NORM, "blk.%d.attn_output_norm" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_LAYER_OUT_NORM, "blk.%d.layer_output_norm" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_CLS, "cls" },
+ { LLM_TENSOR_CLS_OUT, "cls.output" },
+ },
+ },
+ {
+ LLM_ARCH_NOMIC_BERT,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
+ { LLM_TENSOR_TOKEN_TYPES, "token_types" },
+ { LLM_TENSOR_ATTN_OUT_NORM, "blk.%d.attn_output_norm" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_LAYER_OUT_NORM, "blk.%d.layer_output_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_NOMIC_BERT_MOE,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
+ { LLM_TENSOR_TOKEN_TYPES, "token_types" },
+ { LLM_TENSOR_ATTN_OUT_NORM, "blk.%d.attn_output_norm" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_LAYER_OUT_NORM, "blk.%d.layer_output_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ },
+ },
+ {
+ LLM_ARCH_NEO_BERT,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_ENC_OUTPUT_NORM, "enc.output_norm" },
+ { LLM_TENSOR_CLS, "cls" },
+ { LLM_TENSOR_CLS_OUT, "cls.output" },
+ },
+ },
+ {
+ LLM_ARCH_JINA_BERT_V2,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
+ { LLM_TENSOR_TOKEN_TYPES, "token_types" },
+ { LLM_TENSOR_ATTN_NORM_2, "blk.%d.attn_norm_2" },
+ { LLM_TENSOR_ATTN_OUT_NORM, "blk.%d.attn_output_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_LAYER_OUT_NORM, "blk.%d.layer_output_norm" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_CLS, "cls" },
+ },
+ },
+ {
+ LLM_ARCH_BLOOM,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ },
+ },
+ {
+ LLM_ARCH_STABLELM,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
+ },
+ },
+ {
+ LLM_ARCH_QWEN,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_QWEN2,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_QWEN2VL,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_QWEN2MOE,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ { LLM_TENSOR_FFN_GATE_INP_SHEXP, "blk.%d.ffn_gate_inp_shexp" },
+ { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
+ { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
+ { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
+ },
+ },
+ {
+ LLM_ARCH_QWEN3,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_QWEN3MOE,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ },
+ },
+ {
+ LLM_ARCH_PHI2,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_PHI3,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FACTORS_LONG, "rope_factors_long" },
+ { LLM_TENSOR_ROPE_FACTORS_SHORT, "rope_factors_short" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_PHIMOE,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FACTORS_LONG, "rope_factors_long" },
+ { LLM_TENSOR_ROPE_FACTORS_SHORT, "rope_factors_short" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ },
+ },
+ {
+ LLM_ARCH_PLAMO,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_PLAMO2,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_SSM_IN, "blk.%d.ssm_in" },
+ { LLM_TENSOR_SSM_CONV1D, "blk.%d.ssm_conv1d" },
+ { LLM_TENSOR_SSM_X, "blk.%d.ssm_x" },
+ { LLM_TENSOR_SSM_DT, "blk.%d.ssm_dt" },
+ { LLM_TENSOR_SSM_A, "blk.%d.ssm_a" },
+ { LLM_TENSOR_SSM_D, "blk.%d.ssm_d" },
+ { LLM_TENSOR_SSM_OUT, "blk.%d.ssm_out" },
+ { LLM_TENSOR_SSM_DT_NORM, "blk.%d.ssm_dt_norm" },
+ { LLM_TENSOR_SSM_B_NORM, "blk.%d.ssm_b_norm" },
+ { LLM_TENSOR_SSM_C_NORM, "blk.%d.ssm_c_norm" },
+ { LLM_TENSOR_ATTN_POST_NORM, "blk.%d.post_attention_norm" },
+ { LLM_TENSOR_FFN_POST_NORM, "blk.%d.post_ffw_norm" },
+ },
+ },
+ {
+ LLM_ARCH_CODESHELL,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_ORION,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_INTERNLM2,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_MINICPM,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ROPE_FACTORS_LONG, "rope_factors_long" },
+ { LLM_TENSOR_ROPE_FACTORS_SHORT, "rope_factors_short" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_GATE_EXP, "blk.%d.ffn_gate.%d" },
+ { LLM_TENSOR_FFN_DOWN_EXP, "blk.%d.ffn_down.%d" },
+ { LLM_TENSOR_FFN_UP_EXP, "blk.%d.ffn_up.%d" },
+ },
+ },
+ {
+ LLM_ARCH_MINICPM3,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FACTORS_LONG, "rope_factors_long" },
+ { LLM_TENSOR_ROPE_FACTORS_SHORT, "rope_factors_short" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q_A_NORM, "blk.%d.attn_q_a_norm" },
+ { LLM_TENSOR_ATTN_KV_A_NORM, "blk.%d.attn_kv_a_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_Q_A, "blk.%d.attn_q_a" },
+ { LLM_TENSOR_ATTN_Q_B, "blk.%d.attn_q_b" },
+ { LLM_TENSOR_ATTN_KV_A_MQA, "blk.%d.attn_kv_a_mqa" },
+ { LLM_TENSOR_ATTN_KV_B, "blk.%d.attn_kv_b" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ },
+ },
+ {
+ LLM_ARCH_GEMMA,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_GEMMA2,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_POST_NORM, "blk.%d.post_attention_norm" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_POST_NORM, "blk.%d.post_ffw_norm" },
+ },
+ },
+ {
+ LLM_ARCH_GEMMA3,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_POST_NORM, "blk.%d.post_attention_norm" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_POST_NORM, "blk.%d.post_ffw_norm" },
+ },
+ },
+ {
+ LLM_ARCH_GEMMA3N,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_POST_NORM, "blk.%d.post_attention_norm" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_POST_NORM, "blk.%d.post_ffw_norm" },
+ { LLM_TENSOR_PER_LAYER_TOKEN_EMBD, "per_layer_token_embd" },
+ { LLM_TENSOR_PER_LAYER_MODEL_PROJ, "per_layer_model_proj" },
+ { LLM_TENSOR_PER_LAYER_PROJ_NORM, "per_layer_proj_norm" },
+ { LLM_TENSOR_ALTUP_UNEMBD_PROJ, "altup_unembd_proj" },
+ { LLM_TENSOR_ALTUP_PROJ, "altup_proj" },
+ { LLM_TENSOR_PER_LAYER_INP_GATE, "blk.%d.inp_gate" },
+ { LLM_TENSOR_PER_LAYER_PROJ, "blk.%d.proj" },
+ { LLM_TENSOR_PER_LAYER_POST_NORM, "blk.%d.post_norm" },
+ { LLM_TENSOR_ALTUP_CORRECT_COEF, "blk.%d.altup_correct_coef" },
+ { LLM_TENSOR_ALTUP_CORRECT_SCALE, "blk.%d.altup_correct_scale" },
+ { LLM_TENSOR_ALTUP_PREDICT_COEF, "blk.%d.altup_predict_coef" },
+ { LLM_TENSOR_ALTUP_ROUTER, "blk.%d.altup_router" },
+ { LLM_TENSOR_ALTUP_ROUTER_NORM, "blk.%d.altup_router_norm" },
+ { LLM_TENSOR_LAUREL_L, "blk.%d.laurel_l" },
+ { LLM_TENSOR_LAUREL_R, "blk.%d.laurel_r" },
+ { LLM_TENSOR_LAUREL_POST_NORM, "blk.%d.laurel_post_norm" },
+ },
+ },
+ {
+ LLM_ARCH_STARCODER2,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_MAMBA,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_SSM_IN, "blk.%d.ssm_in" },
+ { LLM_TENSOR_SSM_CONV1D, "blk.%d.ssm_conv1d" },
+ { LLM_TENSOR_SSM_X, "blk.%d.ssm_x" },
+ { LLM_TENSOR_SSM_DT, "blk.%d.ssm_dt" },
+ { LLM_TENSOR_SSM_A, "blk.%d.ssm_a" },
+ { LLM_TENSOR_SSM_D, "blk.%d.ssm_d" },
+ { LLM_TENSOR_SSM_OUT, "blk.%d.ssm_out" },
+ },
+ },
+ {
+ LLM_ARCH_MAMBA2,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_SSM_IN, "blk.%d.ssm_in" },
+ { LLM_TENSOR_SSM_CONV1D, "blk.%d.ssm_conv1d" },
+ { LLM_TENSOR_SSM_DT, "blk.%d.ssm_dt" },
+ { LLM_TENSOR_SSM_A, "blk.%d.ssm_a" },
+ { LLM_TENSOR_SSM_D, "blk.%d.ssm_d" },
+ { LLM_TENSOR_SSM_NORM, "blk.%d.ssm_norm" },
+ { LLM_TENSOR_SSM_OUT, "blk.%d.ssm_out" },
+ },
+ },
+ {
+ LLM_ARCH_JAMBA,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_SSM_IN, "blk.%d.ssm_in" },
+ { LLM_TENSOR_SSM_CONV1D, "blk.%d.ssm_conv1d" },
+ { LLM_TENSOR_SSM_X, "blk.%d.ssm_x" },
+ { LLM_TENSOR_SSM_DT, "blk.%d.ssm_dt" },
+ { LLM_TENSOR_SSM_DT_NORM, "blk.%d.ssm_dt_norm" },
+ { LLM_TENSOR_SSM_A, "blk.%d.ssm_a" },
+ { LLM_TENSOR_SSM_B_NORM, "blk.%d.ssm_b_norm" },
+ { LLM_TENSOR_SSM_C_NORM, "blk.%d.ssm_c_norm" },
+ { LLM_TENSOR_SSM_D, "blk.%d.ssm_d" },
+ { LLM_TENSOR_SSM_OUT, "blk.%d.ssm_out" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ },
+ },
+ {
+ LLM_ARCH_FALCON_H1,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_SSM_IN, "blk.%d.ssm_in" },
+ { LLM_TENSOR_SSM_CONV1D, "blk.%d.ssm_conv1d" },
+ { LLM_TENSOR_SSM_DT, "blk.%d.ssm_dt" },
+ { LLM_TENSOR_SSM_A, "blk.%d.ssm_a" },
+ { LLM_TENSOR_SSM_D, "blk.%d.ssm_d" },
+ { LLM_TENSOR_SSM_NORM, "blk.%d.ssm_norm" },
+ { LLM_TENSOR_SSM_OUT, "blk.%d.ssm_out" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_XVERSE,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_COMMAND_R,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
+ },
+ },
+ {
+ LLM_ARCH_COHERE2,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_DBRX,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_OUT_NORM, "blk.%d.attn_output_norm" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ },
+ },
+ {
+ LLM_ARCH_OLMO,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_OLMO2,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_POST_NORM, "blk.%d.post_attention_norm" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
+ { LLM_TENSOR_FFN_POST_NORM, "blk.%d.post_ffw_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_OLMOE,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ },
+ },
+ {
+ LLM_ARCH_OPENELM,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_ARCTIC,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_NORM_EXPS, "blk.%d.ffn_norm_exps" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ },
+ },
+ {
+ LLM_ARCH_DEEPSEEK,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ { LLM_TENSOR_FFN_GATE_INP_SHEXP, "blk.%d.ffn_gate_inp_shexp" },
+ { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
+ { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
+ { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
+ },
+ },
+ {
+ LLM_ARCH_DEEPSEEK2,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q_A_NORM, "blk.%d.attn_q_a_norm" },
+ { LLM_TENSOR_ATTN_KV_A_NORM, "blk.%d.attn_kv_a_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_Q_A, "blk.%d.attn_q_a" },
+ { LLM_TENSOR_ATTN_Q_B, "blk.%d.attn_q_b" },
+ { LLM_TENSOR_ATTN_KV_A_MQA, "blk.%d.attn_kv_a_mqa" },
+ { LLM_TENSOR_ATTN_KV_B, "blk.%d.attn_kv_b" },
+ { LLM_TENSOR_ATTN_K_B, "blk.%d.attn_k_b" },
+ { LLM_TENSOR_ATTN_V_B, "blk.%d.attn_v_b" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ { LLM_TENSOR_FFN_GATE_INP_SHEXP, "blk.%d.ffn_gate_inp_shexp" },
+ { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
+ { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
+ { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
