Upload 6 files

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/data1/yuyijiong/\u81ea\u5df1\u8bad\u7ec3\u7684\u6a21\u578b-merged/qwen-chat-14b-32k-pre-lora-yarn-100step/",
+  "_name_or_path": "/data1/\u9884\u8bad\u7ec3\u6a21\u578b/Qwen-7b-chat/",
   "architectures": [
     "QWenLMHeadModel"
   ],
@@ -12,31 +12,30 @@
   "emb_dropout_prob": 0.0,
   "fp16": false,
   "fp32": false,
-  "hidden_size": 5120,
+  "hidden_size": 4096,
   "initializer_range": 0.02,
-  "intermediate_size": 27392,
+  "intermediate_size": 22016,
   "kv_channels": 128,
   "layer_norm_epsilon": 1e-06,
   "max_position_embeddings": 8192,
   "model_type": "qwen",
   "no_bias": true,
-  "num_attention_heads": 40,
-  "num_hidden_layers": 40,
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
   "onnx_safe": null,
   "rotary_emb_base": 10000,
   "rotary_pct": 1.0,
   "scale_attn_weights": true,
-  "seq_length": 2048,
-  "softmax_in_fp32": false,
+  "seq_length": 8192,
   "tie_word_embeddings": false,
   "tokenizer_class": "QWenTokenizer",
   "torch_dtype": "bfloat16",
-  "transformers_version": "4.36.0.dev0",
+  "transformers_version": "4.36.2",
   "use_cache": true,
   "use_cache_kernel": false,
   "use_cache_quantization": false,
   "use_dynamic_ntk": true,
   "use_flash_attn": true,
   "use_logn_attn": false,
-  "vocab_size": 152064
+  "vocab_size": 151936
 }

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a8006909c3ae914e269deb90f851061d470d72369d6639b0a93a202d16c6743
+size 4988485656

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac6b03f2d8dd1bfce46d7ac09162072d20d241ceecc791841fc3d42486904583
+size 4981246520

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e9e03d7551091f10dfd3fbc1227e3978d0e9fc667b8619d72a18a93ffd3dd64
+size 4228285288

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf5a6b62604f9e150a040e7ee748ac2ba91d6c96affac11f576bb9fcea3d222e
+size 1244659840

model.safetensors.index.json

@@ -1,330 +1,266 @@
 {
   "metadata": {
-    "total_size": 28334581760
+    "total_size": 15442649088
   },
   "weight_map": {
-    "lm_head.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.0.attn.c_attn.bias": "model-00001-of-00006.safetensors",
-    "transformer.h.0.attn.c_attn.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.0.attn.c_proj.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.0.ln_1.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.0.ln_2.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.0.mlp.c_proj.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.0.mlp.w1.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.0.mlp.w2.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.1.attn.c_attn.bias": "model-00001-of-00006.safetensors",
-    "transformer.h.1.attn.c_attn.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.1.attn.c_proj.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.1.ln_1.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.1.ln_2.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.1.mlp.c_proj.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.1.mlp.w1.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.1.mlp.w2.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.10.attn.c_attn.bias": "model-00002-of-00006.safetensors",
-    "transformer.h.10.attn.c_attn.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.10.attn.c_proj.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.10.ln_1.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.10.ln_2.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.10.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.10.mlp.w1.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.10.mlp.w2.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.11.attn.c_attn.bias": "model-00002-of-00006.safetensors",
-    "transformer.h.11.attn.c_attn.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.11.attn.c_proj.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.11.ln_1.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.11.ln_2.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.11.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.11.mlp.w1.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.11.mlp.w2.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.12.attn.c_attn.bias": "model-00002-of-00006.safetensors",
-    "transformer.h.12.attn.c_attn.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.12.attn.c_proj.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.12.ln_1.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.12.ln_2.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.12.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.12.mlp.w1.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.12.mlp.w2.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.13.attn.c_attn.bias": "model-00002-of-00006.safetensors",
-    "transformer.h.13.attn.c_attn.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.13.attn.c_proj.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.13.ln_1.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.13.ln_2.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.13.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.13.mlp.w1.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.13.mlp.w2.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.14.attn.c_attn.bias": "model-00003-of-00006.safetensors",
-    "transformer.h.14.attn.c_attn.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.14.attn.c_proj.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.14.ln_1.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.14.ln_2.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.14.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.14.mlp.w1.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.14.mlp.w2.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.15.attn.c_attn.bias": "model-00003-of-00006.safetensors",
-    "transformer.h.15.attn.c_attn.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.15.attn.c_proj.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.15.ln_1.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.15.ln_2.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.15.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.15.mlp.w1.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.15.mlp.w2.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.16.attn.c_attn.bias": "model-00003-of-00006.safetensors",
-    "transformer.h.16.attn.c_attn.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.16.attn.c_proj.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.16.ln_1.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.16.ln_2.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.16.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.16.mlp.w1.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.16.mlp.w2.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.17.attn.c_attn.bias": "model-00003-of-00006.safetensors",
-    "transformer.h.17.attn.c_attn.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.17.attn.c_proj.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.17.ln_1.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.17.ln_2.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.17.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.17.mlp.w1.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.17.mlp.w2.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.18.attn.c_attn.bias": "model-00003-of-00006.safetensors",
-    "transformer.h.18.attn.c_attn.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.18.attn.c_proj.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.18.ln_1.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.18.ln_2.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.18.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.18.mlp.w1.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.18.mlp.w2.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.19.attn.c_attn.bias": "model-00003-of-00006.safetensors",
