Upload LlamaForCausalLM

config.json

@@ -0,0 +1,27 @@
+{
+  "_name_or_path": "Yi-DPO-v1-34B",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 7168,
+  "initializer_range": 0.02,
+  "intermediate_size": 20480,
+  "max_position_embeddings": 4096,
+  "model_type": "llama",
+  "num_attention_heads": 56,
+  "num_hidden_layers": 60,
+  "num_key_value_heads": 8,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 5000000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.35.2",
+  "use_cache": false,
+  "vocab_size": 64002
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "bos_token_id": 6,
+  "do_sample": true,
+  "eos_token_id": 7,
+  "max_length": 4096,
+  "pad_token_id": 0,
+  "temperature": 0.6,
+  "top_p": 0.8,
+  "transformers_version": "4.35.2"
+}

model-00001-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3bbac63ef59c7780f095743e3ed808132a42b8d7900c3ebd87a681b6fc1f1ee
+size 4536256096

model-00002-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5bd0e497338cbd7f968331b6986b837a15d3d064804b3ac1ff447d61f09ac553
+size 4462856232

model-00003-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee68576472828173adc717b35f8f4091bea4f80e267de70fa90f8d45cba13b80
+size 4462856232

model-00004-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:926f7118c24bb598cdf1750cb40604fc56dde990ea7b20e024d782b7b2fdb660
+size 4462856232

model-00005-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:657c89bf9d81ba58274944f0a9b5e27eec3911484d566f026b8365b6ce014792
+size 4462856232

model-00006-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:def886cbb8d858f035bfd71e1fa9f2135f86fac6e79cfa8ebb81550cbdb9ad2c
+size 4462856248

model-00007-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72557e458dd0f9bcdd0f8011b914ff80a0b4f59c160ad3d46a2796d4d46459dd
+size 4462856248

model-00008-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39a4cbf1b928e324bc363bbd973535848eef5377d0c6ee97a57280cb1fbcc106
+size 4462856248

model-00009-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c48a3f6c345e4d7a7116b8887e877a5822ea616c88ce30f41b1373d908992f0e
+size 4462856248

model-00010-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20d5728daa08090ea7e0b0fbef2623a2b54e4b38b1e13463ae5f6423239acd22
+size 4462856248

model-00011-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a78a1d1b5d334a0f46fa86c36ee700d9145d1196a5505d0cd9cc2df1254513d3
+size 4462856248

model-00012-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bda6928df7b961e3cff7f29fc44fbece35293a87a8d6b4d8612811efe73256c5
+size 4462856248

model-00013-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3727fbb553aec7adbd2eb4d6a5e6e5c584c7b5ae5d110e21e0908b17c65c0d00
+size 4462856248

model-00014-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0aaab329766c4af3f344a72e942fb5f7c5addd7cf85efef01652387639fff5a9
+size 4462856248

model-00015-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:10a39a695f7144bf02db908631491c78105f16251d7d16023f4d0ec95553d82d
+size 4462856248

model-00016-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:983a03d341ddc30bfdf9e2701f187b72635b548784c0f1207ba6d26c02eaacab
+size 4462856248

model-00017-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ae9a8776cf9c296401340d9ae147e19714ee83ccbf7a0b599ebb4177c451271
+size 4462856248

model-00018-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:500866c960ad42acf77ed39f746085f2cccde804f546fca2ab9d458533c6964e
+size 4462856248

model-00019-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f951722f576490c8d7dd1004969a6a5037cc942d8fb66d193fa947fbff0860c
+size 4462856248

model-00020-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24b729cf44fb11a49e397eaa6f29bbd4062fff2b9e080ea61807e8e4c24bfa9e
+size 4462856248

model-00021-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69c7993871811821e3ad9655cfc0526a2286ce9d55f9dd7fa719a987fbd46767
+size 4462856248

model-00022-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:374afdff62f1ef3186d034471ef531f1f9639e41e692926c15e6abdc4340ee45
+size 4462856248

model-00023-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8026fba7b05ea2cae4d06703aa27b47300d2ba21d63a6fba863b632f3cc87f1
+size 4462856248

model-00024-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9bebd2ad94afa7cd7f7a155988d2917b6808342219c26201617c722f0fc1f4b3
+size 4462856248