+ { LLM_TENSOR_FFN_EXP_PROBS_B, "blk.%d.exp_probs_b" },
+ },
+ },
+ {
+ LLM_ARCH_PLM,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_KV_A_MQA, "blk.%d.attn_kv_a_mqa" },
+ { LLM_TENSOR_ATTN_KV_A_NORM, "blk.%d.attn_kv_a_norm" },
+ { LLM_TENSOR_ATTN_KV_B, "blk.%d.attn_kv_b" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_CHATGLM,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ },
+ },
+ {
+ LLM_ARCH_GLM4,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_ATTN_POST_NORM, "blk.%d.post_attention_norm" },
+ { LLM_TENSOR_FFN_POST_NORM, "blk.%d.post_ffw_norm" },
+ },
+ },
+ {
+ LLM_ARCH_GLM4_MOE,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_POST_NORM, "blk.%d.post_attention_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
+ { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
+ { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
+ { LLM_TENSOR_FFN_EXP_PROBS_B, "blk.%d.exp_probs_b" },
+ // NextN/MTP tensors - preserved but unused (in final layer, dynamic layer number)
+ { LLM_TENSOR_NEXTN_EH_PROJ, "blk.%d.nextn.eh_proj" },
+ { LLM_TENSOR_NEXTN_EMBED_TOKENS, "blk.%d.nextn.embed_tokens" },
+ { LLM_TENSOR_NEXTN_ENORM, "blk.%d.nextn.enorm" },
+ { LLM_TENSOR_NEXTN_HNORM, "blk.%d.nextn.hnorm" },
+ { LLM_TENSOR_NEXTN_SHARED_HEAD_HEAD, "blk.%d.nextn.shared_head_head" },
+ { LLM_TENSOR_NEXTN_SHARED_HEAD_NORM, "blk.%d.nextn.shared_head_norm" },
+ },
+ },
+ {
+ LLM_ARCH_BITNET,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_SUB_NORM, "blk.%d.attn_sub_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_SUB_NORM, "blk.%d.ffn_sub_norm" },
+ },
+ },
+ {
+ LLM_ARCH_T5,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_DEC_OUTPUT_NORM, "dec.output_norm" },
+ { LLM_TENSOR_DEC_ATTN_NORM, "dec.blk.%d.attn_norm" },
+ { LLM_TENSOR_DEC_ATTN_Q, "dec.blk.%d.attn_q" },
+ { LLM_TENSOR_DEC_ATTN_K, "dec.blk.%d.attn_k" },
+ { LLM_TENSOR_DEC_ATTN_V, "dec.blk.%d.attn_v" },
+ { LLM_TENSOR_DEC_ATTN_OUT, "dec.blk.%d.attn_o" },
+ { LLM_TENSOR_DEC_ATTN_REL_B, "dec.blk.%d.attn_rel_b" },
+ { LLM_TENSOR_DEC_CROSS_ATTN_NORM, "dec.blk.%d.cross_attn_norm" },
+ { LLM_TENSOR_DEC_CROSS_ATTN_Q, "dec.blk.%d.cross_attn_q" },
+ { LLM_TENSOR_DEC_CROSS_ATTN_K, "dec.blk.%d.cross_attn_k" },
+ { LLM_TENSOR_DEC_CROSS_ATTN_V, "dec.blk.%d.cross_attn_v" },
+ { LLM_TENSOR_DEC_CROSS_ATTN_OUT, "dec.blk.%d.cross_attn_o" },
+ { LLM_TENSOR_DEC_CROSS_ATTN_REL_B, "dec.blk.%d.cross_attn_rel_b" },
+ { LLM_TENSOR_DEC_FFN_NORM, "dec.blk.%d.ffn_norm" },
+ { LLM_TENSOR_DEC_FFN_GATE, "dec.blk.%d.ffn_gate" },
+ { LLM_TENSOR_DEC_FFN_DOWN, "dec.blk.%d.ffn_down" },
+ { LLM_TENSOR_DEC_FFN_UP, "dec.blk.%d.ffn_up" },
+ { LLM_TENSOR_ENC_OUTPUT_NORM, "enc.output_norm" },
+ { LLM_TENSOR_ENC_ATTN_NORM, "enc.blk.%d.attn_norm" },
+ { LLM_TENSOR_ENC_ATTN_Q, "enc.blk.%d.attn_q" },
+ { LLM_TENSOR_ENC_ATTN_K, "enc.blk.%d.attn_k" },
+ { LLM_TENSOR_ENC_ATTN_V, "enc.blk.%d.attn_v" },
+ { LLM_TENSOR_ENC_ATTN_OUT, "enc.blk.%d.attn_o" },
+ { LLM_TENSOR_ENC_ATTN_REL_B, "enc.blk.%d.attn_rel_b" },
+ { LLM_TENSOR_ENC_FFN_NORM, "enc.blk.%d.ffn_norm" },
+ { LLM_TENSOR_ENC_FFN_GATE, "enc.blk.%d.ffn_gate" },
+ { LLM_TENSOR_ENC_FFN_DOWN, "enc.blk.%d.ffn_down" },
+ { LLM_TENSOR_ENC_FFN_UP, "enc.blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_T5ENCODER,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ENC_OUTPUT_NORM, "enc.output_norm" },
+ { LLM_TENSOR_ENC_ATTN_NORM, "enc.blk.%d.attn_norm" },
+ { LLM_TENSOR_ENC_ATTN_Q, "enc.blk.%d.attn_q" },
+ { LLM_TENSOR_ENC_ATTN_K, "enc.blk.%d.attn_k" },
+ { LLM_TENSOR_ENC_ATTN_V, "enc.blk.%d.attn_v" },
+ { LLM_TENSOR_ENC_ATTN_OUT, "enc.blk.%d.attn_o" },
+ { LLM_TENSOR_ENC_ATTN_REL_B, "enc.blk.%d.attn_rel_b" },
+ { LLM_TENSOR_ENC_FFN_NORM, "enc.blk.%d.ffn_norm" },
+ { LLM_TENSOR_ENC_FFN_GATE, "enc.blk.%d.ffn_gate" },
+ { LLM_TENSOR_ENC_FFN_DOWN, "enc.blk.%d.ffn_down" },
+ { LLM_TENSOR_ENC_FFN_UP, "enc.blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_JAIS,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ },
+ },
+ {
+ LLM_ARCH_NEMOTRON,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_EXAONE,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_EXAONE4,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_POST_NORM, "blk.%d.post_attention_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_POST_NORM, "blk.%d.post_ffw_norm" },
+ }
+ },
+ {
+ LLM_ARCH_RWKV6,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_NORM_2, "blk.%d.attn_norm_2" },
+ { LLM_TENSOR_TIME_MIX_W1, "blk.%d.time_mix_w1" },
+ { LLM_TENSOR_TIME_MIX_W2, "blk.%d.time_mix_w2" },
+ { LLM_TENSOR_TIME_MIX_LERP_X, "blk.%d.time_mix_lerp_x" },
+ { LLM_TENSOR_TIME_MIX_LERP_W, "blk.%d.time_mix_lerp_w" },
+ { LLM_TENSOR_TIME_MIX_LERP_K, "blk.%d.time_mix_lerp_k" },
+ { LLM_TENSOR_TIME_MIX_LERP_V, "blk.%d.time_mix_lerp_v" },
+ { LLM_TENSOR_TIME_MIX_LERP_R, "blk.%d.time_mix_lerp_r" },
+ { LLM_TENSOR_TIME_MIX_LERP_G, "blk.%d.time_mix_lerp_g" },
+ { LLM_TENSOR_TIME_MIX_LERP_FUSED, "blk.%d.time_mix_lerp_fused" },
+ { LLM_TENSOR_TIME_MIX_FIRST, "blk.%d.time_mix_first" },
+ { LLM_TENSOR_TIME_MIX_DECAY, "blk.%d.time_mix_decay" },
+ { LLM_TENSOR_TIME_MIX_DECAY_W1, "blk.%d.time_mix_decay_w1" },
+ { LLM_TENSOR_TIME_MIX_DECAY_W2, "blk.%d.time_mix_decay_w2" },
+ { LLM_TENSOR_TIME_MIX_KEY, "blk.%d.time_mix_key" },
+ { LLM_TENSOR_TIME_MIX_VALUE, "blk.%d.time_mix_value" },
+ { LLM_TENSOR_TIME_MIX_RECEPTANCE, "blk.%d.time_mix_receptance" },
+ { LLM_TENSOR_TIME_MIX_GATE, "blk.%d.time_mix_gate" },
+ { LLM_TENSOR_TIME_MIX_LN, "blk.%d.time_mix_ln" },
+ { LLM_TENSOR_TIME_MIX_OUTPUT, "blk.%d.time_mix_output" },
+ { LLM_TENSOR_CHANNEL_MIX_LERP_K, "blk.%d.channel_mix_lerp_k" },
+ { LLM_TENSOR_CHANNEL_MIX_LERP_R, "blk.%d.channel_mix_lerp_r" },
+ { LLM_TENSOR_CHANNEL_MIX_KEY, "blk.%d.channel_mix_key" },
+ { LLM_TENSOR_CHANNEL_MIX_VALUE, "blk.%d.channel_mix_value" },
+ { LLM_TENSOR_CHANNEL_MIX_RECEPTANCE, "blk.%d.channel_mix_receptance" },
+ },
+ },
+ {
+ LLM_ARCH_RWKV6QWEN2,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_TIME_MIX_W1, "blk.%d.time_mix_w1" },
+ { LLM_TENSOR_TIME_MIX_W2, "blk.%d.time_mix_w2" },
+ { LLM_TENSOR_TIME_MIX_LERP_X, "blk.%d.time_mix_lerp_x" },
+ { LLM_TENSOR_TIME_MIX_LERP_FUSED, "blk.%d.time_mix_lerp_fused" },
+ { LLM_TENSOR_TIME_MIX_FIRST, "blk.%d.time_mix_first" },
+ { LLM_TENSOR_TIME_MIX_DECAY, "blk.%d.time_mix_decay" },
+ { LLM_TENSOR_TIME_MIX_DECAY_W1, "blk.%d.time_mix_decay_w1" },
+ { LLM_TENSOR_TIME_MIX_DECAY_W2, "blk.%d.time_mix_decay_w2" },
+ { LLM_TENSOR_TIME_MIX_KEY, "blk.%d.time_mix_key" },
+ { LLM_TENSOR_TIME_MIX_VALUE, "blk.%d.time_mix_value" },
+ { LLM_TENSOR_TIME_MIX_RECEPTANCE, "blk.%d.time_mix_receptance" },
+ { LLM_TENSOR_TIME_MIX_GATE, "blk.%d.time_mix_gate" },
+ { LLM_TENSOR_TIME_MIX_OUTPUT, "blk.%d.time_mix_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_RWKV7,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_NORM_2, "blk.%d.attn_norm_2" },
+ { LLM_TENSOR_TIME_MIX_W0, "blk.%d.time_mix_w0" },
+ { LLM_TENSOR_TIME_MIX_W1, "blk.%d.time_mix_w1" },
+ { LLM_TENSOR_TIME_MIX_W2, "blk.%d.time_mix_w2" },
+ { LLM_TENSOR_TIME_MIX_A0, "blk.%d.time_mix_a0" },
+ { LLM_TENSOR_TIME_MIX_A1, "blk.%d.time_mix_a1" },
+ { LLM_TENSOR_TIME_MIX_A2, "blk.%d.time_mix_a2" },
+ { LLM_TENSOR_TIME_MIX_V0, "blk.%d.time_mix_v0" },
+ { LLM_TENSOR_TIME_MIX_V1, "blk.%d.time_mix_v1" },
+ { LLM_TENSOR_TIME_MIX_V2, "blk.%d.time_mix_v2" },
+ { LLM_TENSOR_TIME_MIX_G1, "blk.%d.time_mix_g1" },
+ { LLM_TENSOR_TIME_MIX_G2, "blk.%d.time_mix_g2" },
+ { LLM_TENSOR_TIME_MIX_K_K, "blk.%d.time_mix_k_k" },
+ { LLM_TENSOR_TIME_MIX_K_A, "blk.%d.time_mix_k_a" },
+ { LLM_TENSOR_TIME_MIX_R_K, "blk.%d.time_mix_r_k" },
+ { LLM_TENSOR_TIME_MIX_LERP_FUSED, "blk.%d.time_mix_lerp_fused" },
+ { LLM_TENSOR_TIME_MIX_KEY, "blk.%d.time_mix_key" },
+ { LLM_TENSOR_TIME_MIX_VALUE, "blk.%d.time_mix_value" },
+ { LLM_TENSOR_TIME_MIX_RECEPTANCE, "blk.%d.time_mix_receptance" },
+ { LLM_TENSOR_TIME_MIX_LN, "blk.%d.time_mix_ln" },
+ { LLM_TENSOR_TIME_MIX_OUTPUT, "blk.%d.time_mix_output" },
+ { LLM_TENSOR_CHANNEL_MIX_LERP_K, "blk.%d.channel_mix_lerp_k" },
+ { LLM_TENSOR_CHANNEL_MIX_KEY, "blk.%d.channel_mix_key" },
+ { LLM_TENSOR_CHANNEL_MIX_VALUE, "blk.%d.channel_mix_value" },
+ },
+ },
+ {
+ LLM_ARCH_ARWKV7,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_TIME_MIX_W0, "blk.%d.time_mix_w0" },
+ { LLM_TENSOR_TIME_MIX_W1, "blk.%d.time_mix_w1" },
+ { LLM_TENSOR_TIME_MIX_W2, "blk.%d.time_mix_w2" },
+ { LLM_TENSOR_TIME_MIX_A0, "blk.%d.time_mix_a0" },
+ { LLM_TENSOR_TIME_MIX_A1, "blk.%d.time_mix_a1" },
+ { LLM_TENSOR_TIME_MIX_A2, "blk.%d.time_mix_a2" },
+ { LLM_TENSOR_TIME_MIX_V0, "blk.%d.time_mix_v0" },
+ { LLM_TENSOR_TIME_MIX_V1, "blk.%d.time_mix_v1" },
+ { LLM_TENSOR_TIME_MIX_V2, "blk.%d.time_mix_v2" },
+ { LLM_TENSOR_TIME_MIX_G1, "blk.%d.time_mix_g1" },
+ { LLM_TENSOR_TIME_MIX_G2, "blk.%d.time_mix_g2" },
+ { LLM_TENSOR_TIME_MIX_K_K, "blk.%d.time_mix_k_k" },
+ { LLM_TENSOR_TIME_MIX_K_A, "blk.%d.time_mix_k_a" },
+ { LLM_TENSOR_TIME_MIX_R_K, "blk.%d.time_mix_r_k" },
+ { LLM_TENSOR_TIME_MIX_LERP_FUSED, "blk.%d.time_mix_lerp_fused" },
+ { LLM_TENSOR_TIME_MIX_KEY, "blk.%d.time_mix_key" },
+ { LLM_TENSOR_TIME_MIX_VALUE, "blk.%d.time_mix_value" },
+ { LLM_TENSOR_TIME_MIX_RECEPTANCE, "blk.%d.time_mix_receptance" },
+ { LLM_TENSOR_TIME_MIX_LN, "blk.%d.time_mix_ln" },
+ { LLM_TENSOR_TIME_MIX_OUTPUT, "blk.%d.time_mix_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_GRANITE,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_GRANITE_MOE,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
+ { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
+ { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
+ },
+ },
+ {
+ LLM_ARCH_GRANITE_HYBRID,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ // mamba(2) ssm layers
+ { LLM_TENSOR_SSM_IN, "blk.%d.ssm_in" },
+ { LLM_TENSOR_SSM_CONV1D, "blk.%d.ssm_conv1d" },
+ { LLM_TENSOR_SSM_DT, "blk.%d.ssm_dt" },
+ { LLM_TENSOR_SSM_A, "blk.%d.ssm_a" },
+ { LLM_TENSOR_SSM_D, "blk.%d.ssm_d" },
+ { LLM_TENSOR_SSM_NORM, "blk.%d.ssm_norm" },
+ { LLM_TENSOR_SSM_OUT, "blk.%d.ssm_out" },
+ // attention layers
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ // dense FFN
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ // moe FFN
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ // shared expert
+ { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
+ { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
+ { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
+ },
+ },
+ {
+ LLM_ARCH_CHAMELEON,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
+ },
+ },
+ {
+ LLM_ARCH_WAVTOKENIZER_DEC,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
+ { LLM_TENSOR_CONV1D, "conv1d" },
+ { LLM_TENSOR_CONVNEXT_DW, "convnext.%d.dw" },
+ { LLM_TENSOR_CONVNEXT_NORM, "convnext.%d.norm" },
+ { LLM_TENSOR_CONVNEXT_PW1, "convnext.%d.pw1" },
+ { LLM_TENSOR_CONVNEXT_PW2, "convnext.%d.pw2" },
+ { LLM_TENSOR_CONVNEXT_GAMMA, "convnext.%d.gamma" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_POS_NET_CONV1, "posnet.%d.conv1" },
+ { LLM_TENSOR_POS_NET_CONV2, "posnet.%d.conv2" },
+ { LLM_TENSOR_POS_NET_NORM, "posnet.%d.norm" },
+ { LLM_TENSOR_POS_NET_NORM1, "posnet.%d.norm1" },
+ { LLM_TENSOR_POS_NET_NORM2, "posnet.%d.norm2" },
+ { LLM_TENSOR_POS_NET_ATTN_NORM, "posnet.%d.attn_norm" },
+ { LLM_TENSOR_POS_NET_ATTN_Q, "posnet.%d.attn_q" },
+ { LLM_TENSOR_POS_NET_ATTN_K, "posnet.%d.attn_k" },
+ { LLM_TENSOR_POS_NET_ATTN_V, "posnet.%d.attn_v" },
+ { LLM_TENSOR_POS_NET_ATTN_OUT, "posnet.%d.attn_output" },
+ },
+ },
+ {
+ LLM_ARCH_BAILINGMOE,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ROPE_FREQS, "rope_freqs" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ { LLM_TENSOR_FFN_GATE_INP_SHEXP, "blk.%d.ffn_gate_inp_shexp" },
+ { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
+ { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
+ { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
+ },
+ },
+ {
+ LLM_ARCH_DOTS1,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ { LLM_TENSOR_FFN_GATE_INP_SHEXP, "blk.%d.ffn_gate_inp_shexp" },
+ { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
+ { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
+ { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
+ { LLM_TENSOR_FFN_EXP_PROBS_B, "blk.%d.exp_probs_b" },
+ }
+ },
+ {
+ LLM_ARCH_ERNIE4_5,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_ERNIE4_5_MOE,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
+ { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
+ { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ { LLM_TENSOR_FFN_EXP_PROBS_B, "blk.%d.exp_probs_b" },
+ },
+ },
+ {
+ LLM_ARCH_HUNYUAN_MOE,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
+ { LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
+ { LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ },
+ },
+ {
+ LLM_ARCH_HUNYUAN_DENSE,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+
+ },
+ },
+ {
+ LLM_ARCH_SMOLLM3,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_OPENAI_MOE,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_POST_NORM, "blk.%d.post_attention_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_SINKS, "blk.%d.attn_sinks" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
+ },
+ },
+ {
+ LLM_ARCH_LFM2,
+ {
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm" },
+ { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_SHORTCONV_CONV, "blk.%d.shortconv.conv" },
+ { LLM_TENSOR_SHORTCONV_INPROJ, "blk.%d.shortconv.in_proj" },
+ { LLM_TENSOR_SHORTCONV_OUTPROJ, "blk.%d.shortconv.out_proj" },
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_TOKEN_EMBD_NORM, "token_embd_norm" },
+ }
+ },
+ {
+ LLM_ARCH_SMALLTHINKER,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ { LLM_TENSOR_FFN_GATE_INP, "blk.%d.ffn_gate_inp" },
+ { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
+ { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
+ { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" }
+ },
+ },
+ {
+ LLM_ARCH_DREAM,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_LLADA,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ { LLM_TENSOR_OUTPUT_NORM, "output_norm" },
+ { LLM_TENSOR_OUTPUT, "output" },
+ { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
+ { LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
+ { LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
+ { LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
+ { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
+ { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
+ { LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
+ { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
+ { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
+ },
+ },
+ {
+ LLM_ARCH_UNKNOWN,
+ {
+ { LLM_TENSOR_TOKEN_EMBD, "token_embd" },
+ },
+ },
+ };
+
+ static const std::map<llm_tensor, llm_tensor_info> LLM_TENSOR_INFOS = {
+ {LLM_TENSOR_TOKEN_EMBD, {LLM_TENSOR_LAYER_INPUT, GGML_OP_GET_ROWS}},
+ {LLM_TENSOR_POS_EMBD, {LLM_TENSOR_LAYER_INPUT, GGML_OP_GET_ROWS}},
+ {LLM_TENSOR_TOKEN_EMBD_NORM, {LLM_TENSOR_LAYER_INPUT, GGML_OP_GET_ROWS}},
+ {LLM_TENSOR_TOKEN_TYPES, {LLM_TENSOR_LAYER_INPUT, GGML_OP_GET_ROWS}},
+ {LLM_TENSOR_OUTPUT, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_CLS, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_CLS_OUT, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_OUTPUT_NORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
+ {LLM_TENSOR_DEC_OUTPUT_NORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
+ {LLM_TENSOR_ENC_OUTPUT_NORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
+ {LLM_TENSOR_ROPE_FREQS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ROPE}},
+ {LLM_TENSOR_ROPE_FACTORS_LONG, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ROPE}},
+ {LLM_TENSOR_ROPE_FACTORS_SHORT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ROPE}},
+ {LLM_TENSOR_ATTN_Q, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_V, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_QKV, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_OUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_FFN_GATE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_FFN_DOWN, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_FFN_UP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_FFN_DOWN_SHEXP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_FFN_GATE_SHEXP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_FFN_UP_SHEXP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_Q_A, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_Q_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_KV_A_MQA, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_KV_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_K_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_V_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_SINKS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_SCALE}},
+ {LLM_TENSOR_DEC_ATTN_Q, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_ATTN_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_ATTN_V, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_ATTN_OUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_CROSS_ATTN_Q, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_CROSS_ATTN_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_CROSS_ATTN_V, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_CROSS_ATTN_OUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_FFN_GATE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_FFN_DOWN, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_FFN_UP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ENC_ATTN_Q, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ENC_ATTN_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ENC_ATTN_V, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ENC_ATTN_OUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ENC_FFN_GATE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ENC_FFN_DOWN, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ENC_FFN_UP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_FFN_GATE_INP_SHEXP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_FFN_GATE_INP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_SSM_IN, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_SSM_X, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_SSM_DT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_SSM_OUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_W1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_W2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_A1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_A2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_V1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_V2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_G1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_G2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_DECAY_W1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_DECAY_W2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_KEY, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_VALUE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_RECEPTANCE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_GATE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_OUTPUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_CHANNEL_MIX_KEY, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_CHANNEL_MIX_RECEPTANCE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_CHANNEL_MIX_VALUE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_FFN_ACT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_DIV}},
+ {LLM_TENSOR_SSM_CONV1D, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_SSM_CONV}},
+ {LLM_TENSOR_SSM_A, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_SSM_SCAN}},
