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-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:514257939a92df8a25e830cac0b8e4751d383b07fbaa82d8f7948734ceec23b3
+size 2919371160

model-00002-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d6a8d99690b735ff70209009e7f0e232f10783b7e67b3e0821a798236771f31
+size 2919371152

model-00003-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69efd8145fc1f096ba187690b57e6adecfb1ce64e1e8432df477d19008b83127
+size 2919371160

model-00004-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a621ed07686176391aff310c19cd4337260655c38ada4a0d3a33cf7a9c4aa527
+size 2952908968

model-00005-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b3632d7e64c250de6175e9d58f24ac75b0e1ea3f81843a8e27bab2dfcb6ea1e8
+size 2952925472

model-00006-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a8bf8f37463cc51a9b124d4885152419fd73254d5a0e5b5d8396f43a04460d9
+size 2919354680

model-00007-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7fb2a003f08823d93586963f7398012c4ba412c8d97cba7e5f1c1286045434d6
+size 2919371192

model-00008-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1acac375bf741d4ad7a6b3307de91c7f357975247324573849606b9de88a9c4f
+size 2919371192

model-00009-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1cbf435e7d7623444489ee8fa0d9ad9e146a838dbfa18f4b0296af6c7b694516
+size 2952908968

model-00010-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d8df22c4eb72cfb67bd73d90eb71e473c0a58cb783764caf403538873ca197a
+size 2952925472

model-00011-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4924d6e1cef893395214b168bd251699ab66b0bdad8f82ef84c0cbe3e12b03dc
+size 2919354680

model-00012-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb5ac6facbcf40ccb57e6fb3738ccbb0a1c17ea1e887228fcf29985329c25639
+size 2919371192

model-00013-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc7d7003faa4969e9acdb41c2a75f571d5e0263fca88302646eb505416f3ad2b
+size 2903264272