-    "transformer.h.19.attn.c_attn.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.19.attn.c_proj.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.19.ln_1.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.19.ln_2.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.19.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.19.mlp.w1.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.19.mlp.w2.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.2.attn.c_attn.bias": "model-00001-of-00006.safetensors",
-    "transformer.h.2.attn.c_attn.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.2.attn.c_proj.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.2.ln_1.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.2.ln_2.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.2.mlp.c_proj.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.2.mlp.w1.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.2.mlp.w2.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.20.attn.c_attn.bias": "model-00003-of-00006.safetensors",
-    "transformer.h.20.attn.c_attn.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.20.attn.c_proj.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.20.ln_1.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.20.ln_2.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.20.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.20.mlp.w1.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.20.mlp.w2.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.21.attn.c_attn.bias": "model-00004-of-00006.safetensors",
-    "transformer.h.21.attn.c_attn.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.21.attn.c_proj.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.21.ln_1.weight": "model-00003-of-00006.safetensors",
-    "transformer.h.21.ln_2.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.21.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.21.mlp.w1.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.21.mlp.w2.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.22.attn.c_attn.bias": "model-00004-of-00006.safetensors",
-    "transformer.h.22.attn.c_attn.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.22.attn.c_proj.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.22.ln_1.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.22.ln_2.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.22.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.22.mlp.w1.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.22.mlp.w2.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.23.attn.c_attn.bias": "model-00004-of-00006.safetensors",
-    "transformer.h.23.attn.c_attn.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.23.attn.c_proj.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.23.ln_1.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.23.ln_2.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.23.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.23.mlp.w1.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.23.mlp.w2.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.24.attn.c_attn.bias": "model-00004-of-00006.safetensors",
-    "transformer.h.24.attn.c_attn.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.24.attn.c_proj.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.24.ln_1.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.24.ln_2.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.24.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.24.mlp.w1.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.24.mlp.w2.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.25.attn.c_attn.bias": "model-00004-of-00006.safetensors",
-    "transformer.h.25.attn.c_attn.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.25.attn.c_proj.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.25.ln_1.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.25.ln_2.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.25.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.25.mlp.w1.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.25.mlp.w2.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.26.attn.c_attn.bias": "model-00004-of-00006.safetensors",
-    "transformer.h.26.attn.c_attn.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.26.attn.c_proj.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.26.ln_1.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.26.ln_2.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.26.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.26.mlp.w1.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.26.mlp.w2.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.27.attn.c_attn.bias": "model-00004-of-00006.safetensors",
-    "transformer.h.27.attn.c_attn.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.27.attn.c_proj.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.27.ln_1.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.27.ln_2.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.27.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.27.mlp.w1.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.27.mlp.w2.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.28.attn.c_attn.bias": "model-00004-of-00006.safetensors",
-    "transformer.h.28.attn.c_attn.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.28.attn.c_proj.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.28.ln_1.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.28.ln_2.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.28.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.28.mlp.w1.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.28.mlp.w2.weight": "model-00004-of-00006.safetensors",
-    "transformer.h.29.attn.c_attn.bias": "model-00005-of-00006.safetensors",
-    "transformer.h.29.attn.c_attn.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.29.attn.c_proj.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.29.ln_1.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.29.ln_2.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.29.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.29.mlp.w1.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.29.mlp.w2.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.3.attn.c_attn.bias": "model-00001-of-00006.safetensors",
-    "transformer.h.3.attn.c_attn.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.3.attn.c_proj.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.3.ln_1.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.3.ln_2.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.3.mlp.c_proj.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.3.mlp.w1.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.3.mlp.w2.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.30.attn.c_attn.bias": "model-00005-of-00006.safetensors",
-    "transformer.h.30.attn.c_attn.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.30.attn.c_proj.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.30.ln_1.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.30.ln_2.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.30.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.30.mlp.w1.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.30.mlp.w2.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.31.attn.c_attn.bias": "model-00005-of-00006.safetensors",
-    "transformer.h.31.attn.c_attn.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.31.attn.c_proj.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.31.ln_1.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.31.ln_2.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.31.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.31.mlp.w1.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.31.mlp.w2.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.32.attn.c_attn.bias": "model-00005-of-00006.safetensors",