model-00025-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd2dc0281fb4e9e5e825e755011b46f7ca54d7a2c88b2960b9753002fe13e6e3
+size 4462856248

model-00026-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:956720dc2ee168527941414d6f945aecb2ef2f4d417c646e3d4f408284ba478a
+size 4462856248

model-00027-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b18ed73b97f6a8961ddc8f5e38882ab81564488296156462433cbf6426464de0
+size 4462856248

model-00028-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd9f67332a143126b424279bd741e0b1bfa2f3371c71ecc1f946fa5fcb9f9e71
+size 4462856248

model-00029-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd5be034da1e930c576c985928fb4fbe25760debbe744b775915b8c2596dc75b
+size 4462856248

model-00030-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:130989cc4589de3df6cdea578d7cdbd4980ff4e167cbe31085ff8e93e638853d
+size 4462856248

model-00031-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c61d001da5f1a3495f07642d4c70c790f604a8d31814126770a51c9ad6ffe869
+size 3596759832

model.safetensors.index.json

@@ -0,0 +1,550 @@
+{
+  "metadata": {
+    "total_size": 137555783680
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00031-of-00031.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00031.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00031.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00031.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00031.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00031.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00031.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00031.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00031.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00031.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00031.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00002-of-00031.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00002-of-00031.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00002-of-00031.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00002-of-00031.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00002-of-00031.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00031.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00031.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00031.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00031.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00006-of-00031.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00006-of-00031.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00006-of-00031.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00006-of-00031.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00006-of-00031.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00006-of-00031.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00006-of-00031.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00006-of-00031.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00006-of-00031.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00007-of-00031.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00007-of-00031.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00007-of-00031.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00007-of-00031.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00007-of-00031.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00006-of-00031.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00006-of-00031.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00006-of-00031.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00006-of-00031.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00007-of-00031.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00007-of-00031.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00007-of-00031.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00007-of-00031.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00007-of-00031.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00007-of-00031.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00007-of-00031.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00007-of-00031.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00007-of-00031.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00008-of-00031.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00008-of-00031.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00008-of-00031.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00008-of-00031.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00008-of-00031.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00007-of-00031.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00007-of-00031.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00007-of-00031.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00007-of-00031.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00008-of-00031.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00008-of-00031.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00008-of-00031.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00008-of-00031.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00008-of-00031.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00008-of-00031.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00008-of-00031.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00008-of-00031.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00008-of-00031.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00009-of-00031.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00009-of-00031.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00009-of-00031.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00009-of-00031.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00009-of-00031.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00008-of-00031.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00008-of-00031.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00008-of-00031.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00008-of-00031.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00009-of-00031.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00009-of-00031.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00009-of-00031.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00009-of-00031.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00009-of-00031.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00009-of-00031.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00009-of-00031.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00009-of-00031.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00009-of-00031.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00010-of-00031.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00010-of-00031.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00010-of-00031.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00010-of-00031.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00010-of-00031.