model.safetensors.index.json

@@ -0,0 +1,443 @@
+{
+    "metadata": {
+        "total_size": 38069813248
+    },
+    "weight_map": {
+        "encoder.layers.0.ffn.ffn.w_in.w_0.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.0.ffn.ffn.w_in.w_1.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.0.ffn.ffn.w_out.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.0.ffn.layernorm_before_ffn.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.0.self_att.layernorm_before_attention.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.0.self_att.self_attention.attention_out.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.0.self_att.self_attention.project_k.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.0.self_att.self_attention.project_q.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.0.self_att.self_attention.project_v.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.1.ffn.ffn.w_in.w_0.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.1.ffn.ffn.w_in.w_1.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.1.ffn.ffn.w_out.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.1.ffn.layernorm_before_ffn.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.1.self_att.layernorm_before_attention.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.1.self_att.self_attention.attention_out.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.1.self_att.self_attention.project_k.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.1.self_att.self_attention.project_q.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.1.self_att.self_attention.project_v.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.10.ffn.ffn.w_in.w_0.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.10.ffn.ffn.w_in.w_1.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.10.ffn.ffn.w_out.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.10.ffn.layernorm_before_ffn.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.10.self_att.layernorm_before_attention.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.10.self_att.self_attention.attention_out.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.10.self_att.self_attention.project_k.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.10.self_att.self_attention.project_q.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.10.self_att.self_attention.project_v.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.11.ffn.ffn.w_in.w_0.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.11.ffn.ffn.w_in.w_1.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.11.ffn.ffn.w_out.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.11.ffn.layernorm_before_ffn.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.11.self_att.layernorm_before_attention.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.11.self_att.self_attention.attention_out.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.11.self_att.self_attention.project_k.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.11.self_att.self_attention.project_q.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.11.self_att.self_attention.project_v.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.12.ffn.ffn.w_in.w_0.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.12.ffn.ffn.w_in.w_1.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.12.ffn.ffn.w_out.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.12.ffn.layernorm_before_ffn.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.12.self_att.layernorm_before_attention.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.12.self_att.self_attention.attention_out.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.12.self_att.self_attention.project_k.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.12.self_att.self_attention.project_q.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.12.self_att.self_attention.project_v.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.13.ffn.ffn.w_in.w_0.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.13.ffn.ffn.w_in.w_1.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.13.ffn.ffn.w_out.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.13.ffn.layernorm_before_ffn.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.13.self_att.layernorm_before_attention.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.13.self_att.self_attention.attention_out.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.13.self_att.self_attention.project_k.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.13.self_att.self_attention.project_q.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.13.self_att.self_attention.project_v.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.14.ffn.ffn.w_in.w_0.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.14.ffn.ffn.w_in.w_1.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.14.ffn.ffn.w_out.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.14.ffn.layernorm_before_ffn.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.14.self_att.layernorm_before_attention.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.14.self_att.self_attention.attention_out.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.14.self_att.self_attention.project_k.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.14.self_att.self_attention.project_q.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.14.self_att.self_attention.project_v.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.15.ffn.ffn.w_in.w_0.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.15.ffn.ffn.w_in.w_1.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.15.ffn.ffn.w_out.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.15.ffn.layernorm_before_ffn.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.15.self_att.layernorm_before_attention.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.15.self_att.self_attention.attention_out.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.15.self_att.self_attention.project_k.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.15.self_att.self_attention.project_q.weight": "model-00004-of-00013.safetensors",
+        "encoder.layers.15.self_att.self_attention.project_v.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.16.ffn.ffn.w_in.w_0.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.16.ffn.ffn.w_in.w_1.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.16.ffn.ffn.w_out.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.16.ffn.layernorm_before_ffn.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.16.self_att.layernorm_before_attention.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.16.self_att.self_attention.attention_out.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.16.self_att.self_attention.project_k.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.16.self_att.self_attention.project_q.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.16.self_att.self_attention.project_v.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.17.ffn.ffn.w_in.w_0.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.17.ffn.ffn.w_in.w_1.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.17.ffn.ffn.w_out.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.17.ffn.layernorm_before_ffn.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.17.self_att.layernorm_before_attention.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.17.self_att.self_attention.attention_out.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.17.self_att.self_attention.project_k.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.17.self_att.self_attention.project_q.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.17.self_att.self_attention.project_v.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.18.ffn.ffn.w_in.w_0.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.18.ffn.ffn.w_in.w_1.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.18.ffn.ffn.w_out.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.18.ffn.layernorm_before_ffn.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.18.self_att.layernorm_before_attention.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.18.self_att.self_attention.attention_out.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.18.self_att.self_attention.project_k.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.18.self_att.self_attention.project_q.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.18.self_att.self_attention.project_v.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.19.ffn.ffn.w_in.w_0.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.19.ffn.ffn.w_in.w_1.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.19.ffn.ffn.w_out.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.19.ffn.layernorm_before_ffn.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.19.self_att.layernorm_before_attention.weight": "model-00005-of-00013.safetensors",
