upload model

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "/dataset/granite_ckpt/rpanda/ibm_models/granite-code-models-release/granite-20b-code/instruct-p1",
+  "activation_function": "gelu",
+  "architectures": [
+    "GPTBigCodeForCausalLM"
+  ],
+  "attention_softmax_in_fp32": true,
+  "attn_pdrop": 0.1,
+  "bos_token_id": 0,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 0,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt_bigcode",
+  "multi_query": true,
+  "n_embd": 6144,
+  "n_head": 48,
+  "n_inner": 24576,
+  "n_layer": 52,
+  "n_positions": 8192,
+  "pad_token_id": 0,
+  "resid_pdrop": 0.1,
+  "scale_attention_softmax_in_fp32": true,
+  "scale_attn_weights": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.41.0.dev0",
+  "use_cache": false,
+  "vocab_size": 49152
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "eos_token_id": 0,
+  "pad_token_id": 0,
+  "transformers_version": "4.41.0.dev0"
+}

model-00001-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3383b79a4dd323d7c73898dab684f5321d46cb1cf8a5b43638bf5471a206fab
+size 4952185256

model-00002-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7d946d1623ff174405492b7ef8e435824aeb396656507fa5f8a04f5eb2119127
+size 4930227792

model-00003-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e957539529055ebdb22a666d45c3e9177ca2babab216bd3a1339c3b544671e28
+size 4927118464

model-00004-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88bccb2ed8573058c40884b24768562c0d435ef01db4587c494b35d75afaeabd
+size 4930227864

model-00005-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dac1caa3a17ff34407b330d576c8c491a18e56ee03768aa3d9733611eeb67aa5
+size 4927118464

model-00006-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f90f340897243e0ccd9ebc4a8be78856d4ae2462ced3bd1b177a95585253e847
+size 4930227864

model-00007-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73872bba2b2dbe29fce015940f234b8ee50b22111f16eec159e162e07df11774
+size 4927118464

model-00008-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1df15bea1ce84ffd84c38a1cdc2edda597fe9d8c9cc91fca004fd06536bfdca1
+size 4930227864

model-00009-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c6e4520259c91595c44c3b4daa1bf4a2b41fa08a0fb47cae5dc765e1ac1d9cb
+size 679601176

