Upload folder using huggingface_hub

config.json

@@ -0,0 +1,39 @@
+{
+  "_name_or_path": "bigcode/starcoder",
+  "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,
+  "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": 6144,
+  "n_head": 48,
+  "n_inner": 24576,
+  "n_layer": 40,
+  "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": "4.41.1",
+  "use_cache": true,
+  "validate_runner_input": true,
+  "vocab_size": 49152
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "eos_token_id": 0,
+  "transformers_version": "4.41.1"
+}

model-00001-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b1ade7205a265cfec4d089971c5a2a002b4076f3e13a032e87da804ff4d593b
+size 4750744032

model-00002-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97fab2a7c4b984c950718212b99b0fac07a2ef1ad98502447936396d13bdbd4e
+size 4549540568

model-00003-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b9b6d7edc498675764677c3d89983073792c1188fbd9adcc4ee40db85c0d7ad
+size 4549540568

model-00004-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b41cd4cfe59ae9477a006f9d54da1f6de6788a46c854acbb6dfd7536ef1ab4a9
+size 4549540568

model-00005-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf0f4b66784e44f9e04d58ca0676205cd756e1a3e7c7981507e872f3da384202
+size 4549540600

model-00006-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:991dc6d2b9c2b11b378204262fcf0bd4375294bb876ad28d8835e3fe627902f3
+size 4549540600

model-00007-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5af5b949082f87f687b5ac940f6116df93e71bd201b78ca1f93b71af6b791e6
+size 4549540600

model-00008-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c40236244e04ee5d18a07740ec3e38a82c23e57daf8b97c5ffcafb17a05d6d6d
+size 4549540600

model-00009-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:391291d49174ba0d9dae2f2596d8d1731952c4fdcd1e3356a884a989b1908a8b
+size 4549540600

model-00010-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25e5eaf0fd3b2b5c70f99d1c1480e176852d4f644238f8deffd5073056383ec6
+size 4549540600

model-00011-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:65edc67c05d93799778c4f2783749681ffe0640e5c1d988e180fe49d050a34b6
+size 4549540600

model-00012-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f1b9efcc08d65076ba161d465f9999e79191dc2aa28d9d8445ef3cfbf836fb5
+size 4549540600

model-00013-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b17f224195d8a058eb502841f1147a72dc1978a64d5d7fdcc2410999575bb36
+size 4549540600

model-00014-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5eb6bc7a723fa40154df3fe32445e1eff4cb11c3a8fc9cd4e8be03156b166844
+size 2724645752

