diff --git a/Readme.md b/Readme.md index af96411..bb8d458 100644 --- a/Readme.md +++ b/Readme.md @@ -5,19 +5,23 @@ input_ids = tokenizer.build_chat_input(query, history=history, role=role) -input_ids -> [1, 6] 1:batch_num 6:sequence_length -inputs_embeds -> [6, 1, 4096] 4096:hidden_size -rotary_pos_emb -> [6, 1, 32, 2] 32:pos的编码维度 2:cos+sin +for + input_ids -> [1, 6] 1:batch_num 6:sequence_length + inputs_embeds -> [6, 1, 4096] 4096:hidden_size + rotary_pos_emb -> [6, 1, 32, 2] 32:pos的编码维度 2:cos+sin -hidden_states = inputs_embeds -for layers : GLMBlock(hidden_states, rotary_pos_emb) -hidden_states = self.final_layernorm(hidden_states) -hidden_states = hidden_states[-1:] -lm_logits = self.output_layer(hidden_states) -lm_logits = lm_logits.transpose(0, 1).contiguous() -> [1, 1, 65024] + hidden_states = inputs_embeds + for layers : GLMBlock(hidden_states, rotary_pos_emb) + hidden_states = self.final_layernorm(hidden_states) + hidden_states = hidden_states[-1:] + lm_logits = self.output_layer(hidden_states) + lm_logits = lm_logits.transpose(0, 1).contiguous() -> [1, 1, 65024] -probs = softmax(lm_logits) -> [1, 65024] -next_tokens = torch.multinomial(probs, num_samples=1) 采样 -> [1] 1:batch_num -input_ids = torch.cat([input_ids, next_tokens) -> [1, 7] 1:batch_num + probs = softmax(lm_logits) -> [1, 65024] + next_tokens = torch.multinomial(probs, num_samples=1) 采样 -> [1] 1:batch_num + + if next_tokens == eos_token_id 推理结束退出循环 + + input_ids = torch.cat([input_ids, next_tokens) -> [1, 7] 1:batch_num response = tokenizer.decode(outputs) \ No newline at end of file diff --git a/chatglm/config.json b/chatglm/config.json new file mode 100644 index 0000000..37933c8 --- /dev/null +++ b/chatglm/config.json @@ -0,0 +1,42 @@ +{ + "_name_or_path": "THUDM/chatglm3-6b", + "model_type": "chatglm", + "architectures": [ + "ChatGLMModel" + ], + "auto_map": { + "AutoConfig": "configuration_chatglm.ChatGLMConfig", + "AutoModel": "modeling_chatglm.ChatGLMForConditionalGeneration", + "AutoModelForCausalLM": "modeling_chatglm.ChatGLMForConditionalGeneration", + "AutoModelForSeq2SeqLM": "modeling_chatglm.ChatGLMForConditionalGeneration", + "AutoModelForSequenceClassification": "modeling_chatglm.ChatGLMForSequenceClassification" + }, + "add_bias_linear": false, + "add_qkv_bias": true, + "apply_query_key_layer_scaling": true, + "apply_residual_connection_post_layernorm": false, + "attention_dropout": 0.0, + "attention_softmax_in_fp32": true, + "bias_dropout_fusion": true, + "ffn_hidden_size": 13696, + "fp32_residual_connection": false, + "hidden_dropout": 0.0, + "hidden_size": 4096, + "kv_channels": 128, + "layernorm_epsilon": 1e-05, + "multi_query_attention": true, + "multi_query_group_num": 2, + "num_attention_heads": 32, + "num_layers": 28, + "original_rope": true, + "padded_vocab_size": 65024, + "post_layer_norm": true, + "rmsnorm": true, + "seq_length": 8192, + "use_cache": true, + "torch_dtype": "float16", + "transformers_version": "4.30.2", + "tie_word_embeddings": false, + "eos_token_id": 2, + "pad_token_id": 0 +} \ No newline at end of file diff --git a/chatglm/configuration.json b/chatglm/configuration.json new file mode 100644 index 0000000..262fd55 --- /dev/null +++ b/chatglm/configuration.json @@ -0,0 +1 @@ +{"framework":"Pytorch","task":"chatbot"} \ No newline at end of file diff --git a/chatglm/modeling_chatglm.py b/chatglm/modeling_chatglm.py index 9e16ea4..a32728c 100644 --- a/chatglm/modeling_chatglm.py +++ b/chatglm/modeling_chatglm.py @@ -700,10 +700,10 @@ class ChatGLMForConditionalGeneration(nn.Module): # finished sentences should add a padding token to next pad_token = pad_token_id * isFinished next_tokens = next_tokens * (1 - isFinished) + pad_token - input_ids = torch.cat([input_ids, next_tokens[:, None]], dim=-1) isFinished = isFinished | next_tokens.eq(eos_token_id_tensor) if isFinished.min() == 1: # all batch is finish break + input_ids = torch.cat([input_ids, next_tokens[:, None]], dim=-1) return input_ids diff --git a/chatglm/pytorch_model.bin.index.json b/chatglm/pytorch_model.bin.index.json new file mode 100644 index 0000000..cd6a94a --- /dev/null +++ b/chatglm/pytorch_model.bin.index.json @@ -0,0 +1,207 @@ +{ + "metadata": { + "total_size": 12487168064 + }, + "weight_map": { + "transformer.embedding.word_embeddings.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.final_layernorm.weight": "pytorch_model-00007-of-00007.bin", + "transformer.encoder.layers.0.input_layernorm.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.0.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.0.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.0.post_attention_layernorm.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.0.self_attention.dense.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.0.self_attention.query_key_value.bias": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.0.self_attention.query_key_value.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.1.input_layernorm.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.1.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.1.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.1.post_attention_layernorm.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.1.self_attention.dense.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.1.self_attention.query_key_value.bias": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.1.self_attention.query_key_value.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.10.input_layernorm.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.10.mlp.dense_4h_to_h.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.10.mlp.dense_h_to_4h.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.10.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.10.self_attention.dense.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.10.self_attention.query_key_value.bias": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.10.self_attention.query_key_value.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.11.input_layernorm.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.11.mlp.dense_4h_to_h.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.11.mlp.dense_h_to_4h.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.11.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.11.self_attention.dense.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.11.self_attention.query_key_value.bias": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.11.self_attention.query_key_value.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.12.input_layernorm.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.12.mlp.dense_4h_to_h.