model.safetensors.index.json

@@ -0,0 +1,635 @@
+{
+  "metadata": {
+    "total_size": 40133986304
+  },
+  "weight_map": {
+    "transformer.h.0.attn.c_attn.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.0.attn.c_attn.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.0.attn.c_proj.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.0.attn.c_proj.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.0.ln_1.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.0.ln_1.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.0.ln_2.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.0.ln_2.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.0.mlp.c_fc.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.0.mlp.c_fc.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.0.mlp.c_proj.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.0.mlp.c_proj.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.1.attn.c_attn.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.1.attn.c_attn.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.1.attn.c_proj.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.1.attn.c_proj.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.1.ln_1.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.1.ln_1.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.1.ln_2.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.1.ln_2.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.1.mlp.c_fc.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.1.mlp.c_fc.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.1.mlp.c_proj.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.1.mlp.c_proj.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.10.attn.c_attn.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.10.attn.c_attn.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.10.attn.c_proj.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.10.attn.c_proj.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.10.ln_1.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.10.ln_1.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.10.ln_2.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.10.ln_2.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.10.mlp.c_fc.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.10.mlp.c_fc.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.10.mlp.c_proj.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.10.mlp.c_proj.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.11.attn.c_attn.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.11.attn.c_attn.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.11.attn.c_proj.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.11.attn.c_proj.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.11.ln_1.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.11.ln_1.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.11.ln_2.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.11.ln_2.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.11.mlp.c_fc.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.11.mlp.c_fc.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.11.mlp.c_proj.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.11.mlp.c_proj.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.12.attn.c_attn.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.12.attn.c_attn.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.12.attn.c_proj.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.12.attn.c_proj.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.12.ln_1.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.12.ln_1.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.12.ln_2.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.12.ln_2.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.12.mlp.c_fc.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.12.mlp.c_fc.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.12.mlp.c_proj.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.12.mlp.c_proj.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.13.attn.c_attn.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.13.attn.c_attn.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.13.attn.c_proj.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.13.attn.c_proj.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.13.ln_1.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.13.ln_1.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.13.ln_2.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.13.ln_2.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.13.mlp.c_fc.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.13.mlp.c_fc.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.13.mlp.c_proj.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.13.mlp.c_proj.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.14.attn.c_attn.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.14.attn.c_attn.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.14.attn.c_proj.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.14.attn.c_proj.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.14.ln_1.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.14.ln_1.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.14.ln_2.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.14.ln_2.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.14.mlp.c_fc.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.14.mlp.c_fc.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.14.mlp.c_proj.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.14.mlp.c_proj.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.15.attn.c_attn.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.15.attn.c_attn.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.15.attn.c_proj.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.15.attn.c_proj.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.15.ln_1.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.15.ln_1.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.15.ln_2.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.15.ln_2.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.15.mlp.c_fc.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.15.mlp.c_fc.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.15.mlp.c_proj.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.15.mlp.c_proj.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.16.attn.c_attn.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.16.attn.c_attn.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.16.attn.c_proj.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.16.attn.c_proj.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.16.ln_1.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.16.ln_1.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.16.ln_2.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.16.ln_2.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.16.mlp.c_fc.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.16.mlp.c_fc.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.16.mlp.c_proj.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.16.mlp.c_proj.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.17.attn.c_attn.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.17.attn.c_attn.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.17.attn.c_proj.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.17.attn.c_proj.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.17.ln_1.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.17.ln_1.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.17.ln_2.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.17.ln_2.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.17.mlp.c_fc.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.17.mlp.c_fc.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.17.mlp.c_proj.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.17.mlp.c_proj.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.18.attn.c_attn.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.18.attn.c_attn.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.18.attn.c_proj.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.18.attn.c_proj.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.18.ln_1.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.18.ln_1.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.18.ln_2.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.18.ln_2.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.18.mlp.c_fc.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.18.mlp.c_fc.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.18.mlp.c_proj.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.18.mlp.c_proj.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.19.attn.c_attn.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.19.attn.c_attn.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.19.attn.c_proj.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.19.attn.c_proj.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.19.ln_1.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.19.ln_1.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.19.ln_2.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.19.ln_2.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.19.mlp.c_fc.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.19.mlp.c_fc.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.19.mlp.c_proj.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.19.mlp.c_proj.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.2.attn.c_attn.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.2.attn.c_attn.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.2.attn.c_proj.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.2.attn.c_proj.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.2.ln_1.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.2.ln_1.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.2.ln_2.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.2.ln_2.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.2.mlp.c_fc.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.2.mlp.c_fc.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.2.mlp.c_proj.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.2.mlp.c_proj.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.20.attn.c_attn.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.20.attn.c_attn.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.20.attn.c_proj.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.20.attn.c_proj.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.20.ln_1.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.20.ln_1.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.20.ln_2.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.20.ln_2.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.20.mlp.c_fc.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.20.mlp.c_fc.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.20.mlp.c_proj.