-    "transformer.h.32.attn.c_attn.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.32.attn.c_proj.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.32.ln_1.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.32.ln_2.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.32.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.32.mlp.w1.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.32.mlp.w2.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.33.attn.c_attn.bias": "model-00005-of-00006.safetensors",
-    "transformer.h.33.attn.c_attn.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.33.attn.c_proj.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.33.ln_1.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.33.ln_2.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.33.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.33.mlp.w1.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.33.mlp.w2.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.34.attn.c_attn.bias": "model-00005-of-00006.safetensors",
-    "transformer.h.34.attn.c_attn.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.34.attn.c_proj.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.34.ln_1.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.34.ln_2.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.34.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.34.mlp.w1.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.34.mlp.w2.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.35.attn.c_attn.bias": "model-00005-of-00006.safetensors",
-    "transformer.h.35.attn.c_attn.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.35.attn.c_proj.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.35.ln_1.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.35.ln_2.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.35.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.35.mlp.w1.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.35.mlp.w2.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.36.attn.c_attn.bias": "model-00005-of-00006.safetensors",
-    "transformer.h.36.attn.c_attn.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.36.attn.c_proj.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.36.ln_1.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.36.ln_2.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.36.mlp.c_proj.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.36.mlp.w1.weight": "model-00005-of-00006.safetensors",
-    "transformer.h.36.mlp.w2.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.37.attn.c_attn.bias": "model-00006-of-00006.safetensors",
-    "transformer.h.37.attn.c_attn.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.37.attn.c_proj.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.37.ln_1.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.37.ln_2.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.37.mlp.c_proj.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.37.mlp.w1.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.37.mlp.w2.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.38.attn.c_attn.bias": "model-00006-of-00006.safetensors",
-    "transformer.h.38.attn.c_attn.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.38.attn.c_proj.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.38.ln_1.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.38.ln_2.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.38.mlp.c_proj.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.38.mlp.w1.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.38.mlp.w2.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.39.attn.c_attn.bias": "model-00006-of-00006.safetensors",
-    "transformer.h.39.attn.c_attn.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.39.attn.c_proj.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.39.ln_1.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.39.ln_2.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.39.mlp.c_proj.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.39.mlp.w1.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.39.mlp.w2.weight": "model-00006-of-00006.safetensors",
-    "transformer.h.4.attn.c_attn.bias": "model-00001-of-00006.safetensors",
-    "transformer.h.4.attn.c_attn.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.4.attn.c_proj.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.4.ln_1.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.4.ln_2.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.4.mlp.c_proj.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.4.mlp.w1.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.4.mlp.w2.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.5.attn.c_attn.bias": "model-00001-of-00006.safetensors",
-    "transformer.h.5.attn.c_attn.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.5.attn.c_proj.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.5.ln_1.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.5.ln_2.weight": "model-00001-of-00006.safetensors",
-    "transformer.h.5.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.5.mlp.w1.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.5.mlp.w2.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.6.attn.c_attn.bias": "model-00002-of-00006.safetensors",
-    "transformer.h.6.attn.c_attn.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.6.attn.c_proj.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.6.ln_1.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.6.ln_2.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.6.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.6.mlp.w1.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.6.mlp.w2.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.7.attn.c_attn.bias": "model-00002-of-00006.safetensors",
-    "transformer.h.7.attn.c_attn.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.7.attn.c_proj.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.7.ln_1.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.7.ln_2.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.7.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.7.mlp.w1.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.7.mlp.w2.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.8.attn.c_attn.bias": "model-00002-of-00006.safetensors",
-    "transformer.h.8.attn.c_attn.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.8.attn.c_proj.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.8.ln_1.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.8.ln_2.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.8.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.8.mlp.w1.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.8.mlp.w2.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.9.attn.c_attn.bias": "model-00002-of-00006.safetensors",
-    "transformer.h.9.attn.c_attn.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.9.attn.c_proj.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.9.ln_1.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.9.ln_2.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.9.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.9.mlp.w1.weight": "model-00002-of-00006.safetensors",
-    "transformer.h.9.mlp.w2.weight": "model-00002-of-00006.safetensors",
-    "transformer.ln_f.weight": "model-00006-of-00006.safetensors",
-    "transformer.wte.weight": "model-00001-of-00006.safetensors"
+    "lm_head.weight": "model-00004-of-00004.safetensors",
+    "transformer.h.0.attn.c_attn.bias": "model-00001-of-00004.safetensors",
+    "transformer.h.0.attn.c_attn.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.0.attn.c_proj.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.0.ln_1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.0.ln_2.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.0.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.0.mlp.w1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.0.mlp.w2.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.1.attn.c_attn.bias": "model-00001-of-00004.safetensors",
