Upload model

README.md

@@ -1,16 +1,16 @@
 ---
+base_model: bigcode/starcoderbase-7b
+library_name: transformers
+license: other
 tags:
 - autotrain
 - text-generation-inference
 - text-generation
 - peft
-library_name: transformers
-base_model: bigcode/starcoderbase-7b
 widget:
-  - messages:
-      - role: user
-        content: What is your favorite condiment?
-license: other
+- messages:
+  - role: user
+    content: What is your favorite condiment?
 ---
 
 # Model Trained Using AutoTrain

config.json

@@ -0,0 +1,39 @@
+{
+  "_name_or_path": "bigcode/starcoderbase-7b",
+  "activation_function": "gelu",
+  "architectures": [
+    "GPTBigCodeModel"
+  ],
+  "attention_softmax_in_fp32": true,
+  "attn_pdrop": 0.1,
+  "bos_token_id": 0,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 0,
+  "inference_runner": 0,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "max_batch_size": null,
+  "max_sequence_length": null,
+  "model_type": "gpt_bigcode",
+  "multi_query": true,
+  "n_embd": 4096,
+  "n_head": 32,
+  "n_inner": 16384,
+  "n_layer": 42,
+  "n_positions": 8192,
+  "pad_key_length": true,
+  "pre_allocate_kv_cache": false,
+  "resid_pdrop": 0.1,
+  "scale_attention_softmax_in_fp32": true,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.45.0",
+  "use_cache": true,
+  "validate_runner_input": true,
+  "vocab_size": 49152
+}

model-00001-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7dacc5c96a3b8907ab1e86eaca4b9f87c2f53dcbf914058923778f7e5929b8e0
+size 4992349040

model-00002-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3984792dfc11c963996d83545b544156dbd6ea2aefc7a82ba3a288b4eb1b5ad
+size 4866736328

model-00003-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:756bebeafde7ea9ea626b471d62066d303b2fba98ab52846ab8bb00595937e73
+size 4996758416

model-00004-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1188b5651a683c975399b3d23ee4de7a506316063dc0b272203fb03020558220
+size 4996742208

model-00005-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6eb7e1aa7772f540044166957397121206ff85d54d68b2a026fd6054a9ed3a9
+size 4866736376

model-00006-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cabdaf39a0504940700ee97b00d416e6d2341d52899955551a6d41a0a3dfb5a8
+size 4589778064

model.safetensors.index.json

@@ -0,0 +1,515 @@
+{
+  "metadata": {
+    "total_size": 29309052928
+  },
+  "weight_map": {
+    "h.0.attn.c_attn.bias": "model-00001-of-00006.safetensors",
+    "h.0.attn.c_attn.weight": "model-00001-of-00006.safetensors",
+    "h.0.attn.c_proj.bias": "model-00001-of-00006.safetensors",
+    "h.0.attn.c_proj.weight": "model-00001-of-00006.safetensors",
+    "h.0.ln_1.bias": "model-00001-of-00006.safetensors",
+    "h.0.ln_1.weight": "model-00001-of-00006.safetensors",
+    "h.0.ln_2.bias": "model-00001-of-00006.safetensors",
+    "h.0.ln_2.weight": "model-00001-of-00006.safetensors",
+    "h.0.mlp.c_fc.bias": "model-00001-of-00006.safetensors",
+    "h.0.mlp.c_fc.weight": "model-00001-of-00006.safetensors",
+    "h.0.mlp.c_proj.bias": "model-00001-of-00006.safetensors",
+    "h.0.mlp.c_proj.weight": "model-00001-of-00006.safetensors",
+    "h.1.attn.c_attn.bias": "model-00001-of-00006.safetensors",
+    "h.1.attn.c_attn.weight": "model-00001-of-00006.safetensors",
+    "h.1.attn.c_proj.bias": "model-00001-of-00006.safetensors",
+    "h.1.attn.c_proj.weight": "model-00001-of-00006.safetensors",
+    "h.1.ln_1.bias": "model-00001-of-00006.safetensors",
+    "h.1.ln_1.weight": "model-00001-of-00006.safetensors",
+    "h.1.ln_2.bias": "model-00001-of-00006.safetensors",
+    "h.1.ln_2.weight": "model-00001-of-00006.safetensors",
+    "h.1.mlp.c_fc.bias": "model-00001-of-00006.safetensors",
+    "h.1.mlp.c_fc.weight": "model-00001-of-00006.safetensors",
+    "h.1.mlp.c_proj.bias": "model-00001-of-00006.safetensors",
+    "h.1.mlp.c_proj.weight": "model-00001-of-00006.safetensors",
+    "h.10.attn.c_attn.bias": "model-00002-of-00006.safetensors",
+    "h.10.attn.c_attn.weight": "model-00002-of-00006.safetensors",
+    "h.10.attn.c_proj.bias": "model-00002-of-00006.safetensors",
+    "h.10.attn.c_proj.weight": "model-00002-of-00006.safetensors",
+    "h.10.ln_1.bias": "model-00002-of-00006.safetensors",
+    "h.10.ln_1.weight": "model-00002-of-00006.safetensors",
+    "h.10.ln_2.bias": "model-00002-of-00006.safetensors",