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.20.mlp.c_proj.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.21.attn.c_attn.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.21.attn.c_attn.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.21.attn.c_proj.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.21.attn.c_proj.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.21.ln_1.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.21.ln_1.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.21.ln_2.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.21.ln_2.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.21.mlp.c_fc.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.21.mlp.c_fc.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.21.mlp.c_proj.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.21.mlp.c_proj.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.22.attn.c_attn.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.22.attn.c_attn.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.22.attn.c_proj.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.22.attn.c_proj.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.22.ln_1.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.22.ln_1.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.22.ln_2.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.22.ln_2.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.22.mlp.c_fc.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.22.mlp.c_fc.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.22.mlp.c_proj.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.22.mlp.c_proj.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.23.attn.c_attn.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.23.attn.c_attn.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.23.attn.c_proj.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.23.attn.c_proj.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.23.ln_1.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.23.ln_1.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.23.ln_2.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.23.ln_2.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.23.mlp.c_fc.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.23.mlp.c_fc.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.23.mlp.c_proj.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.23.mlp.c_proj.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.24.attn.c_attn.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.24.attn.c_attn.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.24.attn.c_proj.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.24.attn.c_proj.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.24.ln_1.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.24.ln_1.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.24.ln_2.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.24.ln_2.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.24.mlp.c_fc.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.24.mlp.c_fc.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.24.mlp.c_proj.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.24.mlp.c_proj.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.25.attn.c_attn.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.25.attn.c_attn.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.25.attn.c_proj.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.25.attn.c_proj.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.25.ln_1.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.25.ln_1.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.25.ln_2.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.25.ln_2.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.25.mlp.c_fc.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.25.mlp.c_fc.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.25.mlp.c_proj.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.25.mlp.c_proj.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.26.attn.c_attn.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.26.attn.c_attn.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.26.attn.c_proj.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.26.attn.c_proj.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.26.ln_1.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.26.ln_1.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.26.ln_2.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.26.ln_2.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.26.mlp.c_fc.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.26.mlp.c_fc.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.26.mlp.c_proj.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.26.mlp.c_proj.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.27.attn.c_attn.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.27.attn.c_attn.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.27.attn.c_proj.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.27.attn.c_proj.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.27.ln_1.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.27.ln_1.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.27.ln_2.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.27.ln_2.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.27.mlp.c_fc.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.27.mlp.c_fc.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.27.mlp.c_proj.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.27.mlp.c_proj.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.28.attn.c_attn.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.28.attn.c_attn.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.28.attn.c_proj.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.28.attn.c_proj.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.28.ln_1.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.28.ln_1.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.28.ln_2.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.28.ln_2.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.28.mlp.c_fc.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.28.mlp.c_fc.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.28.mlp.c_proj.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.28.mlp.c_proj.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.29.attn.c_attn.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.29.attn.c_attn.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.29.attn.c_proj.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.29.attn.c_proj.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.29.ln_1.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.29.ln_1.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.29.ln_2.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.29.ln_2.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.29.mlp.c_fc.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.29.mlp.c_fc.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.29.mlp.c_proj.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.29.mlp.c_proj.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.3.attn.c_attn.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.3.attn.c_attn.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.3.attn.c_proj.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.3.attn.c_proj.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.3.ln_1.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.3.ln_1.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.3.ln_2.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.3.ln_2.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.3.mlp.c_fc.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.3.mlp.c_fc.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.3.mlp.c_proj.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.3.mlp.c_proj.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.30.attn.c_attn.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.30.attn.c_attn.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.30.attn.c_proj.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.30.attn.c_proj.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.30.ln_1.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.30.ln_1.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.30.ln_2.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.30.ln_2.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.30.mlp.c_fc.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.30.mlp.c_fc.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.30.mlp.c_proj.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.30.mlp.c_proj.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.31.attn.c_attn.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.31.attn.c_attn.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.31.attn.c_proj.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.31.attn.c_proj.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.31.ln_1.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.31.ln_1.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.31.ln_2.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.31.ln_2.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.31.mlp.c_fc.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.31.mlp.c_fc.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.31.mlp.c_proj.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.31.mlp.c_proj.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.32.attn.c_attn.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.32.attn.c_attn.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.32.attn.c_proj.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.32.attn.c_proj.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.32.ln_1.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.32.ln_1.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.32.ln_2.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.32.ln_2.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.32.mlp.c_fc.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.32.mlp.c_fc.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.32.mlp.c_proj.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.32.mlp.c_proj.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.33.attn.c_attn.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.33.attn.c_attn.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.33.attn.c_proj.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.33.attn.c_proj.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.33.ln_1.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.33.ln_1.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.33.ln_2.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.33.ln_2.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.33.mlp.c_fc.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.33.mlp.c_fc.