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00009-of-00031.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00009-of-00031.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00009-of-00031.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00009-of-00031.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00010-of-00031.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00010-of-00031.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00010-of-00031.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00010-of-00031.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00010-of-00031.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00010-of-00031.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00010-of-00031.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00010-of-00031.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00010-of-00031.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00011-of-00031.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00011-of-00031.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00011-of-00031.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00011-of-00031.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00011-of-00031.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00010-of-00031.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00010-of-00031.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00010-of-00031.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00010-of-00031.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00002-of-00031.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00002-of-00031.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00002-of-00031.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00002-of-00031.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00002-of-00031.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00002-of-00031.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00002-of-00031.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00002-of-00031.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00002-of-00031.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00011-of-00031.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00011-of-00031.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00011-of-00031.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00011-of-00031.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00011-of-00031.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00011-of-00031.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00011-of-00031.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00011-of-00031.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00011-of-00031.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00012-of-00031.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00012-of-00031.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00012-of-00031.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00012-of-00031.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00012-of-00031.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00011-of-00031.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00011-of-00031.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00011-of-00031.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00011-of-00031.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00012-of-00031.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00012-of-00031.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00012-of-00031.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00012-of-00031.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00012-of-00031.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00012-of-00031.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00012-of-00031.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00012-of-00031.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00012-of-00031.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00013-of-00031.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00013-of-00031.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00013-of-00031.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00013-of-00031.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00013-of-00031.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00012-of-00031.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00012-of-00031.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00012-of-00031.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00012-of-00031.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00013-of-00031.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00013-of-00031.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00013-of-00031.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00013-of-00031.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00013-of-00031.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00013-of-00031.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00013-of-00031.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00013-of-00031.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00013-of-00031.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00014-of-00031.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00014-of-00031.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00014-of-00031.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00014-of-00031.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00014-of-00031.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00013-of-00031.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00013-of-00031.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00013-of-00031.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00013-of-00031.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00014-of-00031.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00014-of-00031.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00014-of-00031.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00014-of-00031.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00014-of-00031.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00014-of-00031.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00014-of-00031.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00014-of-00031.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00014-of-00031.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00015-of-00031.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00015-of-00031.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00015-of-00031.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00015-of-00031.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00015-of-00031.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00014-of-00031.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00014-of-00031.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00014-of-00031.