+    "h.10.ln_2.weight": "model-00002-of-00006.safetensors",
+    "h.10.mlp.c_fc.bias": "model-00002-of-00006.safetensors",
+    "h.10.mlp.c_fc.weight": "model-00002-of-00006.safetensors",
+    "h.10.mlp.c_proj.bias": "model-00002-of-00006.safetensors",
+    "h.10.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
+    "h.11.attn.c_attn.bias": "model-00002-of-00006.safetensors",
+    "h.11.attn.c_attn.weight": "model-00002-of-00006.safetensors",
+    "h.11.attn.c_proj.bias": "model-00002-of-00006.safetensors",
+    "h.11.attn.c_proj.weight": "model-00002-of-00006.safetensors",
+    "h.11.ln_1.bias": "model-00002-of-00006.safetensors",
+    "h.11.ln_1.weight": "model-00002-of-00006.safetensors",
+    "h.11.ln_2.bias": "model-00002-of-00006.safetensors",
+    "h.11.ln_2.weight": "model-00002-of-00006.safetensors",
+    "h.11.mlp.c_fc.bias": "model-00002-of-00006.safetensors",
+    "h.11.mlp.c_fc.weight": "model-00002-of-00006.safetensors",
+    "h.11.mlp.c_proj.bias": "model-00002-of-00006.safetensors",
+    "h.11.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
+    "h.12.attn.c_attn.bias": "model-00002-of-00006.safetensors",
+    "h.12.attn.c_attn.weight": "model-00002-of-00006.safetensors",
+    "h.12.attn.c_proj.bias": "model-00002-of-00006.safetensors",
+    "h.12.attn.c_proj.weight": "model-00002-of-00006.safetensors",
+    "h.12.ln_1.bias": "model-00002-of-00006.safetensors",
+    "h.12.ln_1.weight": "model-00002-of-00006.safetensors",
+    "h.12.ln_2.bias": "model-00002-of-00006.safetensors",
+    "h.12.ln_2.weight": "model-00002-of-00006.safetensors",
+    "h.12.mlp.c_fc.bias": "model-00002-of-00006.safetensors",
+    "h.12.mlp.c_fc.weight": "model-00002-of-00006.safetensors",
+    "h.12.mlp.c_proj.bias": "model-00002-of-00006.safetensors",
+    "h.12.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
+    "h.13.attn.c_attn.bias": "model-00002-of-00006.safetensors",
+    "h.13.attn.c_attn.weight": "model-00002-of-00006.safetensors",
+    "h.13.attn.c_proj.bias": "model-00002-of-00006.safetensors",
+    "h.13.attn.c_proj.weight": "model-00002-of-00006.safetensors",
+    "h.13.ln_1.bias": "model-00002-of-00006.safetensors",
+    "h.13.ln_1.weight": "model-00002-of-00006.safetensors",
+    "h.13.ln_2.bias": "model-00002-of-00006.safetensors",
+    "h.13.ln_2.weight": "model-00002-of-00006.safetensors",
+    "h.13.mlp.c_fc.bias": "model-00003-of-00006.safetensors",
+    "h.13.mlp.c_fc.weight": "model-00003-of-00006.safetensors",
+    "h.13.mlp.c_proj.bias": "model-00003-of-00006.safetensors",
+    "h.13.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
+    "h.14.attn.c_attn.bias": "model-00003-of-00006.safetensors",
+    "h.14.attn.c_attn.weight": "model-00003-of-00006.safetensors",
+    "h.14.attn.c_proj.bias": "model-00003-of-00006.safetensors",
+    "h.14.attn.c_proj.weight": "model-00003-of-00006.safetensors",
+    "h.14.ln_1.bias": "model-00003-of-00006.safetensors",
+    "h.14.ln_1.weight": "model-00003-of-00006.safetensors",
+    "h.14.ln_2.bias": "model-00003-of-00006.safetensors",
+    "h.14.ln_2.weight": "model-00003-of-00006.safetensors",
+    "h.14.mlp.c_fc.bias": "model-00003-of-00006.safetensors",
+    "h.14.mlp.c_fc.weight": "model-00003-of-00006.safetensors",
+    "h.14.mlp.c_proj.bias": "model-00003-of-00006.safetensors",
+    "h.14.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
+    "h.15.attn.c_attn.bias": "model-00003-of-00006.safetensors",
+    "h.15.attn.c_attn.weight": "model-00003-of-00006.safetensors",
+    "h.15.attn.c_proj.bias": "model-00003-of-00006.safetensors",
+    "h.15.attn.c_proj.weight": "model-00003-of-00006.safetensors",
+    "h.15.ln_1.bias": "model-00003-of-00006.safetensors",
+    "h.15.ln_1.weight": "model-00003-of-00006.safetensors",
+    "h.15.ln_2.bias": "model-00003-of-00006.safetensors",
+    "h.15.ln_2.weight": "model-00003-of-00006.safetensors",
+    "h.15.mlp.c_fc.bias": "model-00003-of-00006.safetensors",
+    "h.15.mlp.c_fc.weight": "model-00003-of-00006.safetensors",
+    "h.15.mlp.c_proj.bias": "model-00003-of-00006.safetensors",
+    "h.15.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
+    "h.16.attn.c_attn.bias": "model-00003-of-00006.safetensors",
+    "h.16.attn.c_attn.weight": "model-00003-of-00006.safetensors",
+    "h.16.attn.c_proj.bias": "model-00003-of-00006.safetensors",