+ {LLM_TENSOR_SSM_DT_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_SSM_B_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_SSM_C_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_SSM_D, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_SSM_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_TIME_MIX_LERP_X, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_TIME_MIX_LN, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_CHANNEL_MIX_LERP_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_CHANNEL_MIX_LERP_R, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_TIME_MIX_K_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_TIME_MIX_K_A, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_TIME_MIX_R_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_TIME_MIX_LERP_W, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_LERP_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_LERP_V, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_LERP_R, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_LERP_G, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_LERP_FUSED, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_DECAY, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_W0, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_A0, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_V0, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_FIRST, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_RWKV_WKV6}},
+ {LLM_TENSOR_ATTN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ATTN_NORM_2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ATTN_OUT_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ATTN_POST_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_FFN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_FFN_POST_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_FFN_NORM_EXPS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ATTN_Q_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ATTN_K_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_LAYER_OUT_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ATTN_Q_A_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ATTN_KV_A_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ATTN_SUB_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_FFN_SUB_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_DEC_ATTN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_DEC_CROSS_ATTN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_DEC_FFN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ENC_ATTN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ENC_FFN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_DEC_ATTN_REL_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_GET_ROWS}},
+ {LLM_TENSOR_ENC_ATTN_REL_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_GET_ROWS}},
+ {LLM_TENSOR_FFN_DOWN_EXPS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT_ID}},
+ {LLM_TENSOR_FFN_GATE_EXPS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT_ID}},
+ {LLM_TENSOR_FFN_UP_EXPS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT_ID}},
+ {LLM_TENSOR_FFN_EXP_PROBS_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ // altup / laurel (gemma 3n)
+ {LLM_TENSOR_PER_LAYER_TOKEN_EMBD, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_GET_ROWS}},
+ {LLM_TENSOR_PER_LAYER_MODEL_PROJ, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_PER_LAYER_PROJ_NORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
+ {LLM_TENSOR_ALTUP_PROJ, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ALTUP_UNEMBD_PROJ, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_PER_LAYER_INP_GATE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_PER_LAYER_PROJ, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_PER_LAYER_POST_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ALTUP_CORRECT_COEF, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ALTUP_CORRECT_SCALE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ALTUP_PREDICT_COEF, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ALTUP_ROUTER, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ALTUP_ROUTER_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_LAUREL_L, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_LAUREL_R, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_LAUREL_POST_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ // this tensor is loaded for T5, but never used
+ {LLM_TENSOR_DEC_CROSS_ATTN_REL_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_NONE}},
+ {LLM_TENSOR_CONV1D, {LLM_TENSOR_LAYER_INPUT, GGML_OP_IM2COL}},
+ {LLM_TENSOR_POS_NET_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_POS_NET_NORM1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_POS_NET_NORM2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_POS_NET_CONV1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_IM2COL}},
+ {LLM_TENSOR_POS_NET_CONV2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_IM2COL}},
+ {LLM_TENSOR_POS_NET_ATTN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_POS_NET_ATTN_Q, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_POS_NET_ATTN_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_POS_NET_ATTN_V, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_POS_NET_ATTN_OUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_CONVNEXT_DW, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_IM2COL}},
+ {LLM_TENSOR_CONVNEXT_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_CONVNEXT_PW1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_CONVNEXT_PW2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_CONVNEXT_GAMMA, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_SHORTCONV_CONV, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_SSM_CONV}},
+ {LLM_TENSOR_SHORTCONV_INPROJ, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_SHORTCONV_OUTPROJ, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ // NextN/MTP tensors are currently ignored (reserved for future MTP support)
+ // These tensors only exist in the last layer(s) and are treated as output tensors
+ {LLM_TENSOR_NEXTN_EH_PROJ, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_NEXTN_EMBED_TOKENS, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_GET_ROWS}},
+ {LLM_TENSOR_NEXTN_ENORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_GET_ROWS}},
+ {LLM_TENSOR_NEXTN_HNORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
+ {LLM_TENSOR_NEXTN_SHARED_HEAD_HEAD, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_NEXTN_SHARED_HEAD_NORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
+ };
+ });
+
+ auto it = LLM_TENSOR_NAMES->find(arch);
+ if (it != LLM_TENSOR_NAMES->end()) {
+ if (it->second.find(tensor) != it->second.end()) {
+ return it->second.find(tensor)->second;
+ }
+ }
+ return "unknown";
+}
+
+// Accessor functions for lazy-initialized static maps
+static const std::map<llm_arch, const char *>& get_llm_arch_names() {
+ static std::once_flag flag;
+ static std::map<llm_arch, const char *>* LLM_ARCH_NAMES = nullptr;
+
+ std::call_once(flag, []() {
+ LLM_ARCH_NAMES = new std::map<llm_arch, const char *>{
+ { LLM_ARCH_LLAMA, "llama" },
+ { LLM_ARCH_LLAMA4, "llama4" },
+ { LLM_ARCH_DECI, "deci" },
+ { LLM_ARCH_FALCON, "falcon" },
+ { LLM_ARCH_GROK, "grok" },
+ { LLM_ARCH_GPT2, "gpt2" },
+ { LLM_ARCH_GPTJ, "gptj" },
+ { LLM_ARCH_GPTNEOX, "gptneox" },
+ { LLM_ARCH_MPT, "mpt" },
+ { LLM_ARCH_BAICHUAN, "baichuan" },
+ { LLM_ARCH_STARCODER, "starcoder" },
+ { LLM_ARCH_REFACT, "refact" },
+ { LLM_ARCH_BERT, "bert" },
+ { LLM_ARCH_NOMIC_BERT, "nomic-bert" },
+ { LLM_ARCH_NOMIC_BERT_MOE, "nomic-bert-moe" },
+ { LLM_ARCH_NEO_BERT, "neo-bert" },
+ { LLM_ARCH_JINA_BERT_V2, "jina-bert-v2" },
+ { LLM_ARCH_BLOOM, "bloom" },
+ { LLM_ARCH_STABLELM, "stablelm" },
+ { LLM_ARCH_QWEN, "qwen" },
+ { LLM_ARCH_QWEN2, "qwen2" },
+ { LLM_ARCH_QWEN2MOE, "qwen2moe" },
+ { LLM_ARCH_QWEN2VL, "qwen2vl" },
+ { LLM_ARCH_QWEN3, "qwen3" },
+ { LLM_ARCH_QWEN3MOE, "qwen3moe" },
+ { LLM_ARCH_PHI2, "phi2" },
+ { LLM_ARCH_PHI3, "phi3" },
+ { LLM_ARCH_PHIMOE, "phimoe" },
+ { LLM_ARCH_PLAMO, "plamo" },
+ { LLM_ARCH_PLAMO2, "plamo2" },
+ { LLM_ARCH_CODESHELL, "codeshell" },
+ { LLM_ARCH_ORION, "orion" },
+ { LLM_ARCH_INTERNLM2, "internlm2" },
+ { LLM_ARCH_MINICPM, "minicpm" },
+ { LLM_ARCH_MINICPM3, "minicpm3" },
+ { LLM_ARCH_GEMMA, "gemma" },
+ { LLM_ARCH_GEMMA2, "gemma2" },
+ { LLM_ARCH_GEMMA3, "gemma3" },
+ { LLM_ARCH_GEMMA3N, "gemma3n" },
+ { LLM_ARCH_STARCODER2, "starcoder2" },
+ { LLM_ARCH_MAMBA, "mamba" },
+ { LLM_ARCH_MAMBA2, "mamba2" },
+ { LLM_ARCH_JAMBA, "jamba" },
+ { LLM_ARCH_FALCON_H1, "falcon-h1" },
+ { LLM_ARCH_XVERSE, "xverse" },
+ { LLM_ARCH_COMMAND_R, "command-r" },
+ { LLM_ARCH_COHERE2, "cohere2" },
+ { LLM_ARCH_DBRX, "dbrx" },
+ { LLM_ARCH_OLMO, "olmo" },
+ { LLM_ARCH_OLMO2, "olmo2" },
+ { LLM_ARCH_OLMOE, "olmoe" },
+ { LLM_ARCH_OPENELM, "openelm" },
+ { LLM_ARCH_ARCTIC, "arctic" },
+ { LLM_ARCH_DEEPSEEK, "deepseek" },
+ { LLM_ARCH_DEEPSEEK2, "deepseek2" },
+ { LLM_ARCH_CHATGLM, "chatglm" },
+ { LLM_ARCH_GLM4, "glm4" },
+ { LLM_ARCH_GLM4_MOE, "glm4moe" },
+ { LLM_ARCH_BITNET, "bitnet" },
+ { LLM_ARCH_T5, "t5" },
+ { LLM_ARCH_T5ENCODER, "t5encoder" },
+ { LLM_ARCH_JAIS, "jais" },
+ { LLM_ARCH_NEMOTRON, "nemotron" },
+ { LLM_ARCH_EXAONE, "exaone" },
+ { LLM_ARCH_EXAONE4, "exaone4" },
+ { LLM_ARCH_RWKV6, "rwkv6" },
+ { LLM_ARCH_RWKV6QWEN2, "rwkv6qwen2" },
+ { LLM_ARCH_RWKV7, "rwkv7" },
+ { LLM_ARCH_ARWKV7, "arwkv7" },
+ { LLM_ARCH_GRANITE, "granite" },
+ { LLM_ARCH_GRANITE_MOE, "granitemoe" },
+ { LLM_ARCH_GRANITE_HYBRID, "granitehybrid" },
+ { LLM_ARCH_CHAMELEON, "chameleon" },
+ { LLM_ARCH_WAVTOKENIZER_DEC, "wavtokenizer-dec" },
+ { LLM_ARCH_PLM, "plm" },
+ { LLM_ARCH_BAILINGMOE, "bailingmoe" },
+ { LLM_ARCH_DOTS1, "dots1" },
+ { LLM_ARCH_ARCEE, "arcee" },
+ { LLM_ARCH_ERNIE4_5, "ernie4_5" },
+ { LLM_ARCH_ERNIE4_5_MOE, "ernie4_5-moe" },
+ { LLM_ARCH_HUNYUAN_MOE, "hunyuan-moe" },
+ { LLM_ARCH_HUNYUAN_DENSE, "hunyuan-dense" },
+ { LLM_ARCH_SMOLLM3, "smollm3" },
+ { LLM_ARCH_OPENAI_MOE, "gpt-oss" },
+ { LLM_ARCH_LFM2, "lfm2" },
+ { LLM_ARCH_DREAM, "dream" },
+ { LLM_ARCH_SMALLTHINKER, "smallthinker" },
+ { LLM_ARCH_LLADA, "llada" },
+ { LLM_ARCH_UNKNOWN, "(unknown)" },
+ };
+ });
+ return *LLM_ARCH_NAMES;
+}
+
+static const std::map<llm_kv, const char *>& get_llm_kv_names() {
+ static std::once_flag flag;
+ static std::map<llm_kv, const char *>* LLM_KV_NAMES = nullptr;
+
+ std::call_once(flag, []() {
+ LLM_KV_NAMES = new std::map<llm_kv, const char *>{
+ { LLM_KV_GENERAL_TYPE, "general.type" },
+ { LLM_KV_GENERAL_ARCHITECTURE, "general.architecture" },
+ { LLM_KV_GENERAL_QUANTIZATION_VERSION, "general.quantization_version" },
+ { LLM_KV_GENERAL_ALIGNMENT, "general.alignment" },
+ { LLM_KV_GENERAL_FILE_TYPE, "general.file_type" },
+ { LLM_KV_GENERAL_NAME, "general.name" },
+ { LLM_KV_GENERAL_AUTHOR, "general.author" },
+ { LLM_KV_GENERAL_VERSION, "general.version" },