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.12.mlp.dense_h_to_4h.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.12.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.12.self_attention.dense.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.12.self_attention.query_key_value.bias": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.12.self_attention.query_key_value.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.13.input_layernorm.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.13.mlp.dense_4h_to_h.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.13.mlp.dense_h_to_4h.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.13.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.13.self_attention.dense.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.13.self_attention.query_key_value.bias": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.13.self_attention.query_key_value.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.14.input_layernorm.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.14.mlp.dense_4h_to_h.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.14.mlp.dense_h_to_4h.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.14.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.14.self_attention.dense.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.14.self_attention.query_key_value.bias": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.14.self_attention.query_key_value.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.15.input_layernorm.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.15.mlp.dense_4h_to_h.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.15.mlp.dense_h_to_4h.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.15.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.15.self_attention.dense.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.15.self_attention.query_key_value.bias": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.15.self_attention.query_key_value.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.16.input_layernorm.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.16.mlp.dense_4h_to_h.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.16.mlp.dense_h_to_4h.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.16.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.16.self_attention.dense.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.16.self_attention.query_key_value.bias": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.16.self_attention.query_key_value.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.17.input_layernorm.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.17.mlp.dense_4h_to_h.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.17.mlp.dense_h_to_4h.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.17.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.17.self_attention.dense.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.17.self_attention.query_key_value.bias": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.17.self_attention.query_key_value.weight": "pytorch_model-00004-of-00007.bin", + "transformer.encoder.layers.18.input_layernorm.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.18.mlp.dense_4h_to_h.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.18.mlp.dense_h_to_4h.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.18.post_attention_layernorm.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.18.self_attention.dense.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.18.self_attention.query_key_value.bias": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.18.self_attention.query_key_value.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.19.input_layernorm.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.19.mlp.dense_4h_to_h.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.19.mlp.dense_h_to_4h.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.19.post_attention_layernorm.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.19.self_attention.dense.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.19.self_attention.query_key_value.bias": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.19.self_attention.query_key_value.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.2.input_layernorm.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.2.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.2.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.2.post_attention_layernorm.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.2.self_attention.dense.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.2.self_attention.query_key_value.bias": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.2.self_attention.query_key_value.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.20.input_layernorm.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.20.mlp.dense_4h_to_h.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.20.mlp.dense_h_to_4h.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.20.post_attention_layernorm.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.20.self_attention.dense.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.20.self_attention.query_key_value.bias": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.20.self_attention.query_key_value.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.21.input_layernorm.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.21.mlp.dense_4h_to_h.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.21.mlp.dense_h_to_4h.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.21.post_attention_layernorm.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.21.self_attention.dense.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.21.self_attention.query_key_value.bias": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.21.self_attention.query_key_value.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.22.input_layernorm.weight": "pytorch_model-00005-of-00007.bin", + "transformer.encoder.layers.22.mlp.dense_4h_to_h.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.22.mlp.dense_h_to_4h.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.22.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.22.self_attention.dense.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.22.self_attention.query_key_value.bias": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.22.self_attention.query_key_value.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.23.input_layernorm.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.23.mlp.dense_4h_to_h.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.23.mlp.dense_h_to_4h.