+        "encoder.layers.19.self_att.self_attention.attention_out.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.19.self_att.self_attention.project_k.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.19.self_att.self_attention.project_q.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.19.self_att.self_attention.project_v.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.2.ffn.ffn.w_in.w_0.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.2.ffn.ffn.w_in.w_1.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.2.ffn.ffn.w_out.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.2.ffn.layernorm_before_ffn.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.2.self_att.layernorm_before_attention.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.2.self_att.self_attention.attention_out.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.2.self_att.self_attention.project_k.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.2.self_att.self_attention.project_q.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.2.self_att.self_attention.project_v.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.20.ffn.ffn.w_in.w_0.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.20.ffn.ffn.w_in.w_1.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.20.ffn.ffn.w_out.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.20.ffn.layernorm_before_ffn.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.20.self_att.layernorm_before_attention.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.20.self_att.self_attention.attention_out.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.20.self_att.self_attention.project_k.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.20.self_att.self_attention.project_q.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.20.self_att.self_attention.project_v.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.21.ffn.ffn.w_in.w_0.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.21.ffn.ffn.w_in.w_1.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.21.ffn.ffn.w_out.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.21.ffn.layernorm_before_ffn.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.21.self_att.layernorm_before_attention.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.21.self_att.self_attention.attention_out.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.21.self_att.self_attention.project_k.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.21.self_att.self_attention.project_q.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.21.self_att.self_attention.project_v.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.22.ffn.ffn.w_in.w_0.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.22.ffn.ffn.w_in.w_1.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.22.ffn.ffn.w_out.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.22.ffn.layernorm_before_ffn.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.22.self_att.layernorm_before_attention.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.22.self_att.self_attention.attention_out.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.22.self_att.self_attention.project_k.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.22.self_att.self_attention.project_q.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.22.self_att.self_attention.project_v.weight": "model-00006-of-00013.safetensors",
+        "encoder.layers.23.ffn.ffn.w_in.w_0.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.23.ffn.ffn.w_in.w_1.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.23.ffn.ffn.w_out.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.23.ffn.layernorm_before_ffn.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.23.self_att.layernorm_before_attention.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.23.self_att.self_attention.attention_out.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.23.self_att.self_attention.project_k.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.23.self_att.self_attention.project_q.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.23.self_att.self_attention.project_v.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.24.ffn.ffn.w_in.w_0.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.24.ffn.ffn.w_in.w_1.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.24.ffn.ffn.w_out.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.24.ffn.layernorm_before_ffn.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.24.self_att.layernorm_before_attention.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.24.self_att.self_attention.attention_out.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.24.self_att.self_attention.project_k.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.24.self_att.self_attention.project_q.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.24.self_att.self_attention.project_v.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.25.ffn.ffn.w_in.w_0.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.25.ffn.ffn.w_in.w_1.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.25.ffn.ffn.w_out.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.25.ffn.layernorm_before_ffn.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.25.self_att.layernorm_before_attention.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.25.self_att.self_attention.attention_out.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.25.self_att.self_attention.project_k.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.25.self_att.self_attention.project_q.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.25.self_att.self_attention.project_v.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.26.ffn.ffn.w_in.w_0.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.26.ffn.ffn.w_in.w_1.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.26.ffn.ffn.w_out.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.26.ffn.layernorm_before_ffn.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.26.self_att.layernorm_before_attention.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.26.self_att.self_attention.attention_out.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.26.self_att.self_attention.project_k.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.26.self_att.self_attention.project_q.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.26.self_att.self_attention.project_v.weight": "model-00007-of-00013.safetensors",
+        "encoder.layers.27.ffn.ffn.w_in.w_0.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.27.ffn.ffn.w_in.w_1.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.27.ffn.ffn.w_out.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.27.ffn.layernorm_before_ffn.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.27.self_att.layernorm_before_attention.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.27.self_att.self_attention.attention_out.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.27.self_att.self_attention.project_k.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.27.self_att.self_attention.project_q.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.27.self_att.self_attention.project_v.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.28.ffn.ffn.w_in.w_0.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.28.ffn.ffn.w_in.w_1.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.28.ffn.ffn.w_out.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.28.ffn.layernorm_before_ffn.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.28.self_att.layernorm_before_attention.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.28.self_att.self_attention.attention_out.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.28.self_att.self_attention.project_k.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.28.self_att.self_attention.project_q.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.28.self_att.self_attention.project_v.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.29.ffn.ffn.w_in.w_0.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.29.ffn.ffn.w_in.w_1.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.29.ffn.ffn.w_out.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.29.ffn.layernorm_before_ffn.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.29.self_att.layernorm_before_attention.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.29.self_att.self_attention.attention_out.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.29.self_att.self_attention.project_k.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.29.self_att.self_attention.project_q.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.29.self_att.self_attention.project_v.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.3.ffn.ffn.w_in.w_0.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.3.ffn.ffn.w_in.w_1.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.3.ffn.ffn.w_out.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.3.ffn.layernorm_before_ffn.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.3.self_att.layernorm_before_attention.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.3.self_att.self_attention.attention_out.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.3.self_att.self_attention.project_k.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.3.self_att.self_attention.project_q.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.3.self_att.self_attention.project_v.weight": "model-00001-of-00013.safetensors",
