Adding `safetensors` variant of this model

This is an automated PR created with https://huggingface.co/spaces/safetensors/convert

This new file is equivalent to `pytorch_model.bin` but safe in the sense that
no arbitrary code can be put into it.

These files also happen to load much faster than their pytorch counterpart:
https://colab.research.google.com/github/huggingface/notebooks/blob/main/safetensors_doc/en/speed.ipynb

The widgets on your model page will run using this model even if this is not merged
making sure the file actually works.

If you find any issues: please report here: https://huggingface.co/spaces/safetensors/convert/discussions

Feel free to ignore this PR.

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1f90b82e0c11ca5792623b4b7a1e8fe81980149161e259cf57508943d3d4f35
+size 587829056

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2343cf6a75270630b5b17b167d2c0f38c16811cc7c7051da2f4e72268cdb3c6
+size 625472828

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d22d7fe07e3b7fe05d1d3dbb7327c863325067b8e9d4f354ffa413516fc4dc6
+size 621261980

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb52086e9c5a18a59479b7f8282b77c837edd9b3a4b4a585a8e419f58dd5479f
+size 205852800

model.safetensors.index.json

@@ -0,0 +1,348 @@
+{
+    "metadata": {
+        "total_size": 1688060000.0
+    },
+    "weight_map": {
+        "lm_head.weight": "model-00004-of-00004.safetensors",
+        "transformer.h.0.attn.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.0.attn.c_attn.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.0.attn.c_attn.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.0.attn.c_proj.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.0.attn.c_proj.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.0.attn.masked_bias": "model-00001-of-00004.safetensors",
+        "transformer.h.0.ln_1.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.0.ln_1.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.0.ln_2.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.0.ln_2.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.0.mlp.c_fc.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.0.mlp.c_fc.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.0.mlp.c_proj.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.0.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.1.attn.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.1.attn.c_attn.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.1.attn.c_attn.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.1.attn.c_proj.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.1.attn.c_proj.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.1.attn.masked_bias": "model-00001-of-00004.safetensors",
+        "transformer.h.1.ln_1.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.1.ln_1.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.1.ln_2.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.1.ln_2.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.1.mlp.c_fc.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.1.mlp.c_fc.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.1.mlp.c_proj.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.1.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.10.attn.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.10.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.10.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.10.attn.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.10.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.10.attn.masked_bias": "model-00002-of-00004.safetensors",
+        "transformer.h.10.ln_1.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.10.ln_1.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.10.ln_2.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.10.ln_2.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.10.mlp.c_fc.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.10.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.10.mlp.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.10.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.11.attn.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.11.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.11.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.11.attn.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.11.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.11.attn.masked_bias": "model-00002-of-00004.safetensors",
+        "transformer.h.11.ln_1.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.11.ln_1.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.11.ln_2.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.11.ln_2.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.11.mlp.c_fc.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.11.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.11.mlp.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.11.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.12.attn.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.12.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.12.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.12.attn.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.12.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.12.attn.masked_bias": "model-00002-of-00004.safetensors",
+        "transformer.h.12.ln_1.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.12.ln_1.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.12.ln_2.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.12.ln_2.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.12.mlp.c_fc.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.12.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.12.mlp.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.12.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.13.attn.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.13.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.13.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.13.attn.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.13.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.13.attn.masked_bias": "model-00002-of-00004.safetensors",
+        "transformer.h.13.ln_1.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.13.ln_1.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.13.ln_2.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.13.ln_2.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.13.mlp.c_fc.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.13.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.13.mlp.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.13.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.14.attn.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.14.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.14.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.14.attn.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.14.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.14.attn.masked_bias": "model-00002-of-00004.safetensors",
+        "transformer.h.14.ln_1.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.14.ln_1.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.14.ln_2.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.14.ln_2.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.14.mlp.c_fc.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.14.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.14.mlp.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.14.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.15.attn.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.15.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.15.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.15.attn.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.15.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.15.attn.masked_bias": "model-00003-of-00004.safetensors",
+        "transformer.h.15.ln_1.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.15.ln_1.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.15.ln_2.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.15.ln_2.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.15.mlp.c_fc.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.15.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.15.mlp.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.15.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.16.attn.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.16.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.16.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.16.attn.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.16.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.16.attn.masked_bias": "model-00003-of-00004.safetensors",
+        "transformer.h.16.ln_1.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.16.ln_1.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.16.ln_2.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.16.ln_2.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.16.mlp.c_fc.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.16.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.16.mlp.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.16.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.17.attn.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.17.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.17.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.17.attn.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.17.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.17.attn.masked_bias": "model-00003-of-00004.safetensors",
+        "transformer.h.17.ln_1.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.17.ln_1.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.17.ln_2.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.17.ln_2.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.17.mlp.c_fc.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.17.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.17.mlp.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.17.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.18.attn.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.18.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.18.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.18.attn.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.18.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.18.attn.masked_bias": "model-00003-of-00004.safetensors",
+        "transformer.h.18.ln_1.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.18.ln_1.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.18.ln_2.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.18.ln_2.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.18.mlp.c_fc.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.18.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.18.mlp.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.18.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.19.attn.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.19.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.19.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.19.attn.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.19.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.19.attn.masked_bias": "model-00003-of-00004.safetensors",