+    "h.16.attn.c_proj.weight": "model-00003-of-00006.safetensors",
+    "h.16.ln_1.bias": "model-00003-of-00006.safetensors",
+    "h.16.ln_1.weight": "model-00003-of-00006.safetensors",
+    "h.16.ln_2.bias": "model-00003-of-00006.safetensors",
+    "h.16.ln_2.weight": "model-00003-of-00006.safetensors",
+    "h.16.mlp.c_fc.bias": "model-00003-of-00006.safetensors",
+    "h.16.mlp.c_fc.weight": "model-00003-of-00006.safetensors",
+    "h.16.mlp.c_proj.bias": "model-00003-of-00006.safetensors",
+    "h.16.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
+    "h.17.attn.c_attn.bias": "model-00003-of-00006.safetensors",
+    "h.17.attn.c_attn.weight": "model-00003-of-00006.safetensors",
+    "h.17.attn.c_proj.bias": "model-00003-of-00006.safetensors",
+    "h.17.attn.c_proj.weight": "model-00003-of-00006.safetensors",
+    "h.17.ln_1.bias": "model-00003-of-00006.safetensors",
+    "h.17.ln_1.weight": "model-00003-of-00006.safetensors",
+    "h.17.ln_2.bias": "model-00003-of-00006.safetensors",
+    "h.17.ln_2.weight": "model-00003-of-00006.safetensors",
+    "h.17.mlp.c_fc.bias": "model-00003-of-00006.safetensors",
+    "h.17.mlp.c_fc.weight": "model-00003-of-00006.safetensors",
+    "h.17.mlp.c_proj.bias": "model-00003-of-00006.safetensors",
+    "h.17.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
+    "h.18.attn.c_attn.bias": "model-00003-of-00006.safetensors",
+    "h.18.attn.c_attn.weight": "model-00003-of-00006.safetensors",
+    "h.18.attn.c_proj.bias": "model-00003-of-00006.safetensors",
+    "h.18.attn.c_proj.weight": "model-00003-of-00006.safetensors",
+    "h.18.ln_1.bias": "model-00003-of-00006.safetensors",
+    "h.18.ln_1.weight": "model-00003-of-00006.safetensors",
+    "h.18.ln_2.bias": "model-00003-of-00006.safetensors",
+    "h.18.ln_2.weight": "model-00003-of-00006.safetensors",
+    "h.18.mlp.c_fc.bias": "model-00003-of-00006.safetensors",
+    "h.18.mlp.c_fc.weight": "model-00003-of-00006.safetensors",
+    "h.18.mlp.c_proj.bias": "model-00003-of-00006.safetensors",
+    "h.18.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
+    "h.19.attn.c_attn.bias": "model-00003-of-00006.safetensors",
+    "h.19.attn.c_attn.weight": "model-00003-of-00006.safetensors",
+    "h.19.attn.c_proj.bias": "model-00003-of-00006.safetensors",
+    "h.19.attn.c_proj.weight": "model-00003-of-00006.safetensors",
+    "h.19.ln_1.bias": "model-00003-of-00006.safetensors",
+    "h.19.ln_1.weight": "model-00003-of-00006.safetensors",
+    "h.19.ln_2.bias": "model-00003-of-00006.safetensors",
+    "h.19.ln_2.weight": "model-00003-of-00006.safetensors",
+    "h.19.mlp.c_fc.bias": "model-00003-of-00006.safetensors",
+    "h.19.mlp.c_fc.weight": "model-00003-of-00006.safetensors",
+    "h.19.mlp.c_proj.bias": "model-00003-of-00006.safetensors",
+    "h.19.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
+    "h.2.attn.c_attn.bias": "model-00001-of-00006.safetensors",
+    "h.2.attn.c_attn.weight": "model-00001-of-00006.safetensors",
+    "h.2.attn.c_proj.bias": "model-00001-of-00006.safetensors",
+    "h.2.attn.c_proj.weight": "model-00001-of-00006.safetensors",
+    "h.2.ln_1.bias": "model-00001-of-00006.safetensors",
+    "h.2.ln_1.weight": "model-00001-of-00006.safetensors",
+    "h.2.ln_2.bias": "model-00001-of-00006.safetensors",
+    "h.2.ln_2.weight": "model-00001-of-00006.safetensors",
+    "h.2.mlp.c_fc.bias": "model-00001-of-00006.safetensors",
+    "h.2.mlp.c_fc.weight": "model-00001-of-00006.safetensors",
+    "h.2.mlp.c_proj.bias": "model-00001-of-00006.safetensors",
+    "h.2.mlp.c_proj.weight": "model-00001-of-00006.safetensors",
+    "h.20.attn.c_attn.bias": "model-00003-of-00006.safetensors",
+    "h.20.attn.c_attn.weight": "model-00003-of-00006.safetensors",
+    "h.20.attn.c_proj.bias": "model-00003-of-00006.safetensors",
+    "h.20.attn.c_proj.weight": "model-00003-of-00006.safetensors",
+    "h.20.ln_1.bias": "model-00003-of-00006.safetensors",
+    "h.20.ln_1.weight": "model-00003-of-00006.safetensors",
+    "h.20.ln_2.bias": "model-00003-of-00006.safetensors",
+    "h.20.ln_2.weight": "model-00003-of-00006.safetensors",
+    "h.20.mlp.c_fc.bias": "model-00003-of-00006.safetensors",
+    "h.20.mlp.c_fc.weight": "model-00003-of-00006.safetensors",
+    "h.20.mlp.c_proj.bias": "model-00004-of-00006.safetensors",