+        "encoder.layers.30.ffn.ffn.w_in.w_0.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.30.ffn.ffn.w_in.w_1.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.30.ffn.ffn.w_out.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.30.ffn.layernorm_before_ffn.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.30.self_att.layernorm_before_attention.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.30.self_att.self_attention.attention_out.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.30.self_att.self_attention.project_k.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.30.self_att.self_attention.project_q.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.30.self_att.self_attention.project_v.weight": "model-00008-of-00013.safetensors",
+        "encoder.layers.31.ffn.ffn.w_in.w_0.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.31.ffn.ffn.w_in.w_1.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.31.ffn.ffn.w_out.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.31.ffn.layernorm_before_ffn.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.31.self_att.layernorm_before_attention.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.31.self_att.self_attention.attention_out.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.31.self_att.self_attention.project_k.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.31.self_att.self_attention.project_q.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.31.self_att.self_attention.project_v.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.32.ffn.ffn.w_in.w_0.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.32.ffn.ffn.w_in.w_1.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.32.ffn.ffn.w_out.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.32.ffn.layernorm_before_ffn.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.32.self_att.layernorm_before_attention.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.32.self_att.self_attention.attention_out.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.32.self_att.self_attention.project_k.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.32.self_att.self_attention.project_q.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.32.self_att.self_attention.project_v.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.33.ffn.ffn.w_in.w_0.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.33.ffn.ffn.w_in.w_1.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.33.ffn.ffn.w_out.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.33.ffn.layernorm_before_ffn.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.33.self_att.layernorm_before_attention.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.33.self_att.self_attention.attention_out.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.33.self_att.self_attention.project_k.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.33.self_att.self_attention.project_q.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.33.self_att.self_attention.project_v.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.34.ffn.ffn.w_in.w_0.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.34.ffn.ffn.w_in.w_1.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.34.ffn.ffn.w_out.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.34.ffn.layernorm_before_ffn.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.34.self_att.layernorm_before_attention.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.34.self_att.self_attention.attention_out.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.34.self_att.self_attention.project_k.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.34.self_att.self_attention.project_q.weight": "model-00009-of-00013.safetensors",
+        "encoder.layers.34.self_att.self_attention.project_v.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.35.ffn.ffn.w_in.w_0.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.35.ffn.ffn.w_in.w_1.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.35.ffn.ffn.w_out.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.35.ffn.layernorm_before_ffn.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.35.self_att.layernorm_before_attention.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.35.self_att.self_attention.attention_out.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.35.self_att.self_attention.project_k.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.35.self_att.self_attention.project_q.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.35.self_att.self_attention.project_v.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.36.ffn.ffn.w_in.w_0.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.36.ffn.ffn.w_in.w_1.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.36.ffn.ffn.w_out.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.36.ffn.layernorm_before_ffn.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.36.self_att.layernorm_before_attention.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.36.self_att.self_attention.attention_out.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.36.self_att.self_attention.project_k.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.36.self_att.self_attention.project_q.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.36.self_att.self_attention.project_v.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.37.ffn.ffn.w_in.w_0.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.37.ffn.ffn.w_in.w_1.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.37.ffn.ffn.w_out.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.37.ffn.layernorm_before_ffn.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.37.self_att.layernorm_before_attention.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.37.self_att.self_attention.attention_out.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.37.self_att.self_attention.project_k.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.37.self_att.self_attention.project_q.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.37.self_att.self_attention.project_v.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.38.ffn.ffn.w_in.w_0.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.38.ffn.ffn.w_in.w_1.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.38.ffn.ffn.w_out.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.38.ffn.layernorm_before_ffn.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.38.self_att.layernorm_before_attention.weight": "model-00010-of-00013.safetensors",
+        "encoder.layers.38.self_att.self_attention.attention_out.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.38.self_att.self_attention.project_k.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.38.self_att.self_attention.project_q.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.38.self_att.self_attention.project_v.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.39.ffn.ffn.w_in.w_0.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.39.ffn.ffn.w_in.w_1.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.39.ffn.ffn.w_out.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.39.ffn.layernorm_before_ffn.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.39.self_att.layernorm_before_attention.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.39.self_att.self_attention.attention_out.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.39.self_att.self_attention.project_k.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.39.self_att.self_attention.project_q.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.39.self_att.self_attention.project_v.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.4.ffn.ffn.w_in.w_0.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.4.ffn.ffn.w_in.w_1.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.4.ffn.ffn.w_out.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.4.ffn.layernorm_before_ffn.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.4.self_att.layernorm_before_attention.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.4.self_att.self_attention.attention_out.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.4.self_att.self_attention.project_k.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.4.self_att.self_attention.project_q.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.4.self_att.self_attention.project_v.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.40.ffn.ffn.w_in.w_0.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.40.ffn.ffn.w_in.w_1.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.40.ffn.ffn.w_out.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.40.ffn.layernorm_before_ffn.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.40.self_att.layernorm_before_attention.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.40.self_att.self_attention.attention_out.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.40.self_att.self_attention.project_k.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.40.self_att.self_attention.project_q.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.40.self_att.self_attention.project_v.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.41.ffn.ffn.w_in.w_0.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.41.ffn.ffn.w_in.w_1.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.41.ffn.ffn.w_out.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.41.ffn.layernorm_before_ffn.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.41.self_att.layernorm_before_attention.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.41.self_att.self_attention.attention_out.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.41.self_att.self_attention.project_k.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.41.self_att.self_attention.project_q.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.41.self_att.self_attention.project_v.weight": "model-00011-of-00013.safetensors",