+        "transformer.h.19.ln_1.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.19.ln_1.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.19.ln_2.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.19.ln_2.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.19.mlp.c_fc.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.19.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.19.mlp.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.19.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.2.attn.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.2.attn.c_attn.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.2.attn.c_attn.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.2.attn.c_proj.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.2.attn.c_proj.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.2.attn.masked_bias": "model-00001-of-00004.safetensors",
+        "transformer.h.2.ln_1.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.2.ln_1.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.2.ln_2.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.2.ln_2.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.2.mlp.c_fc.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.2.mlp.c_fc.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.2.mlp.c_proj.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.2.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.20.attn.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.20.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.20.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.20.attn.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.20.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.20.attn.masked_bias": "model-00003-of-00004.safetensors",
+        "transformer.h.20.ln_1.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.20.ln_1.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.20.ln_2.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.20.ln_2.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.20.mlp.c_fc.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.20.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.20.mlp.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.20.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.21.attn.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.21.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.21.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.21.attn.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.21.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.21.attn.masked_bias": "model-00003-of-00004.safetensors",
+        "transformer.h.21.ln_1.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.21.ln_1.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.21.ln_2.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.21.ln_2.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.21.mlp.c_fc.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.21.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.21.mlp.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.21.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.22.attn.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.22.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.22.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.22.attn.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.22.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.22.attn.masked_bias": "model-00003-of-00004.safetensors",
+        "transformer.h.22.ln_1.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.22.ln_1.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.22.ln_2.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.22.ln_2.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.22.mlp.c_fc.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.22.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.22.mlp.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.22.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.23.attn.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.23.attn.c_attn.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.23.attn.c_attn.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.23.attn.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.23.attn.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.23.attn.masked_bias": "model-00003-of-00004.safetensors",
+        "transformer.h.23.ln_1.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.23.ln_1.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.23.ln_2.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.23.ln_2.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.23.mlp.c_fc.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.23.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.23.mlp.c_proj.bias": "model-00003-of-00004.safetensors",
+        "transformer.h.23.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
+        "transformer.h.3.attn.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.3.attn.c_attn.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.3.attn.c_attn.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.3.attn.c_proj.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.3.attn.c_proj.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.3.attn.masked_bias": "model-00001-of-00004.safetensors",
+        "transformer.h.3.ln_1.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.3.ln_1.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.3.ln_2.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.3.ln_2.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.3.mlp.c_fc.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.3.mlp.c_fc.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.3.mlp.c_proj.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.3.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.4.attn.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.4.attn.c_attn.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.4.attn.c_attn.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.4.attn.c_proj.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.4.attn.c_proj.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.4.attn.masked_bias": "model-00001-of-00004.safetensors",
+        "transformer.h.4.ln_1.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.4.ln_1.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.4.ln_2.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.4.ln_2.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.4.mlp.c_fc.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.4.mlp.c_fc.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.4.mlp.c_proj.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.4.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.5.attn.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.5.attn.c_attn.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.5.attn.c_attn.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.5.attn.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.5.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.5.attn.masked_bias": "model-00001-of-00004.safetensors",
+        "transformer.h.5.ln_1.bias": "model-00001-of-00004.safetensors",
+        "transformer.h.5.ln_1.weight": "model-00001-of-00004.safetensors",
+        "transformer.h.5.ln_2.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.5.ln_2.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.5.mlp.c_fc.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.5.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.5.mlp.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.5.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.6.attn.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.6.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.6.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.6.attn.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.6.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.6.attn.masked_bias": "model-00002-of-00004.safetensors",
+        "transformer.h.6.ln_1.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.6.ln_1.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.6.ln_2.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.6.ln_2.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.6.mlp.c_fc.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.6.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.6.mlp.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.6.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.7.attn.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.7.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.7.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.7.attn.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.7.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.7.attn.masked_bias": "model-00002-of-00004.safetensors",
+        "transformer.h.7.ln_1.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.7.ln_1.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.7.ln_2.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.7.ln_2.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.7.mlp.c_fc.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.7.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.7.mlp.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.7.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.8.attn.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.8.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.8.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.8.attn.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.8.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.8.attn.masked_bias": "model-00002-of-00004.safetensors",
+        "transformer.h.8.ln_1.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.8.ln_1.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.8.ln_2.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.8.ln_2.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.8.mlp.c_fc.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.8.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.8.mlp.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.8.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.9.attn.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.9.attn.c_attn.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.9.attn.c_attn.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.9.attn.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.9.attn.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.9.attn.masked_bias": "model-00002-of-00004.safetensors",
+        "transformer.h.9.ln_1.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.9.ln_1.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.9.ln_2.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.9.ln_2.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.9.mlp.c_fc.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.9.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
+        "transformer.h.9.mlp.c_proj.bias": "model-00002-of-00004.safetensors",
+        "transformer.h.9.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
+        "transformer.ln_f.bias": "model-00003-of-00004.safetensors",
+        "transformer.ln_f.weight": "model-00003-of-00004.safetensors",
+        "transformer.wpe.weight": "model-00001-of-00004.safetensors",
+        "transformer.wte.weight": "model-00001-of-00004.safetensors"
+    }
+}