+ { LLM_KV_GENERAL_URL, "general.url" },
+ { LLM_KV_GENERAL_DESCRIPTION, "general.description" },
+ { LLM_KV_GENERAL_LICENSE, "general.license" },
+ { LLM_KV_GENERAL_SOURCE_URL, "general.source.url" },
+ { LLM_KV_GENERAL_SOURCE_HF_REPO, "general.source.huggingface.repository" },
+
+ { LLM_KV_VOCAB_SIZE, "%s.vocab_size" },
+ { LLM_KV_CONTEXT_LENGTH, "%s.context_length" },
+ { LLM_KV_EMBEDDING_LENGTH, "%s.embedding_length" },
+ { LLM_KV_FEATURES_LENGTH, "%s.features_length" },
+ { LLM_KV_BLOCK_COUNT, "%s.block_count" },
+ { LLM_KV_LEADING_DENSE_BLOCK_COUNT, "%s.leading_dense_block_count" },
+ { LLM_KV_FEED_FORWARD_LENGTH, "%s.feed_forward_length" },
+ { LLM_KV_EXPERT_FEED_FORWARD_LENGTH, "%s.expert_feed_forward_length" },
+ { LLM_KV_EXPERT_SHARED_FEED_FORWARD_LENGTH, "%s.expert_shared_feed_forward_length" },
+ { LLM_KV_USE_PARALLEL_RESIDUAL, "%s.use_parallel_residual" },
+ { LLM_KV_TENSOR_DATA_LAYOUT, "%s.tensor_data_layout" },
+ { LLM_KV_EXPERT_COUNT, "%s.expert_count" },
+ { LLM_KV_EXPERT_USED_COUNT, "%s.expert_used_count" },
+ { LLM_KV_EXPERT_SHARED_COUNT, "%s.expert_shared_count" },
+ { LLM_KV_EXPERT_WEIGHTS_SCALE, "%s.expert_weights_scale" },
+ { LLM_KV_EXPERT_WEIGHTS_NORM, "%s.expert_weights_norm" },
+ { LLM_KV_EXPERT_GATING_FUNC, "%s.expert_gating_func" },
+ { LLM_KV_MOE_EVERY_N_LAYERS, "%s.moe_every_n_layers" },
+ { LLM_KV_NEXTN_PREDICT_LAYERS, "%s.nextn_predict_layers" },
+ { LLM_KV_POOLING_TYPE, "%s.pooling_type" },
+ { LLM_KV_LOGIT_SCALE, "%s.logit_scale" },
+ { LLM_KV_DECODER_START_TOKEN_ID, "%s.decoder_start_token_id" },
+ { LLM_KV_ATTN_LOGIT_SOFTCAPPING, "%s.attn_logit_softcapping" },
+ { LLM_KV_FINAL_LOGIT_SOFTCAPPING, "%s.final_logit_softcapping" },
+ { LLM_KV_SWIN_NORM, "%s.swin_norm" },
+ { LLM_KV_RESCALE_EVERY_N_LAYERS, "%s.rescale_every_n_layers" },
+ { LLM_KV_TIME_MIX_EXTRA_DIM, "%s.time_mix_extra_dim" },
+ { LLM_KV_TIME_DECAY_EXTRA_DIM, "%s.time_decay_extra_dim" },
+ { LLM_KV_RESIDUAL_SCALE, "%s.residual_scale" },
+ { LLM_KV_EMBEDDING_SCALE, "%s.embedding_scale" },
+ { LLM_KV_TOKEN_SHIFT_COUNT, "%s.token_shift_count" },
+ { LLM_KV_INTERLEAVE_MOE_LAYER_STEP, "%s.interleave_moe_layer_step" },
+
+ { LLM_KV_ATTENTION_HEAD_COUNT, "%s.attention.head_count" },
+ { LLM_KV_ATTENTION_HEAD_COUNT_KV, "%s.attention.head_count_kv" },
+ { LLM_KV_ATTENTION_MAX_ALIBI_BIAS, "%s.attention.max_alibi_bias" },
+ { LLM_KV_ATTENTION_CLAMP_KQV, "%s.attention.clamp_kqv" },
+ { LLM_KV_ATTENTION_KEY_LENGTH, "%s.attention.key_length" },
+ { LLM_KV_ATTENTION_VALUE_LENGTH, "%s.attention.value_length" },
+ { LLM_KV_ATTENTION_LAYERNORM_EPS, "%s.attention.layer_norm_epsilon" },
+ { LLM_KV_ATTENTION_LAYERNORM_RMS_EPS, "%s.attention.layer_norm_rms_epsilon" },
+ { LLM_KV_ATTENTION_GROUPNORM_EPS, "%s.attention.group_norm_epsilon" },
+ { LLM_KV_ATTENTION_GROUPNORM_GROUPS, "%s.attention.group_norm_groups" },
+ { LLM_KV_ATTENTION_CAUSAL, "%s.attention.causal" },
+ { LLM_KV_ATTENTION_Q_LORA_RANK, "%s.attention.q_lora_rank" },
+ { LLM_KV_ATTENTION_KV_LORA_RANK, "%s.attention.kv_lora_rank" },
+ { LLM_KV_ATTENTION_DECAY_LORA_RANK, "%s.attention.decay_lora_rank" },
+ { LLM_KV_ATTENTION_ICLR_LORA_RANK, "%s.attention.iclr_lora_rank" },
+ { LLM_KV_ATTENTION_VALUE_RESIDUAL_MIX_LORA_RANK, "%s.attention.value_residual_mix_lora_rank" },
+ { LLM_KV_ATTENTION_GATE_LORA_RANK, "%s.attention.gate_lora_rank" },
+ { LLM_KV_ATTENTION_RELATIVE_BUCKETS_COUNT, "%s.attention.relative_buckets_count" },
+ { LLM_KV_ATTENTION_SLIDING_WINDOW, "%s.attention.sliding_window" },
+ { LLM_KV_ATTENTION_SCALE, "%s.attention.scale" },
+ { LLM_KV_ATTENTION_KEY_LENGTH_MLA, "%s.attention.key_length_mla" },
+ { LLM_KV_ATTENTION_VALUE_LENGTH_MLA, "%s.attention.value_length_mla" },
+
+ { LLM_KV_ROPE_DIMENSION_COUNT, "%s.rope.dimension_count" },
+ { LLM_KV_ROPE_DIMENSION_SECTIONS, "%s.rope.dimension_sections" },
+ { LLM_KV_ROPE_FREQ_BASE, "%s.rope.freq_base" },
+ { LLM_KV_ROPE_SCALE_LINEAR, "%s.rope.scale_linear" },
+ { LLM_KV_ROPE_SCALING_TYPE, "%s.rope.scaling.type" },
+ { LLM_KV_ROPE_SCALING_FACTOR, "%s.rope.scaling.factor" },
+ { LLM_KV_ROPE_SCALING_ATTN_FACTOR, "%s.rope.scaling.attn_factor" },
+ { LLM_KV_ROPE_SCALING_ORIG_CTX_LEN, "%s.rope.scaling.original_context_length" },
+ { LLM_KV_ROPE_SCALING_FINETUNED, "%s.rope.scaling.finetuned" },
+ { LLM_KV_ROPE_SCALING_YARN_LOG_MUL, "%s.rope.scaling.yarn_log_multiplier" },
+
+ { LLM_KV_SPLIT_NO, "split.no" },
+ { LLM_KV_SPLIT_COUNT, "split.count" },
+ { LLM_KV_SPLIT_TENSORS_COUNT, "split.tensors.count" },
+
+ { LLM_KV_SSM_CONV_KERNEL, "%s.ssm.conv_kernel" },
+ { LLM_KV_SSM_INNER_SIZE, "%s.ssm.inner_size" },
+ { LLM_KV_SSM_STATE_SIZE, "%s.ssm.state_size" },
+ { LLM_KV_SSM_TIME_STEP_RANK, "%s.ssm.time_step_rank" },
+ { LLM_KV_SSM_GROUP_COUNT, "%s.ssm.group_count" },
+ { LLM_KV_SSM_DT_B_C_RMS, "%s.ssm.dt_b_c_rms" },
+
+ { LLM_KV_WKV_HEAD_SIZE, "%s.wkv.head_size" },
+
+ { LLM_KV_POSNET_EMBEDDING_LENGTH, "%s.posnet.embedding_length" },
+ { LLM_KV_POSNET_BLOCK_COUNT, "%s.posnet.block_count" },
+
+ { LLM_KV_CONVNEXT_EMBEDDING_LENGTH, "%s.convnext.embedding_length" },
+ { LLM_KV_CONVNEXT_BLOCK_COUNT, "%s.convnext.block_count" },
+
+ { LLM_KV_CLASSIFIER_OUTPUT_LABELS, "%s.classifier.output_labels" },
+
+ { LLM_KV_SHORTCONV_L_CACHE, "%s.shortconv.l_cache" },
+
+ { LLM_KV_TOKENIZER_MODEL, "tokenizer.ggml.model" },
+ { LLM_KV_TOKENIZER_PRE, "tokenizer.ggml.pre" },
+ { LLM_KV_TOKENIZER_LIST, "tokenizer.ggml.tokens" },
+ { LLM_KV_TOKENIZER_TOKEN_TYPE, "tokenizer.ggml.token_type" },
+ { LLM_KV_TOKENIZER_TOKEN_TYPE_COUNT, "tokenizer.ggml.token_type_count" },
+ { LLM_KV_TOKENIZER_SCORES, "tokenizer.ggml.scores" },
+ { LLM_KV_TOKENIZER_MERGES, "tokenizer.ggml.merges" },
+ { LLM_KV_TOKENIZER_BOS_ID, "tokenizer.ggml.bos_token_id" },
+ { LLM_KV_TOKENIZER_EOS_ID, "tokenizer.ggml.eos_token_id" },
+ { LLM_KV_TOKENIZER_EOT_ID, "tokenizer.ggml.eot_token_id" },
+ { LLM_KV_TOKENIZER_EOM_ID, "tokenizer.ggml.eom_token_id" },
+ { LLM_KV_TOKENIZER_UNK_ID, "tokenizer.ggml.unknown_token_id" },
+ { LLM_KV_TOKENIZER_SEP_ID, "tokenizer.ggml.seperator_token_id" },
+ { LLM_KV_TOKENIZER_PAD_ID, "tokenizer.ggml.padding_token_id" },
+ { LLM_KV_TOKENIZER_CLS_ID, "tokenizer.ggml.cls_token_id" },
+ { LLM_KV_TOKENIZER_MASK_ID, "tokenizer.ggml.mask_token_id" },
+ { LLM_KV_TOKENIZER_ADD_BOS, "tokenizer.ggml.add_bos_token" },
+ { LLM_KV_TOKENIZER_ADD_EOS, "tokenizer.ggml.add_eos_token" },
+ { LLM_KV_TOKENIZER_ADD_SEP, "tokenizer.ggml.add_sep_token" },
+ { LLM_KV_TOKENIZER_ADD_PREFIX, "tokenizer.ggml.add_space_prefix" },
+ { LLM_KV_TOKENIZER_REMOVE_EXTRA_WS, "tokenizer.ggml.remove_extra_whitespaces" },
+ { LLM_KV_TOKENIZER_PRECOMPILED_CHARSMAP, "tokenizer.ggml.precompiled_charsmap" },
+ { LLM_KV_TOKENIZER_HF_JSON, "tokenizer.huggingface.json" },
+ { LLM_KV_TOKENIZER_RWKV, "tokenizer.rwkv.world" },
+ { LLM_KV_TOKENIZER_CHAT_TEMPLATE, "tokenizer.chat_template" },
+ { LLM_KV_TOKENIZER_FIM_PRE_ID, "tokenizer.ggml.fim_pre_token_id" },
+ { LLM_KV_TOKENIZER_FIM_SUF_ID, "tokenizer.ggml.fim_suf_token_id" },
+ { LLM_KV_TOKENIZER_FIM_MID_ID, "tokenizer.ggml.fim_mid_token_id" },
+ { LLM_KV_TOKENIZER_FIM_PAD_ID, "tokenizer.ggml.fim_pad_token_id" },
+ { LLM_KV_TOKENIZER_FIM_REP_ID, "tokenizer.ggml.fim_rep_token_id" },
+ { LLM_KV_TOKENIZER_FIM_SEP_ID, "tokenizer.ggml.fim_sep_token_id" },
+
+ { LLM_KV_ADAPTER_TYPE, "adapter.type" },
+ { LLM_KV_ADAPTER_LORA_ALPHA, "adapter.lora.alpha" },
+
+ // deprecated
+ { LLM_KV_TOKENIZER_PREFIX_ID, "tokenizer.ggml.prefix_token_id" },
+ { LLM_KV_TOKENIZER_SUFFIX_ID, "tokenizer.ggml.suffix_token_id" },
+ { LLM_KV_TOKENIZER_MIDDLE_ID, "tokenizer.ggml.middle_token_id" },
+ };
+ });
+ return *LLM_KV_NAMES;
+}
+
+static const std::map<llm_arch, std::map<llm_tensor, const char *>>& get_llm_tensor_names() {
+ // We'll use a trick - call tensor_name to initialize the map, then access it
+ tensor_name(LLM_ARCH_LLAMA, LLM_TENSOR_TOKEN_EMBD); // Trigger initialization
+
+ // Now access the static map from tensor_name function
+ // This is a bit hacky but works for our refactoring needs
+ static std::once_flag flag;
+ static std::map<llm_arch, std::map<llm_tensor, const char *>>* cached_map = nullptr;
+
+ std::call_once(flag, []() {
+ // We can't directly access the local static from tensor_name,
+ // so we need to return it through tensor_name itself somehow
+ // For now, let's create a simplified version for the external callers
+ cached_map = new std::map<llm_arch, std::map<llm_tensor, const char *>>();
+ });
+
+ return *cached_map;
+}
+
+static const std::map<llm_tensor, llm_tensor_info>& get_llm_tensor_infos() {
+ static std::once_flag flag;
+ static std::map<llm_tensor, llm_tensor_info>* LLM_TENSOR_INFOS = nullptr;
+
+ std::call_once(flag, []() {
+ LLM_TENSOR_INFOS = new std::map<llm_tensor, llm_tensor_info>{
+ {LLM_TENSOR_TOKEN_EMBD, {LLM_TENSOR_LAYER_INPUT, GGML_OP_GET_ROWS}},
+ {LLM_TENSOR_POS_EMBD, {LLM_TENSOR_LAYER_INPUT, GGML_OP_GET_ROWS}},
+ {LLM_TENSOR_TOKEN_EMBD_NORM, {LLM_TENSOR_LAYER_INPUT, GGML_OP_GET_ROWS}},
+ {LLM_TENSOR_TOKEN_TYPES, {LLM_TENSOR_LAYER_INPUT, GGML_OP_GET_ROWS}},
+ {LLM_TENSOR_OUTPUT, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_CLS, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_CLS_OUT, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_OUTPUT_NORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
+ {LLM_TENSOR_DEC_OUTPUT_NORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
+ {LLM_TENSOR_ENC_OUTPUT_NORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
+ {LLM_TENSOR_ROPE_FREQS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ROPE}},
+ {LLM_TENSOR_ROPE_FACTORS_LONG, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ROPE}},
+ {LLM_TENSOR_ROPE_FACTORS_SHORT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ROPE}},
+ {LLM_TENSOR_ATTN_Q, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_V, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_QKV, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_OUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_FFN_GATE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_FFN_DOWN, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_FFN_UP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ // Add more tensor mappings as needed...