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.33.mlp.c_proj.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.33.mlp.c_proj.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.34.attn.c_attn.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.34.attn.c_attn.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.34.attn.c_proj.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.34.attn.c_proj.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.34.ln_1.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.34.ln_1.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.34.ln_2.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.34.ln_2.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.34.mlp.c_fc.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.34.mlp.c_fc.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.34.mlp.c_proj.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.34.mlp.c_proj.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.35.attn.c_attn.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.35.attn.c_attn.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.35.attn.c_proj.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.35.attn.c_proj.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.35.ln_1.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.35.ln_1.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.35.ln_2.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.35.ln_2.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.35.mlp.c_fc.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.35.mlp.c_fc.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.35.mlp.c_proj.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.35.mlp.c_proj.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.36.attn.c_attn.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.36.attn.c_attn.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.36.attn.c_proj.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.36.attn.c_proj.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.36.ln_1.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.36.ln_1.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.36.ln_2.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.36.ln_2.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.36.mlp.c_fc.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.36.mlp.c_fc.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.36.mlp.c_proj.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.36.mlp.c_proj.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.37.attn.c_attn.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.37.attn.c_attn.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.37.attn.c_proj.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.37.attn.c_proj.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.37.ln_1.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.37.ln_1.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.37.ln_2.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.37.ln_2.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.37.mlp.c_fc.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.37.mlp.c_fc.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.37.mlp.c_proj.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.37.mlp.c_proj.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.38.attn.c_attn.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.38.attn.c_attn.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.38.attn.c_proj.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.38.attn.c_proj.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.38.ln_1.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.38.ln_1.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.38.ln_2.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.38.ln_2.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.38.mlp.c_fc.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.38.mlp.c_fc.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.38.mlp.c_proj.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.38.mlp.c_proj.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.39.attn.c_attn.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.39.attn.c_attn.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.39.attn.c_proj.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.39.attn.c_proj.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.39.ln_1.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.39.ln_1.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.39.ln_2.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.39.ln_2.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.39.mlp.c_fc.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.39.mlp.c_fc.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.39.mlp.c_proj.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.39.mlp.c_proj.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.4.attn.c_attn.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.4.attn.c_attn.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.4.attn.c_proj.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.4.attn.c_proj.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.4.ln_1.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.4.ln_1.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.4.ln_2.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.4.ln_2.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.4.mlp.c_fc.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.4.mlp.c_fc.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.4.mlp.c_proj.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.4.mlp.c_proj.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.40.attn.c_attn.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.40.attn.c_attn.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.40.attn.c_proj.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.40.attn.c_proj.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.40.ln_1.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.40.ln_1.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.40.ln_2.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.40.ln_2.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.40.mlp.c_fc.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.40.mlp.c_fc.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.40.mlp.c_proj.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.40.mlp.c_proj.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.41.attn.c_attn.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.41.attn.c_attn.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.41.attn.c_proj.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.41.attn.c_proj.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.41.ln_1.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.41.ln_1.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.41.ln_2.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.41.ln_2.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.41.mlp.c_fc.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.41.mlp.c_fc.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.41.mlp.c_proj.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.41.mlp.c_proj.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.42.attn.c_attn.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.42.attn.c_attn.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.42.attn.c_proj.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.42.attn.c_proj.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.42.ln_1.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.42.ln_1.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.42.ln_2.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.42.ln_2.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.42.mlp.c_fc.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.42.mlp.c_fc.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.42.mlp.c_proj.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.42.mlp.c_proj.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.43.attn.c_attn.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.43.attn.c_attn.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.43.attn.c_proj.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.43.attn.c_proj.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.43.ln_1.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.43.ln_1.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.43.ln_2.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.43.ln_2.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.43.mlp.c_fc.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.43.mlp.c_fc.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.43.mlp.c_proj.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.43.mlp.c_proj.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.44.attn.c_attn.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.44.attn.c_attn.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.44.attn.c_proj.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.44.attn.c_proj.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.44.ln_1.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.44.ln_1.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.44.ln_2.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.44.ln_2.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.44.mlp.c_fc.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.44.mlp.c_fc.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.44.mlp.c_proj.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.44.mlp.c_proj.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.45.attn.c_attn.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.45.attn.c_attn.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.45.attn.c_proj.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.45.attn.c_proj.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.45.ln_1.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.45.ln_1.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.45.ln_2.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.45.ln_2.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.45.mlp.c_fc.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.45.mlp.c_fc.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.45.mlp.c_proj.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.45.mlp.c_proj.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.46.attn.c_attn.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.46.attn.c_attn.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.46.attn.c_proj.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.46.attn.c_proj.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.46.ln_1.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.46.ln_1.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.46.ln_2.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.46.ln_2.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.46.mlp.c_fc.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.46.mlp.c_fc.