+    "h.20.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
+    "h.21.attn.c_attn.bias": "model-00004-of-00006.safetensors",
+    "h.21.attn.c_attn.weight": "model-00004-of-00006.safetensors",
+    "h.21.attn.c_proj.bias": "model-00004-of-00006.safetensors",
+    "h.21.attn.c_proj.weight": "model-00004-of-00006.safetensors",
+    "h.21.ln_1.bias": "model-00004-of-00006.safetensors",
+    "h.21.ln_1.weight": "model-00004-of-00006.safetensors",
+    "h.21.ln_2.bias": "model-00004-of-00006.safetensors",
+    "h.21.ln_2.weight": "model-00004-of-00006.safetensors",
+    "h.21.mlp.c_fc.bias": "model-00004-of-00006.safetensors",
+    "h.21.mlp.c_fc.weight": "model-00004-of-00006.safetensors",
+    "h.21.mlp.c_proj.bias": "model-00004-of-00006.safetensors",
+    "h.21.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
+    "h.22.attn.c_attn.bias": "model-00004-of-00006.safetensors",
+    "h.22.attn.c_attn.weight": "model-00004-of-00006.safetensors",
+    "h.22.attn.c_proj.bias": "model-00004-of-00006.safetensors",
+    "h.22.attn.c_proj.weight": "model-00004-of-00006.safetensors",
+    "h.22.ln_1.bias": "model-00004-of-00006.safetensors",
+    "h.22.ln_1.weight": "model-00004-of-00006.safetensors",
+    "h.22.ln_2.bias": "model-00004-of-00006.safetensors",
+    "h.22.ln_2.weight": "model-00004-of-00006.safetensors",
+    "h.22.mlp.c_fc.bias": "model-00004-of-00006.safetensors",
+    "h.22.mlp.c_fc.weight": "model-00004-of-00006.safetensors",
+    "h.22.mlp.c_proj.bias": "model-00004-of-00006.safetensors",
+    "h.22.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
+    "h.23.attn.c_attn.bias": "model-00004-of-00006.safetensors",
+    "h.23.attn.c_attn.weight": "model-00004-of-00006.safetensors",
+    "h.23.attn.c_proj.bias": "model-00004-of-00006.safetensors",
+    "h.23.attn.c_proj.weight": "model-00004-of-00006.safetensors",
+    "h.23.ln_1.bias": "model-00004-of-00006.safetensors",
+    "h.23.ln_1.weight": "model-00004-of-00006.safetensors",
+    "h.23.ln_2.bias": "model-00004-of-00006.safetensors",
+    "h.23.ln_2.weight": "model-00004-of-00006.safetensors",
+    "h.23.mlp.c_fc.bias": "model-00004-of-00006.safetensors",
+    "h.23.mlp.c_fc.weight": "model-00004-of-00006.safetensors",
+    "h.23.mlp.c_proj.bias": "model-00004-of-00006.safetensors",
+    "h.23.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
+    "h.24.attn.c_attn.bias": "model-00004-of-00006.safetensors",
+    "h.24.attn.c_attn.weight": "model-00004-of-00006.safetensors",
+    "h.24.attn.c_proj.bias": "model-00004-of-00006.safetensors",
+    "h.24.attn.c_proj.weight": "model-00004-of-00006.safetensors",
+    "h.24.ln_1.bias": "model-00004-of-00006.safetensors",
+    "h.24.ln_1.weight": "model-00004-of-00006.safetensors",
+    "h.24.ln_2.bias": "model-00004-of-00006.safetensors",
+    "h.24.ln_2.weight": "model-00004-of-00006.safetensors",
+    "h.24.mlp.c_fc.bias": "model-00004-of-00006.safetensors",
+    "h.24.mlp.c_fc.weight": "model-00004-of-00006.safetensors",
+    "h.24.mlp.c_proj.bias": "model-00004-of-00006.safetensors",
+    "h.24.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
+    "h.25.attn.c_attn.bias": "model-00004-of-00006.safetensors",
+    "h.25.attn.c_attn.weight": "model-00004-of-00006.safetensors",
+    "h.25.attn.c_proj.bias": "model-00004-of-00006.safetensors",
+    "h.25.attn.c_proj.weight": "model-00004-of-00006.safetensors",
+    "h.25.ln_1.bias": "model-00004-of-00006.safetensors",
+    "h.25.ln_1.weight": "model-00004-of-00006.safetensors",
+    "h.25.ln_2.bias": "model-00004-of-00006.safetensors",
+    "h.25.ln_2.weight": "model-00004-of-00006.safetensors",
+    "h.25.mlp.c_fc.bias": "model-00004-of-00006.safetensors",
+    "h.25.mlp.c_fc.weight": "model-00004-of-00006.safetensors",
+    "h.25.mlp.c_proj.bias": "model-00004-of-00006.safetensors",
+    "h.25.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
+    "h.26.attn.c_attn.bias": "model-00004-of-00006.safetensors",
+    "h.26.attn.c_attn.weight": "model-00004-of-00006.safetensors",
+    "h.26.attn.c_proj.bias": "model-00004-of-00006.safetensors",
+    "h.26.attn.c_proj.weight": "model-00004-of-00006.safetensors",
+    "h.26.ln_1.bias": "model-00004-of-00006.safetensors",
+    "h.26.ln_1.weight": "model-00004-of-00006.safetensors",
+    "h.26.ln_2.bias": "model-00004-of-00006.safetensors",