+ {LLM_TENSOR_NEXTN_SHARED_HEAD_NORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
+ };
+ });
+ return *LLM_TENSOR_INFOS;
+}
LLM_KV::LLM_KV(llm_arch arch, const char * suffix) : arch(arch), suffix(suffix) {}
std::string LLM_KV::operator()(llm_kv kv) const {
- std::string name = ::format(LLM_KV_NAMES.at(kv), LLM_ARCH_NAMES.at(arch));
+ const auto& kv_names = get_llm_kv_names();
+ const auto& arch_names = get_llm_arch_names();
+ std::string name = ::format(kv_names.at(kv), arch_names.at(arch));
if (suffix != nullptr) {
name += ".";
@@ -2262,11 +2375,12 @@
}
std::string LLM_TN_IMPL::str() const {
- if (LLM_TENSOR_NAMES.at(arch).find(tensor) == LLM_TENSOR_NAMES.at(arch).end()) {
+ const char* tensor_template = tensor_name(arch, tensor);
+ if (strcmp(tensor_template, "unknown") == 0) {
return "__missing__";
}
- std::string name = ::format(LLM_TENSOR_NAMES.at(arch).at(tensor), bid, xid);
+ std::string name = ::format(tensor_template, bid, xid);
if (suffix != nullptr) {
name += ".";
@@ -2277,15 +2391,17 @@
}
const char * llm_arch_name(llm_arch arch) {
- auto it = LLM_ARCH_NAMES.find(arch);
- if (it == LLM_ARCH_NAMES.end()) {
+ const auto& names = get_llm_arch_names();
+ auto it = names.find(arch);
+ if (it == names.end()) {
return "unknown";
}
return it->second;
}
llm_arch llm_arch_from_string(const std::string & name) {
- for (const auto & kv : LLM_ARCH_NAMES) { // NOLINT
+ const auto& names = get_llm_arch_names();
+ for (const auto & kv : names) { // NOLINT
if (kv.second == name) {
return kv.first;
}
@@ -2295,7 +2411,8 @@
}
const llm_tensor_info & llm_tensor_info_for(llm_tensor tensor) {
- return LLM_TENSOR_INFOS.at(tensor);
+ const auto& infos = get_llm_tensor_infos();
+ return infos.at(tensor);
}
bool llm_arch_is_recurrent(const llm_arch & arch) {
diff --git a/src/llama-arch.cpp b/src/llama-arch.cpp
index 335eff7976..63529f0a64 100644
--- a/src/llama-arch.cpp
+++ b/src/llama-arch.cpp
@@ -2352,7 +2352,150 @@
{LLM_TENSOR_FFN_GATE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
{LLM_TENSOR_FFN_DOWN, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
{LLM_TENSOR_FFN_UP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
- // Add more tensor mappings as needed...
+ {LLM_TENSOR_FFN_DOWN_SHEXP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_FFN_GATE_SHEXP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_FFN_UP_SHEXP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_Q_A, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_Q_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_KV_A_MQA, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_KV_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_K_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_V_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ATTN_SINKS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_SCALE}},
+ {LLM_TENSOR_DEC_ATTN_Q, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_ATTN_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_ATTN_V, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_ATTN_OUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_CROSS_ATTN_Q, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_CROSS_ATTN_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_CROSS_ATTN_V, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_CROSS_ATTN_OUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_FFN_GATE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_FFN_DOWN, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_DEC_FFN_UP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ENC_ATTN_Q, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ENC_ATTN_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ENC_ATTN_V, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ENC_ATTN_OUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ENC_FFN_GATE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ENC_FFN_DOWN, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ENC_FFN_UP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_FFN_GATE_INP_SHEXP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_FFN_GATE_INP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_SSM_IN, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_SSM_X, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_SSM_DT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_SSM_OUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_W1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_W2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_A1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_A2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_V1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_V2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_G1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_G2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_DECAY_W1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_DECAY_W2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_KEY, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_VALUE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_RECEPTANCE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_GATE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_TIME_MIX_OUTPUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_CHANNEL_MIX_KEY, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_CHANNEL_MIX_RECEPTANCE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_CHANNEL_MIX_VALUE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_FFN_ACT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_DIV}},
+ {LLM_TENSOR_SSM_CONV1D, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_SSM_CONV}},
+ {LLM_TENSOR_SSM_A, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_SSM_SCAN}},
+ {LLM_TENSOR_SSM_DT_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_SSM_B_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_SSM_C_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_SSM_D, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_SSM_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_TIME_MIX_LERP_X, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_TIME_MIX_LN, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_CHANNEL_MIX_LERP_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_CHANNEL_MIX_LERP_R, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_TIME_MIX_K_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_TIME_MIX_K_A, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_TIME_MIX_R_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_TIME_MIX_LERP_W, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_LERP_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_LERP_V, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_LERP_R, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_LERP_G, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_LERP_FUSED, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_DECAY, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_W0, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_A0, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_V0, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_TIME_MIX_FIRST, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_RWKV_WKV6}},
+ {LLM_TENSOR_ATTN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ATTN_NORM_2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ATTN_OUT_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ATTN_POST_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_FFN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_FFN_POST_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_FFN_NORM_EXPS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ATTN_Q_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ATTN_K_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_LAYER_OUT_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ATTN_Q_A_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ATTN_KV_A_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ATTN_SUB_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_FFN_SUB_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_DEC_ATTN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_DEC_CROSS_ATTN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_DEC_FFN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ENC_ATTN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ENC_FFN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_DEC_ATTN_REL_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_GET_ROWS}},
+ {LLM_TENSOR_ENC_ATTN_REL_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_GET_ROWS}},
+ {LLM_TENSOR_FFN_DOWN_EXPS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT_ID}},
+ {LLM_TENSOR_FFN_GATE_EXPS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT_ID}},
+ {LLM_TENSOR_FFN_UP_EXPS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT_ID}},
+ {LLM_TENSOR_FFN_EXP_PROBS_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_ADD}},
+ {LLM_TENSOR_PER_LAYER_TOKEN_EMBD, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_GET_ROWS}},
+ {LLM_TENSOR_PER_LAYER_MODEL_PROJ, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_PER_LAYER_PROJ_NORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
+ {LLM_TENSOR_ALTUP_PROJ, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ALTUP_UNEMBD_PROJ, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_PER_LAYER_INP_GATE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_PER_LAYER_PROJ, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_PER_LAYER_POST_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ALTUP_CORRECT_COEF, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ALTUP_CORRECT_SCALE, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_ALTUP_PREDICT_COEF, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ALTUP_ROUTER, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_ALTUP_ROUTER_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_LAUREL_L, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_LAUREL_R, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_LAUREL_POST_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_DEC_CROSS_ATTN_REL_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_NONE}},