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.46.mlp.c_proj.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.46.mlp.c_proj.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.47.attn.c_attn.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.47.attn.c_attn.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.47.attn.c_proj.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.47.attn.c_proj.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.47.ln_1.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.47.ln_1.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.47.ln_2.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.47.ln_2.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.47.mlp.c_fc.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.47.mlp.c_fc.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.47.mlp.c_proj.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.47.mlp.c_proj.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.48.attn.c_attn.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.48.attn.c_attn.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.48.attn.c_proj.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.48.attn.c_proj.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.48.ln_1.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.48.ln_1.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.48.ln_2.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.48.ln_2.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.48.mlp.c_fc.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.48.mlp.c_fc.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.48.mlp.c_proj.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.48.mlp.c_proj.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.49.attn.c_attn.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.49.attn.c_attn.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.49.attn.c_proj.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.49.attn.c_proj.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.49.ln_1.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.49.ln_1.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.49.ln_2.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.49.ln_2.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.49.mlp.c_fc.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.49.mlp.c_fc.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.49.mlp.c_proj.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.49.mlp.c_proj.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.5.attn.c_attn.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.5.attn.c_attn.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.5.attn.c_proj.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.5.attn.c_proj.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.5.ln_1.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.5.ln_1.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.5.ln_2.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.5.ln_2.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.5.mlp.c_fc.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.5.mlp.c_fc.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.5.mlp.c_proj.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.5.mlp.c_proj.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.50.attn.c_attn.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.50.attn.c_attn.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.50.attn.c_proj.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.50.attn.c_proj.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.50.ln_1.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.50.ln_1.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.50.ln_2.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.50.ln_2.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.50.mlp.c_fc.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.50.mlp.c_fc.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.50.mlp.c_proj.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.50.mlp.c_proj.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.51.attn.c_attn.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.51.attn.c_attn.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.51.attn.c_proj.bias": "model-00009-of-00009.safetensors",
+    "transformer.h.51.attn.c_proj.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.51.ln_1.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.51.ln_1.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.51.ln_2.bias": "model-00009-of-00009.safetensors",
+    "transformer.h.51.ln_2.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.51.mlp.c_fc.bias": "model-00009-of-00009.safetensors",
+    "transformer.h.51.mlp.c_fc.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.51.mlp.c_proj.bias": "model-00009-of-00009.safetensors",
+    "transformer.h.51.mlp.c_proj.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.6.attn.c_attn.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.6.attn.c_attn.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.6.attn.c_proj.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.6.attn.c_proj.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.6.ln_1.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.6.ln_1.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.6.ln_2.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.6.ln_2.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.6.mlp.c_fc.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.6.mlp.c_fc.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.6.mlp.c_proj.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.6.mlp.c_proj.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.7.attn.c_attn.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.7.attn.c_attn.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.7.attn.c_proj.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.7.attn.c_proj.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.7.ln_1.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.7.ln_1.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.7.ln_2.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.7.ln_2.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.7.mlp.c_fc.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.7.mlp.c_fc.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.7.mlp.c_proj.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.7.mlp.c_proj.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.8.attn.c_attn.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.8.attn.c_attn.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.8.attn.c_proj.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.8.attn.c_proj.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.8.ln_1.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.8.ln_1.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.8.ln_2.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.8.ln_2.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.8.mlp.c_fc.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.8.mlp.c_fc.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.8.mlp.c_proj.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.8.mlp.c_proj.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.9.attn.c_attn.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.9.attn.c_attn.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.9.attn.c_proj.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.9.attn.c_proj.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.9.ln_1.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.9.ln_1.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.9.ln_2.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.9.ln_2.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.9.mlp.c_fc.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.9.mlp.c_fc.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.9.mlp.c_proj.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.9.mlp.c_proj.weight": "model-00002-of-00009.safetensors",
+    "transformer.ln_f.bias": "model-00009-of-00009.safetensors",
+    "transformer.ln_f.weight": "model-00009-of-00009.safetensors",
+    "transformer.wpe.weight": "model-00001-of-00009.safetensors",
+    "transformer.wte.weight": "model-00001-of-00009.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,51 @@
+{
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "<fim_prefix>",
+    "<fim_middle>",
+    "<fim_suffix>",
+    "<fim_pad>",
+    "<filename>",
+    "<gh_stars>",
+    "<issue_start>",
+    "<issue_comment>",
+    "<issue_closed>",
+    "<jupyter_start>",
+    "<jupyter_text>",
+    "<jupyter_code>",
+    "<jupyter_output>",
+    "<empty_output>",
+    "<commit_before>",
+    "<commit_msg>",
+    "<commit_after>",
+    "<reponame>"
+  ],
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,187 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<fim_prefix>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<fim_middle>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<fim_suffix>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<fim_pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<filename>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "6": {
+      "content": "<gh_stars>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "7": {
+      "content": "<issue_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "8": {
+      "content": "<issue_comment>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "9": {
+      "content": "<issue_closed>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "10": {
+      "content": "<jupyter_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "11": {
+      "content": "<jupyter_text>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "12": {
+      "content": "<jupyter_code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "13": {
+      "content": "<jupyter_output>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "14": {
+      "content": "<empty_output>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "15": {
+      "content": "<commit_before>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "16": {
+      "content": "<commit_msg>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "17": {
+      "content": "<commit_after>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "18": {
+      "content": "<reponame>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "<fim_prefix>",
+    "<fim_middle>",
+    "<fim_suffix>",
+    "<fim_pad>",
+    "<filename>",
+    "<gh_stars>",
+    "<issue_start>",
+    "<issue_comment>",
+    "<issue_closed>",
+    "<jupyter_start>",
+    "<jupyter_text>",
+    "<jupyter_code>",
+    "<jupyter_output>",
+    "<empty_output>",
+    "<commit_before>",
+    "<commit_msg>",
+    "<commit_after>",
+    "<reponame>"
+  ],
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
+  "model_max_length": 8192,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>",
+  "vocab_size": 49152
+}