+    "h.26.ln_2.weight": "model-00004-of-00006.safetensors",
+    "h.26.mlp.c_fc.bias": "model-00004-of-00006.safetensors",
+    "h.26.mlp.c_fc.weight": "model-00004-of-00006.safetensors",
+    "h.26.mlp.c_proj.bias": "model-00004-of-00006.safetensors",
+    "h.26.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
+    "h.27.attn.c_attn.bias": "model-00004-of-00006.safetensors",
+    "h.27.attn.c_attn.weight": "model-00004-of-00006.safetensors",
+    "h.27.attn.c_proj.bias": "model-00004-of-00006.safetensors",
+    "h.27.attn.c_proj.weight": "model-00004-of-00006.safetensors",
+    "h.27.ln_1.bias": "model-00004-of-00006.safetensors",
+    "h.27.ln_1.weight": "model-00004-of-00006.safetensors",
+    "h.27.ln_2.bias": "model-00004-of-00006.safetensors",
+    "h.27.ln_2.weight": "model-00004-of-00006.safetensors",
+    "h.27.mlp.c_fc.bias": "model-00004-of-00006.safetensors",
+    "h.27.mlp.c_fc.weight": "model-00004-of-00006.safetensors",
+    "h.27.mlp.c_proj.bias": "model-00004-of-00006.safetensors",
+    "h.27.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
+    "h.28.attn.c_attn.bias": "model-00005-of-00006.safetensors",
+    "h.28.attn.c_attn.weight": "model-00005-of-00006.safetensors",
+    "h.28.attn.c_proj.bias": "model-00005-of-00006.safetensors",
+    "h.28.attn.c_proj.weight": "model-00005-of-00006.safetensors",
+    "h.28.ln_1.bias": "model-00004-of-00006.safetensors",
+    "h.28.ln_1.weight": "model-00004-of-00006.safetensors",
+    "h.28.ln_2.bias": "model-00005-of-00006.safetensors",
+    "h.28.ln_2.weight": "model-00005-of-00006.safetensors",
+    "h.28.mlp.c_fc.bias": "model-00005-of-00006.safetensors",
+    "h.28.mlp.c_fc.weight": "model-00005-of-00006.safetensors",
+    "h.28.mlp.c_proj.bias": "model-00005-of-00006.safetensors",
+    "h.28.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
+    "h.29.attn.c_attn.bias": "model-00005-of-00006.safetensors",
+    "h.29.attn.c_attn.weight": "model-00005-of-00006.safetensors",
+    "h.29.attn.c_proj.bias": "model-00005-of-00006.safetensors",
+    "h.29.attn.c_proj.weight": "model-00005-of-00006.safetensors",
+    "h.29.ln_1.bias": "model-00005-of-00006.safetensors",
+    "h.29.ln_1.weight": "model-00005-of-00006.safetensors",
+    "h.29.ln_2.bias": "model-00005-of-00006.safetensors",
+    "h.29.ln_2.weight": "model-00005-of-00006.safetensors",
+    "h.29.mlp.c_fc.bias": "model-00005-of-00006.safetensors",
+    "h.29.mlp.c_fc.weight": "model-00005-of-00006.safetensors",
+    "h.29.mlp.c_proj.bias": "model-00005-of-00006.safetensors",
+    "h.29.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
+    "h.3.attn.c_attn.bias": "model-00001-of-00006.safetensors",
+    "h.3.attn.c_attn.weight": "model-00001-of-00006.safetensors",
+    "h.3.attn.c_proj.bias": "model-00001-of-00006.safetensors",
+    "h.3.attn.c_proj.weight": "model-00001-of-00006.safetensors",
+    "h.3.ln_1.bias": "model-00001-of-00006.safetensors",
+    "h.3.ln_1.weight": "model-00001-of-00006.safetensors",
+    "h.3.ln_2.bias": "model-00001-of-00006.safetensors",
+    "h.3.ln_2.weight": "model-00001-of-00006.safetensors",
+    "h.3.mlp.c_fc.bias": "model-00001-of-00006.safetensors",
+    "h.3.mlp.c_fc.weight": "model-00001-of-00006.safetensors",
+    "h.3.mlp.c_proj.bias": "model-00001-of-00006.safetensors",
+    "h.3.mlp.c_proj.weight": "model-00001-of-00006.safetensors",
+    "h.30.attn.c_attn.bias": "model-00005-of-00006.safetensors",
+    "h.30.attn.c_attn.weight": "model-00005-of-00006.safetensors",
+    "h.30.attn.c_proj.bias": "model-00005-of-00006.safetensors",
+    "h.30.attn.c_proj.weight": "model-00005-of-00006.safetensors",
+    "h.30.ln_1.bias": "model-00005-of-00006.safetensors",
+    "h.30.ln_1.weight": "model-00005-of-00006.safetensors",
+    "h.30.ln_2.bias": "model-00005-of-00006.safetensors",
+    "h.30.ln_2.weight": "model-00005-of-00006.safetensors",
+    "h.30.mlp.c_fc.bias": "model-00005-of-00006.safetensors",
+    "h.30.mlp.c_fc.weight": "model-00005-of-00006.safetensors",
+    "h.30.mlp.c_proj.bias": "model-00005-of-00006.safetensors",
+    "h.30.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
+    "h.31.attn.c_attn.bias": "model-00005-of-00006.safetensors",
+    "h.31.attn.c_attn.weight": "model-00005-of-00006.safetensors",
+    "h.31.attn.c_proj.bias": "model-00005-of-00006.safetensors",