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00014-of-00031.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00015-of-00031.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00015-of-00031.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00015-of-00031.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00015-of-00031.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00015-of-00031.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00015-of-00031.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00015-of-00031.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00015-of-00031.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00015-of-00031.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00016-of-00031.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00016-of-00031.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00016-of-00031.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00016-of-00031.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00016-of-00031.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00015-of-00031.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00015-of-00031.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00015-of-00031.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00015-of-00031.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00003-of-00031.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00003-of-00031.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00003-of-00031.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00003-of-00031.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00003-of-00031.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00002-of-00031.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00002-of-00031.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00002-of-00031.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00002-of-00031.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00016-of-00031.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00016-of-00031.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00016-of-00031.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00016-of-00031.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00016-of-00031.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00016-of-00031.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00016-of-00031.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00016-of-00031.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00016-of-00031.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00017-of-00031.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00017-of-00031.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00017-of-00031.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00017-of-00031.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00017-of-00031.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00016-of-00031.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00016-of-00031.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00016-of-00031.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00016-of-00031.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00017-of-00031.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-00017-of-00031.safetensors",
+    "model.layers.32.mlp.gate_proj.weight": "model-00017-of-00031.safetensors",
+    "model.layers.32.mlp.up_proj.weight": "model-00017-of-00031.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00017-of-00031.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00017-of-00031.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00017-of-00031.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00017-of-00031.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00017-of-00031.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00018-of-00031.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-00018-of-00031.safetensors",
+    "model.layers.33.mlp.gate_proj.weight": "model-00018-of-00031.safetensors",
+    "model.layers.33.mlp.up_proj.weight": "model-00018-of-00031.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00018-of-00031.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00017-of-00031.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00017-of-00031.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00017-of-00031.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00017-of-00031.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00018-of-00031.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-00018-of-00031.safetensors",
+    "model.layers.34.mlp.gate_proj.weight": "model-00018-of-00031.safetensors",
+    "model.layers.34.mlp.up_proj.weight": "model-00018-of-00031.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00018-of-00031.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00018-of-00031.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00018-of-00031.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00018-of-00031.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00018-of-00031.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00019-of-00031.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-00019-of-00031.safetensors",
+    "model.layers.35.mlp.gate_proj.weight": "model-00019-of-00031.safetensors",
+    "model.layers.35.mlp.up_proj.weight": "model-00019-of-00031.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00019-of-00031.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00018-of-00031.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00018-of-00031.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00018-of-00031.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00018-of-00031.safetensors",
+    "model.layers.36.input_layernorm.weight": "model-00019-of-00031.safetensors",
+    "model.layers.36.mlp.down_proj.weight": "model-00019-of-00031.safetensors",
+    "model.layers.36.mlp.gate_proj.weight": "model-00019-of-00031.safetensors",
+    "model.layers.36.mlp.up_proj.weight": "model-00019-of-00031.safetensors",
+    "model.layers.36.post_attention_layernorm.weight": "model-00019-of-00031.safetensors",
+    "model.layers.36.self_attn.k_proj.weight": "model-00019-of-00031.safetensors",
+    "model.layers.36.self_attn.o_proj.weight": "model-00019-of-00031.safetensors",
+    "model.layers.36.self_attn.q_proj.weight": "model-00019-of-00031.safetensors",
+    "model.layers.36.self_attn.v_proj.weight": "model-00019-of-00031.safetensors",
+    "model.layers.37.input_layernorm.weight": "model-00020-of-00031.safetensors",
+    "model.layers.37.mlp.down_proj.weight": "model-00020-of-00031.safetensors",
+    "model.layers.37.mlp.gate_proj.weight": "model-00020-of-00031.safetensors",
+    "model.layers.37.mlp.up_proj.weight": "model-00020-of-00031.safetensors",
+    "model.layers.37.post_attention_layernorm.weight": "model-00020-of-00031.safetensors",
+    "model.layers.37.self_attn.k_proj.weight": "model-00019-of-00031.safetensors",
+    "model.layers.37.self_attn.o_proj.weight": "model-00019-of-00031.safetensors",