model.safetensors.index.json

@@ -0,0 +1,491 @@
+{
+  "metadata": {
+    "total_size": 62069825536
+  },
+  "weight_map": {
+    "transformer.h.0.attn.c_attn.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.0.attn.c_attn.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.0.attn.c_proj.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.0.attn.c_proj.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.0.ln_1.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.0.ln_1.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.0.ln_2.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.0.ln_2.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.0.mlp.c_fc.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.0.mlp.c_fc.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.0.mlp.c_proj.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.0.mlp.c_proj.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.1.attn.c_attn.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.1.attn.c_attn.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.1.attn.c_proj.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.1.attn.c_proj.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.1.ln_1.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.1.ln_1.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.1.ln_2.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.1.ln_2.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.1.mlp.c_fc.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.1.mlp.c_fc.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.1.mlp.c_proj.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.1.mlp.c_proj.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.10.attn.c_attn.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.10.attn.c_attn.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.10.attn.c_proj.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.10.attn.c_proj.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.10.ln_1.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.10.ln_1.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.10.ln_2.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.10.ln_2.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.10.mlp.c_fc.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.10.mlp.c_fc.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.10.mlp.c_proj.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.10.mlp.c_proj.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.11.attn.c_attn.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.11.attn.c_attn.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.11.attn.c_proj.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.11.attn.c_proj.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.11.ln_1.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.11.ln_1.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.11.ln_2.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.11.ln_2.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.11.mlp.c_fc.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.11.mlp.c_fc.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.11.mlp.c_proj.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.11.mlp.c_proj.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.12.attn.c_attn.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.12.attn.c_attn.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.12.attn.c_proj.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.12.attn.c_proj.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.12.ln_1.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.12.ln_1.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.12.ln_2.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.12.ln_2.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.12.mlp.c_fc.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.12.mlp.c_fc.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.12.mlp.c_proj.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.12.mlp.c_proj.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.13.attn.c_attn.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.13.attn.c_attn.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.13.attn.c_proj.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.13.attn.c_proj.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.13.ln_1.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.13.ln_1.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.13.ln_2.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.13.ln_2.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.13.mlp.c_fc.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.13.mlp.c_fc.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.13.mlp.c_proj.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.13.mlp.c_proj.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.14.attn.c_attn.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.14.attn.c_attn.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.14.attn.c_proj.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.14.attn.c_proj.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.14.ln_1.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.14.ln_1.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.14.ln_2.bias": "model-00005-of-00014.safetensors",
+    "transformer.h.14.ln_2.weight": "model-00005-of-00014.safetensors",
+    "transformer.h.14.mlp.c_fc.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.14.mlp.c_fc.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.14.mlp.c_proj.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.14.mlp.c_proj.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.15.attn.c_attn.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.15.attn.c_attn.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.15.attn.c_proj.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.15.attn.c_proj.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.15.ln_1.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.15.ln_1.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.15.ln_2.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.15.ln_2.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.15.mlp.c_fc.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.15.mlp.c_fc.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.15.mlp.c_proj.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.15.mlp.c_proj.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.16.attn.c_attn.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.16.attn.c_attn.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.16.attn.c_proj.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.16.attn.c_proj.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.16.ln_1.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.16.ln_1.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.16.ln_2.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.16.ln_2.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.16.mlp.c_fc.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.16.mlp.c_fc.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.16.mlp.c_proj.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.16.mlp.c_proj.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.17.attn.c_attn.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.17.attn.c_attn.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.17.attn.c_proj.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.17.attn.c_proj.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.17.ln_1.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.17.ln_1.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.17.ln_2.bias": "model-00006-of-00014.safetensors",
+    "transformer.h.17.ln_2.weight": "model-00006-of-00014.safetensors",
+    "transformer.h.17.mlp.c_fc.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.17.mlp.c_fc.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.17.mlp.c_proj.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.17.mlp.c_proj.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.18.attn.c_attn.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.18.attn.c_attn.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.18.attn.c_proj.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.18.attn.c_proj.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.18.ln_1.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.18.ln_1.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.18.ln_2.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.18.ln_2.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.18.mlp.c_fc.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.18.mlp.c_fc.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.18.mlp.c_proj.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.18.mlp.c_proj.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.19.attn.c_attn.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.19.attn.c_attn.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.19.attn.c_proj.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.19.attn.c_proj.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.19.ln_1.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.19.ln_1.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.19.ln_2.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.19.ln_2.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.19.mlp.c_fc.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.19.mlp.c_fc.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.19.mlp.c_proj.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.19.mlp.c_proj.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.2.attn.c_attn.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.2.attn.c_attn.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.2.attn.c_proj.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.2.attn.c_proj.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.2.ln_1.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.2.ln_1.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.2.ln_2.bias": "model-00001-of-00014.safetensors",
+    "transformer.h.2.ln_2.weight": "model-00001-of-00014.safetensors",
+    "transformer.h.2.mlp.c_fc.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.2.mlp.c_fc.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.2.mlp.c_proj.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.2.mlp.c_proj.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.20.attn.c_attn.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.20.attn.c_attn.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.20.attn.c_proj.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.20.attn.c_proj.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.20.ln_1.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.20.ln_1.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.20.ln_2.bias": "model-00007-of-00014.safetensors",