+    "h.31.attn.c_proj.weight": "model-00005-of-00006.safetensors",
+    "h.31.ln_1.bias": "model-00005-of-00006.safetensors",
+    "h.31.ln_1.weight": "model-00005-of-00006.safetensors",
+    "h.31.ln_2.bias": "model-00005-of-00006.safetensors",
+    "h.31.ln_2.weight": "model-00005-of-00006.safetensors",
+    "h.31.mlp.c_fc.bias": "model-00005-of-00006.safetensors",
+    "h.31.mlp.c_fc.weight": "model-00005-of-00006.safetensors",
+    "h.31.mlp.c_proj.bias": "model-00005-of-00006.safetensors",
+    "h.31.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
+    "h.32.attn.c_attn.bias": "model-00005-of-00006.safetensors",
+    "h.32.attn.c_attn.weight": "model-00005-of-00006.safetensors",
+    "h.32.attn.c_proj.bias": "model-00005-of-00006.safetensors",
+    "h.32.attn.c_proj.weight": "model-00005-of-00006.safetensors",
+    "h.32.ln_1.bias": "model-00005-of-00006.safetensors",
+    "h.32.ln_1.weight": "model-00005-of-00006.safetensors",
+    "h.32.ln_2.bias": "model-00005-of-00006.safetensors",
+    "h.32.ln_2.weight": "model-00005-of-00006.safetensors",
+    "h.32.mlp.c_fc.bias": "model-00005-of-00006.safetensors",
+    "h.32.mlp.c_fc.weight": "model-00005-of-00006.safetensors",
+    "h.32.mlp.c_proj.bias": "model-00005-of-00006.safetensors",
+    "h.32.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
+    "h.33.attn.c_attn.bias": "model-00005-of-00006.safetensors",
+    "h.33.attn.c_attn.weight": "model-00005-of-00006.safetensors",
+    "h.33.attn.c_proj.bias": "model-00005-of-00006.safetensors",
+    "h.33.attn.c_proj.weight": "model-00005-of-00006.safetensors",
+    "h.33.ln_1.bias": "model-00005-of-00006.safetensors",
+    "h.33.ln_1.weight": "model-00005-of-00006.safetensors",
+    "h.33.ln_2.bias": "model-00005-of-00006.safetensors",
+    "h.33.ln_2.weight": "model-00005-of-00006.safetensors",
+    "h.33.mlp.c_fc.bias": "model-00005-of-00006.safetensors",
+    "h.33.mlp.c_fc.weight": "model-00005-of-00006.safetensors",
+    "h.33.mlp.c_proj.bias": "model-00005-of-00006.safetensors",
+    "h.33.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
+    "h.34.attn.c_attn.bias": "model-00005-of-00006.safetensors",
+    "h.34.attn.c_attn.weight": "model-00005-of-00006.safetensors",
+    "h.34.attn.c_proj.bias": "model-00005-of-00006.safetensors",
+    "h.34.attn.c_proj.weight": "model-00005-of-00006.safetensors",
+    "h.34.ln_1.bias": "model-00005-of-00006.safetensors",
+    "h.34.ln_1.weight": "model-00005-of-00006.safetensors",
+    "h.34.ln_2.bias": "model-00005-of-00006.safetensors",
+    "h.34.ln_2.weight": "model-00005-of-00006.safetensors",
+    "h.34.mlp.c_fc.bias": "model-00005-of-00006.safetensors",
+    "h.34.mlp.c_fc.weight": "model-00005-of-00006.safetensors",
+    "h.34.mlp.c_proj.bias": "model-00005-of-00006.safetensors",
+    "h.34.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
+    "h.35.attn.c_attn.bias": "model-00005-of-00006.safetensors",
+    "h.35.attn.c_attn.weight": "model-00005-of-00006.safetensors",
+    "h.35.attn.c_proj.bias": "model-00005-of-00006.safetensors",
+    "h.35.attn.c_proj.weight": "model-00005-of-00006.safetensors",
+    "h.35.ln_1.bias": "model-00005-of-00006.safetensors",
+    "h.35.ln_1.weight": "model-00005-of-00006.safetensors",
+    "h.35.ln_2.bias": "model-00005-of-00006.safetensors",
+    "h.35.ln_2.weight": "model-00005-of-00006.safetensors",
+    "h.35.mlp.c_fc.bias": "model-00006-of-00006.safetensors",
+    "h.35.mlp.c_fc.weight": "model-00006-of-00006.safetensors",
+    "h.35.mlp.c_proj.bias": "model-00006-of-00006.safetensors",
+    "h.35.mlp.c_proj.weight": "model-00006-of-00006.safetensors",
+    "h.36.attn.c_attn.bias": "model-00006-of-00006.safetensors",
+    "h.36.attn.c_attn.weight": "model-00006-of-00006.safetensors",
+    "h.36.attn.c_proj.bias": "model-00006-of-00006.safetensors",
+    "h.36.attn.c_proj.weight": "model-00006-of-00006.safetensors",
+    "h.36.ln_1.bias": "model-00006-of-00006.safetensors",
+    "h.36.ln_1.weight": "model-00006-of-00006.safetensors",
+    "h.36.ln_2.bias": "model-00006-of-00006.safetensors",
+    "h.36.ln_2.weight": "model-00006-of-00006.safetensors",
+    "h.36.mlp.c_fc.bias": "model-00006-of-00006.safetensors",
+    "h.36.mlp.c_fc.weight": "model-00006-of-00006.safetensors",
+    "h.36.mlp.c_proj.bias": "model-00006-of-00006.safetensors",