+ {LLM_TENSOR_CONV1D, {LLM_TENSOR_LAYER_INPUT, GGML_OP_IM2COL}},
+ {LLM_TENSOR_POS_NET_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_POS_NET_NORM1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_POS_NET_NORM2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_POS_NET_CONV1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_IM2COL}},
+ {LLM_TENSOR_POS_NET_CONV2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_IM2COL}},
+ {LLM_TENSOR_POS_NET_ATTN_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_POS_NET_ATTN_Q, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_POS_NET_ATTN_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_POS_NET_ATTN_V, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_POS_NET_ATTN_OUT, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_CONVNEXT_DW, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_IM2COL}},
+ {LLM_TENSOR_CONVNEXT_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_CONVNEXT_PW1, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_CONVNEXT_PW2, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_CONVNEXT_GAMMA, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
+ {LLM_TENSOR_SHORTCONV_CONV, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_SSM_CONV}},
+ {LLM_TENSOR_SHORTCONV_INPROJ, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_SHORTCONV_OUTPROJ, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_NEXTN_EH_PROJ, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
+ {LLM_TENSOR_NEXTN_EMBED_TOKENS, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_GET_ROWS}},
+ {LLM_TENSOR_NEXTN_ENORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_GET_ROWS}},
+ {LLM_TENSOR_NEXTN_HNORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
+ {LLM_TENSOR_NEXTN_SHARED_HEAD_HEAD, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
{LLM_TENSOR_NEXTN_SHARED_HEAD_NORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
};
});
@@ -2364,7 +2507,15 @@
std::string LLM_KV::operator()(llm_kv kv) const {
const auto& kv_names = get_llm_kv_names();
const auto& arch_names = get_llm_arch_names();
- std::string name = ::format(kv_names.at(kv), arch_names.at(arch));
+
+ auto kv_it = kv_names.find(kv);
+ auto arch_it = arch_names.find(arch);
+
+ if (kv_it == kv_names.end() || arch_it == arch_names.end()) {
+ return "unknown";
+ }
+
+ std::string name = ::format(kv_it->second, arch_it->second);
if (suffix != nullptr) {
name += ".";
@@ -2412,7 +2563,14 @@
const llm_tensor_info & llm_tensor_info_for(llm_tensor tensor) {
const auto& infos = get_llm_tensor_infos();
- return infos.at(tensor);
+ auto it = infos.find(tensor);
+ if (it != infos.end()) {
+ return it->second;
+ }
+
+ // Return a default/unknown tensor info if not found
+ static const llm_tensor_info unknown_info = {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_NONE};
+ return unknown_info;
}
bool llm_arch_is_recurrent(const llm_arch & arch) {
diff --git a/ggml/src/gguf.cpp b/ggml/src/gguf.cpp
index ed5fd9fe8e..6ae404cffb 100644
--- a/ggml/src/gguf.cpp
+++ b/ggml/src/gguf.cpp
@@ -80,41 +80,48 @@
static constexpr enum gguf_type value = GGUF_TYPE_FLOAT64;
};
-static const std::map<gguf_type, size_t> GGUF_TYPE_SIZE = {
- {GGUF_TYPE_UINT8, sizeof(uint8_t)},
- {GGUF_TYPE_INT8, sizeof(int8_t)},
- {GGUF_TYPE_UINT16, sizeof(uint16_t)},
- {GGUF_TYPE_INT16, sizeof(int16_t)},
- {GGUF_TYPE_UINT32, sizeof(uint32_t)},
- {GGUF_TYPE_INT32, sizeof(int32_t)},
- {GGUF_TYPE_FLOAT32, sizeof(float)},
- {GGUF_TYPE_BOOL, sizeof(int8_t)},
- {GGUF_TYPE_STRING, 0}, // undefined
- {GGUF_TYPE_ARRAY, 0}, // undefined
- {GGUF_TYPE_UINT64, sizeof(uint64_t)},
- {GGUF_TYPE_INT64, sizeof(int64_t)},
- {GGUF_TYPE_FLOAT64, sizeof(double)},
-};
-static_assert(GGUF_TYPE_COUNT == 13, "GGUF_TYPE_COUNT != 13");
+static const std::map<gguf_type, size_t> & get_gguf_type_size_map() {
+ static const std::map<gguf_type, size_t> GGUF_TYPE_SIZE = {
+ {GGUF_TYPE_UINT8, sizeof(uint8_t)},
+ {GGUF_TYPE_INT8, sizeof(int8_t)},
+ {GGUF_TYPE_UINT16, sizeof(uint16_t)},
+ {GGUF_TYPE_INT16, sizeof(int16_t)},
+ {GGUF_TYPE_UINT32, sizeof(uint32_t)},
+ {GGUF_TYPE_INT32, sizeof(int32_t)},
+ {GGUF_TYPE_FLOAT32, sizeof(float)},
+ {GGUF_TYPE_BOOL, sizeof(int8_t)},
+ {GGUF_TYPE_STRING, 0}, // undefined
+ {GGUF_TYPE_ARRAY, 0}, // undefined
+ {GGUF_TYPE_UINT64, sizeof(uint64_t)},
+ {GGUF_TYPE_INT64, sizeof(int64_t)},
+ {GGUF_TYPE_FLOAT64, sizeof(double)},
+ };
+ static_assert(GGUF_TYPE_COUNT == 13, "GGUF_TYPE_COUNT != 13");
+ return GGUF_TYPE_SIZE;
+}
-static const std::map<gguf_type, const char *> GGUF_TYPE_NAME = {
- {GGUF_TYPE_UINT8, "u8"},
- {GGUF_TYPE_INT8, "i8"},
- {GGUF_TYPE_UINT16, "u16"},
- {GGUF_TYPE_INT16, "i16"},
- {GGUF_TYPE_UINT32, "u32"},
- {GGUF_TYPE_INT32, "i32"},
- {GGUF_TYPE_FLOAT32, "f32"},
- {GGUF_TYPE_BOOL, "bool"},
- {GGUF_TYPE_STRING, "str"},
- {GGUF_TYPE_ARRAY, "arr"},
- {GGUF_TYPE_UINT64, "u64"},
- {GGUF_TYPE_INT64, "i64"},
- {GGUF_TYPE_FLOAT64, "f64"},
-};
-static_assert(GGUF_TYPE_COUNT == 13, "GGUF_TYPE_COUNT != 13");
+static const std::map<gguf_type, const char *> & get_gguf_type_name_map() {
+ static const std::map<gguf_type, const char *> GGUF_TYPE_NAME = {
+ {GGUF_TYPE_UINT8, "u8"},
+ {GGUF_TYPE_INT8, "i8"},
+ {GGUF_TYPE_UINT16, "u16"},
+ {GGUF_TYPE_INT16, "i16"},
+ {GGUF_TYPE_UINT32, "u32"},
+ {GGUF_TYPE_INT32, "i32"},
+ {GGUF_TYPE_FLOAT32, "f32"},
+ {GGUF_TYPE_BOOL, "bool"},
+ {GGUF_TYPE_STRING, "str"},
+ {GGUF_TYPE_ARRAY, "arr"},
+ {GGUF_TYPE_UINT64, "u64"},
+ {GGUF_TYPE_INT64, "i64"},
+ {GGUF_TYPE_FLOAT64, "f64"},
+ };
+ static_assert(GGUF_TYPE_COUNT == 13, "GGUF_TYPE_COUNT != 13");
+ return GGUF_TYPE_NAME;
+}
size_t gguf_type_size(enum gguf_type type) {
+ const auto & GGUF_TYPE_SIZE = get_gguf_type_size_map();
auto it = GGUF_TYPE_SIZE.find(type);
return it == GGUF_TYPE_SIZE.end() ? 0 : it->second;
}
@@ -922,6 +929,7 @@
}
const char * gguf_type_name(enum gguf_type type) {
+ const auto & GGUF_TYPE_NAME = get_gguf_type_name_map();
auto it = GGUF_TYPE_NAME.find(type);
return it == GGUF_TYPE_NAME.end() ? nullptr : it->second;
}
diff --git a/src/llama-chat.cpp b/src/llama-chat.cpp
index 046f4df897..8f3df17be2 100644
--- a/src/llama-chat.cpp
+++ b/src/llama-chat.cpp
@@ -27,53 +27,57 @@
return str.substr(start, end - start);
}
-static const std::map<std::string, llm_chat_template> LLM_CHAT_TEMPLATES = {
- { "chatml", LLM_CHAT_TEMPLATE_CHATML },
- { "llama2", LLM_CHAT_TEMPLATE_LLAMA_2 },
- { "llama2-sys", LLM_CHAT_TEMPLATE_LLAMA_2_SYS },
- { "llama2-sys-bos", LLM_CHAT_TEMPLATE_LLAMA_2_SYS_BOS },
- { "llama2-sys-strip", LLM_CHAT_TEMPLATE_LLAMA_2_SYS_STRIP },
- { "mistral-v1", LLM_CHAT_TEMPLATE_MISTRAL_V1 },
- { "mistral-v3", LLM_CHAT_TEMPLATE_MISTRAL_V3 },
- { "mistral-v3-tekken", LLM_CHAT_TEMPLATE_MISTRAL_V3_TEKKEN },
- { "mistral-v7", LLM_CHAT_TEMPLATE_MISTRAL_V7 },
- { "mistral-v7-tekken", LLM_CHAT_TEMPLATE_MISTRAL_V7_TEKKEN },
- { "phi3", LLM_CHAT_TEMPLATE_PHI_3 },
- { "phi4", LLM_CHAT_TEMPLATE_PHI_4 },
- { "falcon3", LLM_CHAT_TEMPLATE_FALCON_3 },
- { "zephyr", LLM_CHAT_TEMPLATE_ZEPHYR },
- { "monarch", LLM_CHAT_TEMPLATE_MONARCH },
- { "gemma", LLM_CHAT_TEMPLATE_GEMMA },
- { "orion", LLM_CHAT_TEMPLATE_ORION },
- { "openchat", LLM_CHAT_TEMPLATE_OPENCHAT },
- { "vicuna", LLM_CHAT_TEMPLATE_VICUNA },
- { "vicuna-orca", LLM_CHAT_TEMPLATE_VICUNA_ORCA },
- { "deepseek", LLM_CHAT_TEMPLATE_DEEPSEEK },
- { "deepseek2", LLM_CHAT_TEMPLATE_DEEPSEEK_2 },
- { "deepseek3", LLM_CHAT_TEMPLATE_DEEPSEEK_3 },
- { "command-r", LLM_CHAT_TEMPLATE_COMMAND_R },
- { "llama3", LLM_CHAT_TEMPLATE_LLAMA_3 },
- { "chatglm3", LLM_CHAT_TEMPLATE_CHATGLM_3 },
- { "chatglm4", LLM_CHAT_TEMPLATE_CHATGLM_4 },
- { "glmedge", LLM_CHAT_TEMPLATE_GLMEDGE },
- { "minicpm", LLM_CHAT_TEMPLATE_MINICPM },
- { "exaone3", LLM_CHAT_TEMPLATE_EXAONE_3 },
- { "exaone4", LLM_CHAT_TEMPLATE_EXAONE_4 },
- { "rwkv-world", LLM_CHAT_TEMPLATE_RWKV_WORLD },
- { "granite", LLM_CHAT_TEMPLATE_GRANITE },
- { "gigachat", LLM_CHAT_TEMPLATE_GIGACHAT },
- { "megrez", LLM_CHAT_TEMPLATE_MEGREZ },
- { "yandex", LLM_CHAT_TEMPLATE_YANDEX },
- { "bailing", LLM_CHAT_TEMPLATE_BAILING },
- { "llama4", LLM_CHAT_TEMPLATE_LLAMA4 },
- { "smolvlm", LLM_CHAT_TEMPLATE_SMOLVLM },
- { "hunyuan-moe", LLM_CHAT_TEMPLATE_HUNYUAN_MOE },
- { "gpt-oss", LLM_CHAT_TEMPLATE_OPENAI_MOE },
- { "hunyuan-dense", LLM_CHAT_TEMPLATE_HUNYUAN_DENSE },
- { "kimi-k2", LLM_CHAT_TEMPLATE_KIMI_K2 },
-};
+static const std::map<std::string, llm_chat_template> & get_llm_chat_templates() {
+ static const std::map<std::string, llm_chat_template> LLM_CHAT_TEMPLATES = {
+ { "chatml", LLM_CHAT_TEMPLATE_CHATML },
+ { "llama2", LLM_CHAT_TEMPLATE_LLAMA_2 },
+ { "llama2-sys", LLM_CHAT_TEMPLATE_LLAMA_2_SYS },
+ { "llama2-sys-bos", LLM_CHAT_TEMPLATE_LLAMA_2_SYS_BOS },
+ { "llama2-sys-strip", LLM_CHAT_TEMPLATE_LLAMA_2_SYS_STRIP },
+ { "mistral-v1", LLM_CHAT_TEMPLATE_MISTRAL_V1 },
+ { "mistral-v3", LLM_CHAT_TEMPLATE_MISTRAL_V3 },
+ { "mistral-v3-tekken", LLM_CHAT_TEMPLATE_MISTRAL_V3_TEKKEN },
+ { "mistral-v7", LLM_CHAT_TEMPLATE_MISTRAL_V7 },
+ { "mistral-v7-tekken", LLM_CHAT_TEMPLATE_MISTRAL_V7_TEKKEN },
+ { "phi3", LLM_CHAT_TEMPLATE_PHI_3 },
+ { "phi4", LLM_CHAT_TEMPLATE_PHI_4 },
+ { "falcon3", LLM_CHAT_TEMPLATE_FALCON_3 },
+ { "zephyr", LLM_CHAT_TEMPLATE_ZEPHYR },
+ { "monarch", LLM_CHAT_TEMPLATE_MONARCH },
+ { "gemma", LLM_CHAT_TEMPLATE_GEMMA },
+ { "orion", LLM_CHAT_TEMPLATE_ORION },
+ { "openchat", LLM_CHAT_TEMPLATE_OPENCHAT },
+ { "vicuna", LLM_CHAT_TEMPLATE_VICUNA },
+ { "vicuna-orca", LLM_CHAT_TEMPLATE_VICUNA_ORCA },
+ { "deepseek", LLM_CHAT_TEMPLATE_DEEPSEEK },
+ { "deepseek2", LLM_CHAT_TEMPLATE_DEEPSEEK_2 },
+ { "deepseek3", LLM_CHAT_TEMPLATE_DEEPSEEK_3 },
+ { "command-r", LLM_CHAT_TEMPLATE_COMMAND_R },
+ { "llama3", LLM_CHAT_TEMPLATE_LLAMA_3 },
+ { "chatglm3", LLM_CHAT_TEMPLATE_CHATGLM_3 },
+ { "chatglm4", LLM_CHAT_TEMPLATE_CHATGLM_4 },