+    "transformer.h.1.attn.c_attn.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.1.attn.c_proj.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.1.ln_1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.1.ln_2.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.1.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.1.mlp.w1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.1.mlp.w2.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.10.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+    "transformer.h.10.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.10.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.10.ln_1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.10.ln_2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.10.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.10.mlp.w1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.10.mlp.w2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.11.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+    "transformer.h.11.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.11.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.11.ln_1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.11.ln_2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.11.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.11.mlp.w1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.11.mlp.w2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.12.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+    "transformer.h.12.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.12.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.12.ln_1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.12.ln_2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.12.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.12.mlp.w1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.12.mlp.w2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.13.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+    "transformer.h.13.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.13.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.13.ln_1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.13.ln_2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.13.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.13.mlp.w1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.13.mlp.w2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.14.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+    "transformer.h.14.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.14.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.14.ln_1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.14.ln_2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.14.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.14.mlp.w1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.14.mlp.w2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.15.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+    "transformer.h.15.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.15.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.15.ln_1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.15.ln_2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.15.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.15.mlp.w1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.15.mlp.w2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.16.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+    "transformer.h.16.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.16.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.16.ln_1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.16.ln_2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.16.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.16.mlp.w1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.16.mlp.w2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.17.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+    "transformer.h.17.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.17.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.17.ln_1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.17.ln_2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.17.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.17.mlp.w1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.17.mlp.w2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.18.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+    "transformer.h.18.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.18.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.18.ln_1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.18.ln_2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.18.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.18.mlp.w1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.18.mlp.w2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.19.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+    "transformer.h.19.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.19.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.19.ln_1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.19.ln_2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.19.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.19.mlp.w1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.19.mlp.w2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.2.attn.c_attn.bias": "model-00001-of-00004.safetensors",
+    "transformer.h.2.attn.c_attn.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.2.attn.c_proj.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.2.ln_1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.2.ln_2.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.2.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.2.mlp.w1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.2.mlp.w2.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.20.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+    "transformer.h.20.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.20.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.20.ln_1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.20.ln_2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.20.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.20.mlp.w1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.20.mlp.w2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.21.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+    "transformer.h.21.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.21.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.21.ln_1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.21.ln_2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.21.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.21.mlp.w1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.21.mlp.w2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.22.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+    "transformer.h.22.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.22.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.22.ln_1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.22.ln_2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.22.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.22.mlp.w1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.22.mlp.w2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.23.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+    "transformer.h.23.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.23.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.23.ln_1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.23.ln_2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.23.