+    "h.36.mlp.c_proj.weight": "model-00006-of-00006.safetensors",
+    "h.37.attn.c_attn.bias": "model-00006-of-00006.safetensors",
+    "h.37.attn.c_attn.weight": "model-00006-of-00006.safetensors",
+    "h.37.attn.c_proj.bias": "model-00006-of-00006.safetensors",
+    "h.37.attn.c_proj.weight": "model-00006-of-00006.safetensors",
+    "h.37.ln_1.bias": "model-00006-of-00006.safetensors",
+    "h.37.ln_1.weight": "model-00006-of-00006.safetensors",
+    "h.37.ln_2.bias": "model-00006-of-00006.safetensors",
+    "h.37.ln_2.weight": "model-00006-of-00006.safetensors",
+    "h.37.mlp.c_fc.bias": "model-00006-of-00006.safetensors",
+    "h.37.mlp.c_fc.weight": "model-00006-of-00006.safetensors",
+    "h.37.mlp.c_proj.bias": "model-00006-of-00006.safetensors",
+    "h.37.mlp.c_proj.weight": "model-00006-of-00006.safetensors",
+    "h.38.attn.c_attn.bias": "model-00006-of-00006.safetensors",
+    "h.38.attn.c_attn.weight": "model-00006-of-00006.safetensors",
+    "h.38.attn.c_proj.bias": "model-00006-of-00006.safetensors",
+    "h.38.attn.c_proj.weight": "model-00006-of-00006.safetensors",
+    "h.38.ln_1.bias": "model-00006-of-00006.safetensors",
+    "h.38.ln_1.weight": "model-00006-of-00006.safetensors",
+    "h.38.ln_2.bias": "model-00006-of-00006.safetensors",
+    "h.38.ln_2.weight": "model-00006-of-00006.safetensors",
+    "h.38.mlp.c_fc.bias": "model-00006-of-00006.safetensors",
+    "h.38.mlp.c_fc.weight": "model-00006-of-00006.safetensors",
+    "h.38.mlp.c_proj.bias": "model-00006-of-00006.safetensors",
+    "h.38.mlp.c_proj.weight": "model-00006-of-00006.safetensors",
+    "h.39.attn.c_attn.bias": "model-00006-of-00006.safetensors",
+    "h.39.attn.c_attn.weight": "model-00006-of-00006.safetensors",
+    "h.39.attn.c_proj.bias": "model-00006-of-00006.safetensors",
+    "h.39.attn.c_proj.weight": "model-00006-of-00006.safetensors",
+    "h.39.ln_1.bias": "model-00006-of-00006.safetensors",
+    "h.39.ln_1.weight": "model-00006-of-00006.safetensors",
+    "h.39.ln_2.bias": "model-00006-of-00006.safetensors",
+    "h.39.ln_2.weight": "model-00006-of-00006.safetensors",
+    "h.39.mlp.c_fc.bias": "model-00006-of-00006.safetensors",
+    "h.39.mlp.c_fc.weight": "model-00006-of-00006.safetensors",
+    "h.39.mlp.c_proj.bias": "model-00006-of-00006.safetensors",
+    "h.39.mlp.c_proj.weight": "model-00006-of-00006.safetensors",
+    "h.4.attn.c_attn.bias": "model-00001-of-00006.safetensors",
+    "h.4.attn.c_attn.weight": "model-00001-of-00006.safetensors",
+    "h.4.attn.c_proj.bias": "model-00001-of-00006.safetensors",
+    "h.4.attn.c_proj.weight": "model-00001-of-00006.safetensors",
+    "h.4.ln_1.bias": "model-00001-of-00006.safetensors",
+    "h.4.ln_1.weight": "model-00001-of-00006.safetensors",
+    "h.4.ln_2.bias": "model-00001-of-00006.safetensors",
+    "h.4.ln_2.weight": "model-00001-of-00006.safetensors",
+    "h.4.mlp.c_fc.bias": "model-00001-of-00006.safetensors",
+    "h.4.mlp.c_fc.weight": "model-00001-of-00006.safetensors",
+    "h.4.mlp.c_proj.bias": "model-00001-of-00006.safetensors",
+    "h.4.mlp.c_proj.weight": "model-00001-of-00006.safetensors",
+    "h.40.attn.c_attn.bias": "model-00006-of-00006.safetensors",
+    "h.40.attn.c_attn.weight": "model-00006-of-00006.safetensors",
+    "h.40.attn.c_proj.bias": "model-00006-of-00006.safetensors",
+    "h.40.attn.c_proj.weight": "model-00006-of-00006.safetensors",
+    "h.40.ln_1.bias": "model-00006-of-00006.safetensors",
+    "h.40.ln_1.weight": "model-00006-of-00006.safetensors",
+    "h.40.ln_2.bias": "model-00006-of-00006.safetensors",
+    "h.40.ln_2.weight": "model-00006-of-00006.safetensors",
+    "h.40.mlp.c_fc.bias": "model-00006-of-00006.safetensors",
+    "h.40.mlp.c_fc.weight": "model-00006-of-00006.safetensors",
+    "h.40.mlp.c_proj.bias": "model-00006-of-00006.safetensors",
+    "h.40.mlp.c_proj.weight": "model-00006-of-00006.safetensors",
+    "h.41.attn.c_attn.bias": "model-00006-of-00006.safetensors",
+    "h.41.attn.c_attn.weight": "model-00006-of-00006.safetensors",
+    "h.41.attn.c_proj.bias": "model-00006-of-00006.safetensors",
+    "h.41.attn.c_proj.weight": "model-00006-of-00006.safetensors",
+    "h.41.ln_1.bias": "model-00006-of-00006.safetensors",
+    "h.41.ln_1.weight": "model-00006-of-00006.safetensors",
+    "h.41.ln_2.bias": "model-00006-of-00006.safetensors",