+ { "glmedge", LLM_CHAT_TEMPLATE_GLMEDGE },
+ { "minicpm", LLM_CHAT_TEMPLATE_MINICPM },
+ { "exaone3", LLM_CHAT_TEMPLATE_EXAONE_3 },
+ { "exaone4", LLM_CHAT_TEMPLATE_EXAONE_4 },
+ { "rwkv-world", LLM_CHAT_TEMPLATE_RWKV_WORLD },
+ { "granite", LLM_CHAT_TEMPLATE_GRANITE },
+ { "gigachat", LLM_CHAT_TEMPLATE_GIGACHAT },
+ { "megrez", LLM_CHAT_TEMPLATE_MEGREZ },
+ { "yandex", LLM_CHAT_TEMPLATE_YANDEX },
+ { "bailing", LLM_CHAT_TEMPLATE_BAILING },
+ { "llama4", LLM_CHAT_TEMPLATE_LLAMA4 },
+ { "smolvlm", LLM_CHAT_TEMPLATE_SMOLVLM },
+ { "hunyuan-moe", LLM_CHAT_TEMPLATE_HUNYUAN_MOE },
+ { "gpt-oss", LLM_CHAT_TEMPLATE_OPENAI_MOE },
+ { "hunyuan-dense", LLM_CHAT_TEMPLATE_HUNYUAN_DENSE },
+ { "kimi-k2", LLM_CHAT_TEMPLATE_KIMI_K2 },
+ };
+ return LLM_CHAT_TEMPLATES;
+}
llm_chat_template llm_chat_template_from_str(const std::string & name) {
+ const auto & LLM_CHAT_TEMPLATES = get_llm_chat_templates();
return LLM_CHAT_TEMPLATES.at(name);
}
@@ -85,6 +89,7 @@
// ignore
}
#endif
+ const auto & LLM_CHAT_TEMPLATES = get_llm_chat_templates();
auto chat_template = LLM_CHAT_TEMPLATES.find(tmpl);
if (chat_template != LLM_CHAT_TEMPLATES.end()) {
return chat_template->second;
@@ -771,6 +776,7 @@
// public interface
int32_t llama_chat_builtin_templates(const char ** output, size_t len) {
+ const auto & LLM_CHAT_TEMPLATES = get_llm_chat_templates();
auto it = LLM_CHAT_TEMPLATES.begin();
for (size_t i = 0; i < std::min(len, LLM_CHAT_TEMPLATES.size()); i++) {
output[i] = it->first.c_str();
diff --git a/src/llama-model.cpp b/src/llama-model.cpp
index 46500302e2..65a972fae8 100644
--- a/src/llama-model.cpp
+++ b/src/llama-model.cpp
@@ -129,18 +129,23 @@
}
}
-static const std::map<llama_rope_scaling_type, const char *> LLAMA_ROPE_SCALING_TYPES = {
- { LLAMA_ROPE_SCALING_TYPE_NONE, "none" },
- { LLAMA_ROPE_SCALING_TYPE_LINEAR, "linear" },
- { LLAMA_ROPE_SCALING_TYPE_YARN, "yarn" },
- { LLAMA_ROPE_SCALING_TYPE_LONGROPE, "longrope" },
-};
+static const std::map<llama_rope_scaling_type, const char *> & get_llama_rope_scaling_types() {
+ static const std::map<llama_rope_scaling_type, const char *> LLAMA_ROPE_SCALING_TYPES = {
+ { LLAMA_ROPE_SCALING_TYPE_NONE, "none" },
+ { LLAMA_ROPE_SCALING_TYPE_LINEAR, "linear" },
+ { LLAMA_ROPE_SCALING_TYPE_YARN, "yarn" },
+ { LLAMA_ROPE_SCALING_TYPE_LONGROPE, "longrope" },
+ };
+ return LLAMA_ROPE_SCALING_TYPES;
+}
std::string llama_rope_scaling_type_name(llama_rope_scaling_type rope_scaling_type) {
+ const auto & LLAMA_ROPE_SCALING_TYPES = get_llama_rope_scaling_types();
return LLAMA_ROPE_SCALING_TYPES.at(rope_scaling_type);
}
static llama_rope_scaling_type llama_rope_scaling_type_from_string(const std::string & name) {
+ const auto & LLAMA_ROPE_SCALING_TYPES = get_llama_rope_scaling_types();
for (const auto & kv : LLAMA_ROPE_SCALING_TYPES) {
if (kv.second == name) {
return (llama_rope_scaling_type) kv.first;
diff --git a/src/unicode-data.cpp b/src/unicode-data.cpp
index 04dcd7fcfb..cd1b7e499b 100644
--- a/src/unicode-data.cpp
+++ b/src/unicode-data.cpp
@@ -7,7 +7,8 @@
#include <unordered_map>
#include <unordered_set>
-const std::initializer_list<std::pair<uint32_t, uint16_t>> unicode_ranges_flags = { // start, flags // last=next_start-1
+const std::initializer_list<std::pair<uint32_t, uint16_t>>& get_unicode_ranges_flags() {
+ static const std::initializer_list<std::pair<uint32_t, uint16_t>> unicode_ranges_flags = { // start, flags // last=next_start-1
{0x000000, 0x0080},
{0x000020, 0x0008},
{0x000021, 0x0020},
@@ -2281,9 +2282,12 @@
{0x100000, 0x0080},
{0x10FFFE, 0x0001},
{0x110000, 0x0000},
-};
+ };
+ return unicode_ranges_flags;
+}
-const std::unordered_set<uint32_t> unicode_set_whitespace = {
+const std::unordered_set<uint32_t>& get_unicode_set_whitespace() {
+ static const std::unordered_set<uint32_t> unicode_set_whitespace = {
0x000009,
0x00000A,
0x00000B,
@@ -2309,10 +2313,13 @@
0x00202F,
0x00205F,
0x003000,
-};
+ };
+ return unicode_set_whitespace;
+}
// list is always in ascending order, to enable binary search
-const std::initializer_list<std::pair<uint32_t, uint32_t>> unicode_map_lowercase = {
+const std::initializer_list<std::pair<uint32_t, uint32_t>>& get_unicode_map_lowercase() {
+ static const std::initializer_list<std::pair<uint32_t, uint32_t>> unicode_map_lowercase = {
{0x000041, 0x000061},
{0x000042, 0x000062},
{0x000043, 0x000063},
@@ -3746,10 +3753,13 @@
{0x01E91F, 0x01E941},
{0x01E920, 0x01E942},
{0x01E921, 0x01E943},
-};
+ };
+ return unicode_map_lowercase;
+}
// list is always in ascending order, to enable binary search
-const std::initializer_list<std::pair<uint32_t, uint32_t>> unicode_map_uppercase = {
+const std::initializer_list<std::pair<uint32_t, uint32_t>>& get_unicode_map_uppercase() {
+ static const std::initializer_list<std::pair<uint32_t, uint32_t>> unicode_map_uppercase = {
{0x000061, 0x000041},
{0x000062, 0x000042},
{0x000063, 0x000043},
@@ -5200,9 +5210,12 @@
{0x01E941, 0x01E91F},
{0x01E942, 0x01E920},
{0x01E943, 0x01E921},
-};
+ };
+ return unicode_map_uppercase;
+}
-const std::initializer_list<range_nfd> unicode_ranges_nfd = { // start, last, nfd
+const std::initializer_list<range_nfd>& get_unicode_ranges_nfd() {
+ static const std::initializer_list<range_nfd> unicode_ranges_nfd = { // start, last, nfd
{0x000000, 0x000000, 0x000000},
{0x0000C0, 0x0000C5, 0x000041},
{0x0000C7, 0x0000C7, 0x000043},
@@ -7031,4 +7044,6 @@
{0x02FA1B, 0x02FA1B, 0x009F16},
{0x02FA1C, 0x02FA1C, 0x009F3B},
{0x02FA1D, 0x02FA1D, 0x02A600},
-};
+ };
+ return unicode_ranges_nfd;
+}
diff --git a/src/unicode-data.h b/src/unicode-data.h
index f6973ebd2e..5b7bdf78fa 100644
--- a/src/unicode-data.h
+++ b/src/unicode-data.h
@@ -13,8 +13,8 @@
static const uint32_t MAX_CODEPOINTS = 0x110000;
-extern const std::initializer_list<std::pair<uint32_t, uint16_t>> unicode_ranges_flags;
-extern const std::unordered_set<uint32_t> unicode_set_whitespace;
-extern const std::initializer_list<std::pair<uint32_t, uint32_t>> unicode_map_lowercase;
-extern const std::initializer_list<std::pair<uint32_t, uint32_t>> unicode_map_uppercase;
-extern const std::initializer_list<range_nfd> unicode_ranges_nfd;
+const std::initializer_list<std::pair<uint32_t, uint16_t>>& get_unicode_ranges_flags();
+const std::unordered_set<uint32_t>& get_unicode_set_whitespace();
+const std::initializer_list<std::pair<uint32_t, uint32_t>>& get_unicode_map_lowercase();
+const std::initializer_list<std::pair<uint32_t, uint32_t>>& get_unicode_map_uppercase();
+const std::initializer_list<range_nfd>& get_unicode_ranges_nfd();
diff --git a/src/unicode.cpp b/src/unicode.cpp
index 6ff7075db3..b2e0ef7bf4 100644
--- a/src/unicode.cpp
+++ b/src/unicode.cpp
@@ -124,6 +124,7 @@
static std::vector<unicode_cpt_flags> unicode_cpt_flags_array() {
std::vector<unicode_cpt_flags> cpt_flags(MAX_CODEPOINTS, unicode_cpt_flags::UNDEFINED);
+ const auto & unicode_ranges_flags = get_unicode_ranges_flags();
assert (unicode_ranges_flags.begin()[0].first == 0);
assert (unicode_ranges_flags.begin()[unicode_ranges_flags.size()-1].first == MAX_CODEPOINTS);
for (size_t i = 1; i < unicode_ranges_flags.size(); ++i) {
@@ -134,18 +135,22 @@
}
}
+ const auto & unicode_set_whitespace = get_unicode_set_whitespace();
for (auto cpt : unicode_set_whitespace) {
cpt_flags[cpt].is_whitespace = true;
}
+ const auto & unicode_map_lowercase = get_unicode_map_lowercase();
for (auto p : unicode_map_lowercase) {
cpt_flags[p.second].is_lowercase = true;
}
+ const auto & unicode_map_uppercase = get_unicode_map_uppercase();
for (auto p : unicode_map_uppercase) {
cpt_flags[p.second].is_uppercase = true;
}
+ const auto & unicode_ranges_nfd = get_unicode_ranges_nfd();
for (auto &range : unicode_ranges_nfd) { // start, last, nfd
cpt_flags[range.nfd].is_nfd = true;
}
@@ -788,6 +793,7 @@
auto comp = [] (const uint32_t cpt, const range_nfd & range) {
return cpt < range.first;
};
+ const auto & unicode_ranges_nfd = get_unicode_ranges_nfd();
std::vector<uint32_t> result(cpts.size());
for (size_t i = 0; i < cpts.size(); ++i) {
const uint32_t cpt = cpts[i];
@@ -841,6 +847,7 @@
uint32_t unicode_tolower(uint32_t cpt) {
// binary search
+ const auto & unicode_map_lowercase = get_unicode_map_lowercase();
auto it = std::lower_bound(unicode_map_lowercase.begin(), unicode_map_lowercase.end(), cpt,
[](const std::pair<uint32_t, uint32_t> & pair, uint32_t value) {
return pair.first < value;
@@ -883,8 +890,7 @@
return false;
}
-std::vector<std::string> unicode_regex_split(const std::string & text, const std::vector<std::string> & regex_exprs) {
- // unicode categories
+static const std::map<std::string, int> & get_k_ucat_enum() {
static const std::map<std::string, int> k_ucat_enum = {
{ "\\p{N}", unicode_cpt_flags::NUMBER },
{ "\\p{L}", unicode_cpt_flags::LETTER },
@@ -892,7 +898,10 @@
{ "\\p{M}", unicode_cpt_flags::ACCENT_MARK },
{ "\\p{S}", unicode_cpt_flags::SYMBOL },
};
+ return k_ucat_enum;
+}
+static const std::map<int, int> & get_k_ucat_cpt() {
static const std::map<int, int> k_ucat_cpt = {
{ unicode_cpt_flags::NUMBER, 0xD1 },
{ unicode_cpt_flags::LETTER, 0xD2 },
@@ -900,7 +909,10 @@
{ unicode_cpt_flags::ACCENT_MARK, 0xD4 },
{ unicode_cpt_flags::SYMBOL, 0xD5 },
};
+ return k_ucat_cpt;
+}
+static const std::map<int, std::string> & get_k_ucat_map() {
static const std::map<int, std::string> k_ucat_map = {
{ unicode_cpt_flags::NUMBER, "\x30-\x39" }, // 0-9
{ unicode_cpt_flags::LETTER, "\x41-\x5A\x61-\x7A" }, // A-Za-z
@@ -908,6 +920,14 @@
{ unicode_cpt_flags::ACCENT_MARK, "" }, // no sub-128 codepoints
{ unicode_cpt_flags::SYMBOL, "\\\x24\\\x2B\x3C-\x3E\x5E\x60\\\x7C" }, // $+<=>^`|
};
+ return k_ucat_map;
+}
+
+std::vector<std::string> unicode_regex_split(const std::string & text, const std::vector<std::string> & regex_exprs) {
+ // unicode categories
+ const auto & k_ucat_enum = get_k_ucat_enum();
+ const auto & k_ucat_cpt = get_k_ucat_cpt();
+ const auto & k_ucat_map = get_k_ucat_map();
// compute collapsed codepoints only if needed by at least one regex
bool need_collapse = false;