+    "transformer.h.20.ln_2.weight": "model-00007-of-00014.safetensors",
+    "transformer.h.20.mlp.c_fc.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.20.mlp.c_fc.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.20.mlp.c_proj.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.20.mlp.c_proj.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.21.attn.c_attn.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.21.attn.c_attn.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.21.attn.c_proj.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.21.attn.c_proj.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.21.ln_1.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.21.ln_1.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.21.ln_2.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.21.ln_2.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.21.mlp.c_fc.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.21.mlp.c_fc.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.21.mlp.c_proj.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.21.mlp.c_proj.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.22.attn.c_attn.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.22.attn.c_attn.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.22.attn.c_proj.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.22.attn.c_proj.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.22.ln_1.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.22.ln_1.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.22.ln_2.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.22.ln_2.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.22.mlp.c_fc.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.22.mlp.c_fc.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.22.mlp.c_proj.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.22.mlp.c_proj.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.23.attn.c_attn.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.23.attn.c_attn.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.23.attn.c_proj.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.23.attn.c_proj.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.23.ln_1.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.23.ln_1.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.23.ln_2.bias": "model-00008-of-00014.safetensors",
+    "transformer.h.23.ln_2.weight": "model-00008-of-00014.safetensors",
+    "transformer.h.23.mlp.c_fc.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.23.mlp.c_fc.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.23.mlp.c_proj.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.23.mlp.c_proj.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.24.attn.c_attn.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.24.attn.c_attn.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.24.attn.c_proj.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.24.attn.c_proj.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.24.ln_1.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.24.ln_1.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.24.ln_2.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.24.ln_2.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.24.mlp.c_fc.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.24.mlp.c_fc.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.24.mlp.c_proj.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.24.mlp.c_proj.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.25.attn.c_attn.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.25.attn.c_attn.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.25.attn.c_proj.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.25.attn.c_proj.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.25.ln_1.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.25.ln_1.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.25.ln_2.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.25.ln_2.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.25.mlp.c_fc.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.25.mlp.c_fc.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.25.mlp.c_proj.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.25.mlp.c_proj.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.26.attn.c_attn.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.26.attn.c_attn.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.26.attn.c_proj.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.26.attn.c_proj.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.26.ln_1.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.26.ln_1.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.26.ln_2.bias": "model-00009-of-00014.safetensors",
+    "transformer.h.26.ln_2.weight": "model-00009-of-00014.safetensors",
+    "transformer.h.26.mlp.c_fc.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.26.mlp.c_fc.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.26.mlp.c_proj.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.26.mlp.c_proj.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.27.attn.c_attn.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.27.attn.c_attn.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.27.attn.c_proj.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.27.attn.c_proj.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.27.ln_1.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.27.ln_1.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.27.ln_2.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.27.ln_2.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.27.mlp.c_fc.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.27.mlp.c_fc.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.27.mlp.c_proj.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.27.mlp.c_proj.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.28.attn.c_attn.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.28.attn.c_attn.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.28.attn.c_proj.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.28.attn.c_proj.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.28.ln_1.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.28.ln_1.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.28.ln_2.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.28.ln_2.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.28.mlp.c_fc.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.28.mlp.c_fc.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.28.mlp.c_proj.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.28.mlp.c_proj.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.29.attn.c_attn.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.29.attn.c_attn.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.29.attn.c_proj.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.29.attn.c_proj.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.29.ln_1.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.29.ln_1.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.29.ln_2.bias": "model-00010-of-00014.safetensors",
+    "transformer.h.29.ln_2.weight": "model-00010-of-00014.safetensors",
+    "transformer.h.29.mlp.c_fc.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.29.mlp.c_fc.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.29.mlp.c_proj.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.29.mlp.c_proj.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.3.attn.c_attn.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.3.attn.c_attn.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.3.attn.c_proj.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.3.attn.c_proj.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.3.ln_1.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.3.ln_1.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.3.ln_2.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.3.ln_2.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.3.mlp.c_fc.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.3.mlp.c_fc.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.3.mlp.c_proj.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.3.mlp.c_proj.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.30.attn.c_attn.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.30.attn.c_attn.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.30.attn.c_proj.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.30.attn.c_proj.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.30.ln_1.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.30.ln_1.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.30.ln_2.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.30.ln_2.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.30.mlp.c_fc.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.30.mlp.c_fc.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.30.mlp.c_proj.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.30.mlp.c_proj.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.31.attn.c_attn.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.31.attn.c_attn.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.31.attn.c_proj.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.31.attn.c_proj.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.31.ln_1.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.31.ln_1.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.31.ln_2.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.31.ln_2.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.31.mlp.c_fc.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.31.mlp.c_fc.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.31.mlp.c_proj.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.31.mlp.c_proj.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.32.attn.c_attn.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.32.attn.c_attn.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.32.attn.c_proj.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.32.attn.c_proj.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.32.ln_1.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.32.ln_1.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.32.ln_2.bias": "model-00011-of-00014.safetensors",
+    "transformer.h.32.ln_2.weight": "model-00011-of-00014.safetensors",