+    "h.41.ln_2.weight": "model-00006-of-00006.safetensors",
+    "h.41.mlp.c_fc.bias": "model-00006-of-00006.safetensors",
+    "h.41.mlp.c_fc.weight": "model-00006-of-00006.safetensors",
+    "h.41.mlp.c_proj.bias": "model-00006-of-00006.safetensors",
+    "h.41.mlp.c_proj.weight": "model-00006-of-00006.safetensors",
+    "h.5.attn.c_attn.bias": "model-00001-of-00006.safetensors",
+    "h.5.attn.c_attn.weight": "model-00001-of-00006.safetensors",
+    "h.5.attn.c_proj.bias": "model-00001-of-00006.safetensors",
+    "h.5.attn.c_proj.weight": "model-00001-of-00006.safetensors",
+    "h.5.ln_1.bias": "model-00001-of-00006.safetensors",
+    "h.5.ln_1.weight": "model-00001-of-00006.safetensors",
+    "h.5.ln_2.bias": "model-00001-of-00006.safetensors",
+    "h.5.ln_2.weight": "model-00001-of-00006.safetensors",
+    "h.5.mlp.c_fc.bias": "model-00001-of-00006.safetensors",
+    "h.5.mlp.c_fc.weight": "model-00001-of-00006.safetensors",
+    "h.5.mlp.c_proj.bias": "model-00001-of-00006.safetensors",
+    "h.5.mlp.c_proj.weight": "model-00001-of-00006.safetensors",
+    "h.6.attn.c_attn.bias": "model-00002-of-00006.safetensors",
+    "h.6.attn.c_attn.weight": "model-00002-of-00006.safetensors",
+    "h.6.attn.c_proj.bias": "model-00002-of-00006.safetensors",
+    "h.6.attn.c_proj.weight": "model-00002-of-00006.safetensors",
+    "h.6.ln_1.bias": "model-00001-of-00006.safetensors",
+    "h.6.ln_1.weight": "model-00001-of-00006.safetensors",
+    "h.6.ln_2.bias": "model-00002-of-00006.safetensors",
+    "h.6.ln_2.weight": "model-00002-of-00006.safetensors",
+    "h.6.mlp.c_fc.bias": "model-00002-of-00006.safetensors",
+    "h.6.mlp.c_fc.weight": "model-00002-of-00006.safetensors",
+    "h.6.mlp.c_proj.bias": "model-00002-of-00006.safetensors",
+    "h.6.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
+    "h.7.attn.c_attn.bias": "model-00002-of-00006.safetensors",
+    "h.7.attn.c_attn.weight": "model-00002-of-00006.safetensors",
+    "h.7.attn.c_proj.bias": "model-00002-of-00006.safetensors",
+    "h.7.attn.c_proj.weight": "model-00002-of-00006.safetensors",
+    "h.7.ln_1.bias": "model-00002-of-00006.safetensors",
+    "h.7.ln_1.weight": "model-00002-of-00006.safetensors",
+    "h.7.ln_2.bias": "model-00002-of-00006.safetensors",
+    "h.7.ln_2.weight": "model-00002-of-00006.safetensors",
+    "h.7.mlp.c_fc.bias": "model-00002-of-00006.safetensors",
+    "h.7.mlp.c_fc.weight": "model-00002-of-00006.safetensors",
+    "h.7.mlp.c_proj.bias": "model-00002-of-00006.safetensors",
+    "h.7.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
+    "h.8.attn.c_attn.bias": "model-00002-of-00006.safetensors",
+    "h.8.attn.c_attn.weight": "model-00002-of-00006.safetensors",
+    "h.8.attn.c_proj.bias": "model-00002-of-00006.safetensors",
+    "h.8.attn.c_proj.weight": "model-00002-of-00006.safetensors",
+    "h.8.ln_1.bias": "model-00002-of-00006.safetensors",
+    "h.8.ln_1.weight": "model-00002-of-00006.safetensors",
+    "h.8.ln_2.bias": "model-00002-of-00006.safetensors",
+    "h.8.ln_2.weight": "model-00002-of-00006.safetensors",
+    "h.8.mlp.c_fc.bias": "model-00002-of-00006.safetensors",
+    "h.8.mlp.c_fc.weight": "model-00002-of-00006.safetensors",
+    "h.8.mlp.c_proj.bias": "model-00002-of-00006.safetensors",
+    "h.8.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
+    "h.9.attn.c_attn.bias": "model-00002-of-00006.safetensors",
+    "h.9.attn.c_attn.weight": "model-00002-of-00006.safetensors",
+    "h.9.attn.c_proj.bias": "model-00002-of-00006.safetensors",
+    "h.9.attn.c_proj.weight": "model-00002-of-00006.safetensors",
+    "h.9.ln_1.bias": "model-00002-of-00006.safetensors",
+    "h.9.ln_1.weight": "model-00002-of-00006.safetensors",
+    "h.9.ln_2.bias": "model-00002-of-00006.safetensors",
+    "h.9.ln_2.weight": "model-00002-of-00006.safetensors",
+    "h.9.mlp.c_fc.bias": "model-00002-of-00006.safetensors",
+    "h.9.mlp.c_fc.weight": "model-00002-of-00006.safetensors",
+    "h.9.mlp.c_proj.bias": "model-00002-of-00006.safetensors",
+    "h.9.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
+    "ln_f.bias": "model-00006-of-00006.safetensors",
+    "ln_f.weight": "model-00006-of-00006.safetensors",
+    "wpe.weight": "model-00001-of-00006.safetensors",
+    "wte.weight": "model-00001-of-00006.safetensors"
+  }
+}