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.23.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.23.self_attention.dense.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.23.self_attention.query_key_value.bias": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.23.self_attention.query_key_value.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.24.input_layernorm.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.24.mlp.dense_4h_to_h.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.24.mlp.dense_h_to_4h.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.24.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.24.self_attention.dense.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.24.self_attention.query_key_value.bias": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.24.self_attention.query_key_value.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.25.input_layernorm.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.25.mlp.dense_4h_to_h.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.25.mlp.dense_h_to_4h.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.25.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.25.self_attention.dense.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.25.self_attention.query_key_value.bias": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.25.self_attention.query_key_value.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.26.input_layernorm.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.26.mlp.dense_4h_to_h.weight": "pytorch_model-00007-of-00007.bin", + "transformer.encoder.layers.26.mlp.dense_h_to_4h.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.26.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.26.self_attention.dense.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.26.self_attention.query_key_value.bias": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.26.self_attention.query_key_value.weight": "pytorch_model-00006-of-00007.bin", + "transformer.encoder.layers.27.input_layernorm.weight": "pytorch_model-00007-of-00007.bin", + "transformer.encoder.layers.27.mlp.dense_4h_to_h.weight": "pytorch_model-00007-of-00007.bin", + "transformer.encoder.layers.27.mlp.dense_h_to_4h.weight": "pytorch_model-00007-of-00007.bin", + "transformer.encoder.layers.27.post_attention_layernorm.weight": "pytorch_model-00007-of-00007.bin", + "transformer.encoder.layers.27.self_attention.dense.weight": "pytorch_model-00007-of-00007.bin", + "transformer.encoder.layers.27.self_attention.query_key_value.bias": "pytorch_model-00007-of-00007.bin", + "transformer.encoder.layers.27.self_attention.query_key_value.weight": "pytorch_model-00007-of-00007.bin", + "transformer.encoder.layers.3.input_layernorm.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.3.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.3.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.3.post_attention_layernorm.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.3.self_attention.dense.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.3.self_attention.query_key_value.bias": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.3.self_attention.query_key_value.weight": "pytorch_model-00001-of-00007.bin", + "transformer.encoder.layers.4.input_layernorm.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.4.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.4.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.4.post_attention_layernorm.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.4.self_attention.dense.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.4.self_attention.query_key_value.bias": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.4.self_attention.query_key_value.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.5.input_layernorm.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.5.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.5.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.5.post_attention_layernorm.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.5.self_attention.dense.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.5.self_attention.query_key_value.bias": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.5.self_attention.query_key_value.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.6.input_layernorm.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.6.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.6.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.6.post_attention_layernorm.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.6.self_attention.dense.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.6.self_attention.query_key_value.bias": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.6.self_attention.query_key_value.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.7.input_layernorm.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.7.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.7.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.7.post_attention_layernorm.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.7.self_attention.dense.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.7.self_attention.query_key_value.bias": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.7.self_attention.query_key_value.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.8.input_layernorm.weight": "pytorch_model-00002-of-00007.bin", + "transformer.encoder.layers.8.mlp.dense_4h_to_h.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.8.mlp.dense_h_to_4h.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.8.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.8.self_attention.dense.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.8.self_attention.query_key_value.bias": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.8.self_attention.query_key_value.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.9.input_layernorm.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.9.mlp.dense_4h_to_h.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.9.mlp.dense_h_to_4h.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.9.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.9.self_attention.dense.weight": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.9.self_attention.query_key_value.bias": "pytorch_model-00003-of-00007.bin", + "transformer.encoder.layers.9.self_attention.query_key_value.weight": "pytorch_model-00003-of-00007.bin", + "transformer.output_layer.weight": "pytorch_model-00007-of-00007.bin", + "transformer.rotary_pos_emb.inv_freq": "pytorch_model-00001-of-00007.bin" + } +} \ No newline at end of file