+    "transformer.h.32.mlp.c_fc.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.32.mlp.c_fc.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.32.mlp.c_proj.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.32.mlp.c_proj.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.33.attn.c_attn.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.33.attn.c_attn.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.33.attn.c_proj.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.33.attn.c_proj.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.33.ln_1.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.33.ln_1.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.33.ln_2.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.33.ln_2.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.33.mlp.c_fc.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.33.mlp.c_fc.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.33.mlp.c_proj.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.33.mlp.c_proj.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.34.attn.c_attn.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.34.attn.c_attn.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.34.attn.c_proj.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.34.attn.c_proj.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.34.ln_1.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.34.ln_1.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.34.ln_2.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.34.ln_2.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.34.mlp.c_fc.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.34.mlp.c_fc.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.34.mlp.c_proj.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.34.mlp.c_proj.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.35.attn.c_attn.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.35.attn.c_attn.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.35.attn.c_proj.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.35.attn.c_proj.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.35.ln_1.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.35.ln_1.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.35.ln_2.bias": "model-00012-of-00014.safetensors",
+    "transformer.h.35.ln_2.weight": "model-00012-of-00014.safetensors",
+    "transformer.h.35.mlp.c_fc.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.35.mlp.c_fc.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.35.mlp.c_proj.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.35.mlp.c_proj.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.36.attn.c_attn.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.36.attn.c_attn.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.36.attn.c_proj.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.36.attn.c_proj.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.36.ln_1.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.36.ln_1.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.36.ln_2.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.36.ln_2.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.36.mlp.c_fc.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.36.mlp.c_fc.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.36.mlp.c_proj.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.36.mlp.c_proj.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.37.attn.c_attn.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.37.attn.c_attn.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.37.attn.c_proj.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.37.attn.c_proj.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.37.ln_1.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.37.ln_1.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.37.ln_2.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.37.ln_2.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.37.mlp.c_fc.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.37.mlp.c_fc.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.37.mlp.c_proj.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.37.mlp.c_proj.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.38.attn.c_attn.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.38.attn.c_attn.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.38.attn.c_proj.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.38.attn.c_proj.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.38.ln_1.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.38.ln_1.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.38.ln_2.bias": "model-00013-of-00014.safetensors",
+    "transformer.h.38.ln_2.weight": "model-00013-of-00014.safetensors",
+    "transformer.h.38.mlp.c_fc.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.38.mlp.c_fc.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.38.mlp.c_proj.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.38.mlp.c_proj.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.39.attn.c_attn.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.39.attn.c_attn.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.39.attn.c_proj.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.39.attn.c_proj.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.39.ln_1.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.39.ln_1.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.39.ln_2.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.39.ln_2.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.39.mlp.c_fc.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.39.mlp.c_fc.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.39.mlp.c_proj.bias": "model-00014-of-00014.safetensors",
+    "transformer.h.39.mlp.c_proj.weight": "model-00014-of-00014.safetensors",
+    "transformer.h.4.attn.c_attn.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.4.attn.c_attn.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.4.attn.c_proj.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.4.attn.c_proj.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.4.ln_1.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.4.ln_1.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.4.ln_2.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.4.ln_2.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.4.mlp.c_fc.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.4.mlp.c_fc.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.4.mlp.c_proj.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.4.mlp.c_proj.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.5.attn.c_attn.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.5.attn.c_attn.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.5.attn.c_proj.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.5.attn.c_proj.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.5.ln_1.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.5.ln_1.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.5.ln_2.bias": "model-00002-of-00014.safetensors",
+    "transformer.h.5.ln_2.weight": "model-00002-of-00014.safetensors",
+    "transformer.h.5.mlp.c_fc.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.5.mlp.c_fc.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.5.mlp.c_proj.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.5.mlp.c_proj.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.6.attn.c_attn.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.6.attn.c_attn.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.6.attn.c_proj.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.6.attn.c_proj.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.6.ln_1.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.6.ln_1.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.6.ln_2.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.6.ln_2.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.6.mlp.c_fc.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.6.mlp.c_fc.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.6.mlp.c_proj.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.6.mlp.c_proj.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.7.attn.c_attn.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.7.attn.c_attn.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.7.attn.c_proj.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.7.attn.c_proj.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.7.ln_1.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.7.ln_1.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.7.ln_2.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.7.ln_2.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.7.mlp.c_fc.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.7.mlp.c_fc.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.7.mlp.c_proj.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.7.mlp.c_proj.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.8.attn.c_attn.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.8.attn.c_attn.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.8.attn.c_proj.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.8.attn.c_proj.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.8.ln_1.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.8.ln_1.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.8.ln_2.bias": "model-00003-of-00014.safetensors",
+    "transformer.h.8.ln_2.weight": "model-00003-of-00014.safetensors",
+    "transformer.h.8.mlp.c_fc.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.8.mlp.c_fc.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.8.mlp.c_proj.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.8.mlp.c_proj.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.9.attn.c_attn.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.9.attn.c_attn.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.9.attn.c_proj.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.9.attn.c_proj.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.9.ln_1.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.9.ln_1.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.9.ln_2.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.9.ln_2.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.9.mlp.c_fc.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.9.mlp.c_fc.weight": "model-00004-of-00014.safetensors",
+    "transformer.h.9.mlp.c_proj.bias": "model-00004-of-00014.safetensors",
+    "transformer.h.9.mlp.c_proj.weight": "model-00004-of-00014.safetensors",
+    "transformer.ln_f.bias": "model-00014-of-00014.safetensors",
+    "transformer.ln_f.weight": "model-00014-of-00014.safetensors",
+    "transformer.wpe.weight": "model-00001-of-00014.safetensors",
+    "transformer.wte.weight": "model-00001-of-00014.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,44 @@
+{
+  "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
+  },
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,185 @@
+{
+  "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": 1000000000000000019884624838656,
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>",
+  "vocab_size": 49152
+}