+        "encoder.layers.42.ffn.ffn.w_in.w_0.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.42.ffn.ffn.w_in.w_1.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.42.ffn.ffn.w_out.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.42.ffn.layernorm_before_ffn.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.42.self_att.layernorm_before_attention.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.42.self_att.self_attention.attention_out.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.42.self_att.self_attention.project_k.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.42.self_att.self_attention.project_q.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.42.self_att.self_attention.project_v.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.43.ffn.ffn.w_in.w_0.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.43.ffn.ffn.w_in.w_1.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.43.ffn.ffn.w_out.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.43.ffn.layernorm_before_ffn.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.43.self_att.layernorm_before_attention.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.43.self_att.self_attention.attention_out.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.43.self_att.self_attention.project_k.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.43.self_att.self_attention.project_q.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.43.self_att.self_attention.project_v.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.44.ffn.ffn.w_in.w_0.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.44.ffn.ffn.w_in.w_1.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.44.ffn.ffn.w_out.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.44.ffn.layernorm_before_ffn.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.44.self_att.layernorm_before_attention.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.44.self_att.self_attention.attention_out.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.44.self_att.self_attention.project_k.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.44.self_att.self_attention.project_q.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.44.self_att.self_attention.project_v.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.45.ffn.ffn.w_in.w_0.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.45.ffn.ffn.w_in.w_1.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.45.ffn.ffn.w_out.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.45.ffn.layernorm_before_ffn.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.45.self_att.layernorm_before_attention.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.45.self_att.self_attention.attention_out.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.45.self_att.self_attention.project_k.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.45.self_att.self_attention.project_q.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.45.self_att.self_attention.project_v.weight": "model-00012-of-00013.safetensors",
+        "encoder.layers.46.ffn.ffn.w_in.w_0.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.46.ffn.ffn.w_in.w_1.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.46.ffn.ffn.w_out.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.46.ffn.layernorm_before_ffn.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.46.self_att.layernorm_before_attention.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.46.self_att.self_attention.attention_out.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.46.self_att.self_attention.project_k.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.46.self_att.self_attention.project_q.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.46.self_att.self_attention.project_v.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.47.ffn.ffn.w_in.w_0.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.47.ffn.ffn.w_in.w_1.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.47.ffn.ffn.w_out.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.47.ffn.layernorm_before_ffn.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.47.self_att.layernorm_before_attention.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.47.self_att.self_attention.attention_out.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.47.self_att.self_attention.project_k.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.47.self_att.self_attention.project_q.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.47.self_att.self_attention.project_v.weight": "model-00013-of-00013.safetensors",
+        "encoder.layers.5.ffn.ffn.w_in.w_0.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.5.ffn.ffn.w_in.w_1.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.5.ffn.ffn.w_out.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.5.ffn.layernorm_before_ffn.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.5.self_att.layernorm_before_attention.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.5.self_att.self_attention.attention_out.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.5.self_att.self_attention.project_k.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.5.self_att.self_attention.project_q.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.5.self_att.self_attention.project_v.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.6.ffn.ffn.w_in.w_0.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.6.ffn.ffn.w_in.w_1.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.6.ffn.ffn.w_out.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.6.ffn.layernorm_before_ffn.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.6.self_att.layernorm_before_attention.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.6.self_att.self_attention.attention_out.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.6.self_att.self_attention.project_k.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.6.self_att.self_attention.project_q.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.6.self_att.self_attention.project_v.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.7.ffn.ffn.w_in.w_0.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.7.ffn.ffn.w_in.w_1.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.7.ffn.ffn.w_out.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.7.ffn.layernorm_before_ffn.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.7.self_att.layernorm_before_attention.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.7.self_att.self_attention.attention_out.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.7.self_att.self_attention.project_k.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.7.self_att.self_attention.project_q.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.7.self_att.self_attention.project_v.weight": "model-00002-of-00013.safetensors",
+        "encoder.layers.8.ffn.ffn.w_in.w_0.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.8.ffn.ffn.w_in.w_1.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.8.ffn.ffn.w_out.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.8.ffn.layernorm_before_ffn.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.8.self_att.layernorm_before_attention.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.8.self_att.self_attention.attention_out.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.8.self_att.self_attention.project_k.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.8.self_att.self_attention.project_q.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.8.self_att.self_attention.project_v.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.9.ffn.ffn.w_in.w_0.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.9.ffn.ffn.w_in.w_1.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.9.ffn.ffn.w_out.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.9.ffn.layernorm_before_ffn.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.9.self_att.layernorm_before_attention.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.9.self_att.self_attention.attention_out.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.9.self_att.self_attention.project_k.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.9.self_att.self_attention.project_q.weight": "model-00003-of-00013.safetensors",
+        "encoder.layers.9.self_att.self_attention.project_v.weight": "model-00003-of-00013.safetensors",
+        "encoder.output_layernorm.weight": "model-00013-of-00013.safetensors",
+        "input_embedding.weight": "model-00013-of-00013.safetensors",
+        "position_bias.relative_attention_bias": "model-00013-of-00013.safetensors",
+        "segment_embedding.weight": "model-00013-of-00013.safetensors"
+    }
+}