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.23.mlp.w1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.23.mlp.w2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.24.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+    "transformer.h.24.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.24.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.24.ln_1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.24.ln_2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.24.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.24.mlp.w1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.24.mlp.w2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.25.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+    "transformer.h.25.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.25.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.25.ln_1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.25.ln_2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.25.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.25.mlp.w1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.25.mlp.w2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.26.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+    "transformer.h.26.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.26.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.26.ln_1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.26.ln_2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.26.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.26.mlp.w1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.26.mlp.w2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.27.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+    "transformer.h.27.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.27.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.27.ln_1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.27.ln_2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.27.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.27.mlp.w1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.27.mlp.w2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.28.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+    "transformer.h.28.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.28.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.28.ln_1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.28.ln_2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.28.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.28.mlp.w1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.28.mlp.w2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.29.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+    "transformer.h.29.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.29.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.29.ln_1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.29.ln_2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.29.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.29.mlp.w1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.29.mlp.w2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.3.attn.c_attn.bias": "model-00001-of-00004.safetensors",
+    "transformer.h.3.attn.c_attn.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.3.attn.c_proj.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.3.ln_1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.3.ln_2.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.3.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.3.mlp.w1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.3.mlp.w2.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.30.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+    "transformer.h.30.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.30.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.30.ln_1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.30.ln_2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.30.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.30.mlp.w1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.30.mlp.w2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.31.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+    "transformer.h.31.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.31.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.31.ln_1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.31.ln_2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.31.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.31.mlp.w1.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.31.mlp.w2.weight": "model-00003-of-00004.safetensors",
+    "transformer.h.4.attn.c_attn.bias": "model-00001-of-00004.safetensors",
+    "transformer.h.4.attn.c_attn.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.4.attn.c_proj.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.4.ln_1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.4.ln_2.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.4.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.4.mlp.w1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.4.mlp.w2.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.5.attn.c_attn.bias": "model-00001-of-00004.safetensors",
+    "transformer.h.5.attn.c_attn.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.5.attn.c_proj.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.5.ln_1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.5.ln_2.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.5.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.5.mlp.w1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.5.mlp.w2.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.6.attn.c_attn.bias": "model-00001-of-00004.safetensors",
+    "transformer.h.6.attn.c_attn.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.6.attn.c_proj.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.6.ln_1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.6.ln_2.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.6.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.6.mlp.w1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.6.mlp.w2.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.7.attn.c_attn.bias": "model-00001-of-00004.safetensors",
+    "transformer.h.7.attn.c_attn.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.7.attn.c_proj.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.7.ln_1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.7.ln_2.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.7.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.7.mlp.w1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.7.mlp.w2.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.8.attn.c_attn.bias": "model-00001-of-00004.safetensors",
+    "transformer.h.8.attn.c_attn.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.8.attn.c_proj.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.8.ln_1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.8.ln_2.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.8.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.8.mlp.w1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.8.mlp.w2.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.9.attn.c_attn.bias": "model-00001-of-00004.safetensors",
+    "transformer.h.9.attn.c_attn.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.9.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.9.ln_1.weight": "model-00001-of-00004.safetensors",
+    "transformer.h.9.ln_2.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.9.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.9.mlp.w1.weight": "model-00002-of-00004.safetensors",
+    "transformer.h.9.mlp.w2.weight": "model-00002-of-00004.safetensors",
+    "transformer.ln_f.weight": "model-00003-of-00004.safetensors",
+    "transformer.wte.weight": "model-00001-of-00004.safetensors"
   }
 }