+    "model.layers.37.self_attn.q_proj.weight": "model-00019-of-00031.safetensors",
+    "model.layers.37.self_attn.v_proj.weight": "model-00019-of-00031.safetensors",
+    "model.layers.38.input_layernorm.weight": "model-00020-of-00031.safetensors",
+    "model.layers.38.mlp.down_proj.weight": "model-00020-of-00031.safetensors",
+    "model.layers.38.mlp.gate_proj.weight": "model-00020-of-00031.safetensors",
+    "model.layers.38.mlp.up_proj.weight": "model-00020-of-00031.safetensors",
+    "model.layers.38.post_attention_layernorm.weight": "model-00020-of-00031.safetensors",
+    "model.layers.38.self_attn.k_proj.weight": "model-00020-of-00031.safetensors",
+    "model.layers.38.self_attn.o_proj.weight": "model-00020-of-00031.safetensors",
+    "model.layers.38.self_attn.q_proj.weight": "model-00020-of-00031.safetensors",
+    "model.layers.38.self_attn.v_proj.weight": "model-00020-of-00031.safetensors",
+    "model.layers.39.input_layernorm.weight": "model-00021-of-00031.safetensors",
+    "model.layers.39.mlp.down_proj.weight": "model-00021-of-00031.safetensors",
+    "model.layers.39.mlp.gate_proj.weight": "model-00021-of-00031.safetensors",
+    "model.layers.39.mlp.up_proj.weight": "model-00021-of-00031.safetensors",
+    "model.layers.39.post_attention_layernorm.weight": "model-00021-of-00031.safetensors",
+    "model.layers.39.self_attn.k_proj.weight": "model-00020-of-00031.safetensors",
+    "model.layers.39.self_attn.o_proj.weight": "model-00020-of-00031.safetensors",
+    "model.layers.39.self_attn.q_proj.weight": "model-00020-of-00031.safetensors",
+    "model.layers.39.self_attn.v_proj.weight": "model-00020-of-00031.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00003-of-00031.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00003-of-00031.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00003-of-00031.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00003-of-00031.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00003-of-00031.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00003-of-00031.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00003-of-00031.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00003-of-00031.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00003-of-00031.safetensors",
+    "model.layers.40.input_layernorm.weight": "model-00021-of-00031.safetensors",
+    "model.layers.40.mlp.down_proj.weight": "model-00021-of-00031.safetensors",
+    "model.layers.40.mlp.gate_proj.weight": "model-00021-of-00031.safetensors",
+    "model.layers.40.mlp.up_proj.weight": "model-00021-of-00031.safetensors",
+    "model.layers.40.post_attention_layernorm.weight": "model-00021-of-00031.safetensors",
+    "model.layers.40.self_attn.k_proj.weight": "model-00021-of-00031.safetensors",
+    "model.layers.40.self_attn.o_proj.weight": "model-00021-of-00031.safetensors",
+    "model.layers.40.self_attn.q_proj.weight": "model-00021-of-00031.safetensors",
+    "model.layers.40.self_attn.v_proj.weight": "model-00021-of-00031.safetensors",
+    "model.layers.41.input_layernorm.weight": "model-00022-of-00031.safetensors",
+    "model.layers.41.mlp.down_proj.weight": "model-00022-of-00031.safetensors",
+    "model.layers.41.mlp.gate_proj.weight": "model-00022-of-00031.safetensors",
+    "model.layers.41.mlp.up_proj.weight": "model-00022-of-00031.safetensors",
+    "model.layers.41.post_attention_layernorm.weight": "model-00022-of-00031.safetensors",
+    "model.layers.41.self_attn.k_proj.weight": "model-00021-of-00031.safetensors",
+    "model.layers.41.self_attn.o_proj.weight": "model-00021-of-00031.safetensors",
+    "model.layers.41.self_attn.q_proj.weight": "model-00021-of-00031.safetensors",
+    "model.layers.41.self_attn.v_proj.weight": "model-00021-of-00031.safetensors",
+    "model.layers.42.input_layernorm.weight": "model-00022-of-00031.safetensors",
+    "model.layers.42.mlp.down_proj.weight": "model-00022-of-00031.safetensors",
+    "model.layers.42.mlp.gate_proj.weight": "model-00022-of-00031.safetensors",
+    "model.layers.42.mlp.up_proj.weight": "model-00022-of-00031.safetensors",
+    "model.layers.42.post_attention_layernorm.weight": "model-00022-of-00031.safetensors",
+    "model.layers.42.self_attn.k_proj.weight": "model-00022-of-00031.safetensors",
+    "model.layers.42.self_attn.o_proj.weight": "model-00022-of-00031.safetensors",
+    "model.layers.42.self_attn.q_proj.weight": "model-00022-of-00031.safetensors",
+    "model.layers.42.self_attn.v_proj.weight": "model-00022-of-00031.safetensors",
+    "model.layers.43.input_layernorm.weight": "model-00023-of-00031.safetensors",
+    "model.layers.43.mlp.down_proj.weight": "model-00023-of-00031.safetensors",
+    "model.layers.43.mlp.gate_proj.weight": "model-00023-of-00031.safetensors",
+    "model.layers.43.mlp.up_proj.weight": "model-00023-of-00031.safetensors",
+    "model.layers.43.post_attention_layernorm.weight": "model-00023-of-00031.safetensors",
+    "model.layers.43.self_attn.k_proj.weight": "model-00022-of-00031.safetensors",
+    "model.layers.43.self_attn.o_proj.weight": "model-00022-of-00031.safetensors",
+    "model.layers.43.self_attn.q_proj.weight": "model-00022-of-00031.safetensors",
+    "model.layers.43.self_attn.v_proj.weight": "model-00022-of-00031.safetensors",
+    "model.layers.44.input_layernorm.weight": "model-00023-of-00031.safetensors",
+    "model.layers.44.mlp.down_proj.weight": "model-00023-of-00031.safetensors",
+    "model.layers.44.mlp.gate_proj.weight": "model-00023-of-00031.safetensors",
+    "model.layers.44.mlp.up_proj.weight": "model-00023-of-00031.safetensors",
+    "model.layers.44.post_attention_layernorm.weight": "model-00023-of-00031.safetensors",
+    "model.layers.44.self_attn.k_proj.weight": "model-00023-of-00031.safetensors",
+    "model.layers.44.self_attn.o_proj.weight": "model-00023-of-00031.safetensors",
+    "model.layers.44.self_attn.q_proj.weight": "model-00023-of-00031.safetensors",
+    "model.layers.44.self_attn.v_proj.weight": "model-00023-of-00031.safetensors",
+    "model.layers.45.input_layernorm.weight": "model-00024-of-00031.safetensors",
+    "model.layers.45.mlp.down_proj.weight": "model-00024-of-00031.safetensors",
+    "model.layers.45.mlp.gate_proj.weight": "model-00024-of-00031.safetensors",
+    "model.layers.45.mlp.up_proj.weight": "model-00024-of-00031.safetensors",
+    "model.layers.45.post_attention_layernorm.weight": "model-00024-of-00031.safetensors",
+    "model.layers.45.self_attn.k_proj.weight": "model-00023-of-00031.safetensors",
+    "model.layers.45.self_attn.o_proj.weight": "model-00023-of-00031.safetensors",
+    "model.layers.45.self_attn.q_proj.weight": "model-00023-of-00031.safetensors",
+    "model.layers.45.self_attn.v_proj.weight": "model-00023-of-00031.safetensors",
+    "model.layers.46.input_layernorm.weight": "model-00024-of-00031.safetensors",
+    "model.layers.46.mlp.down_proj.weight": "model-00024-of-00031.safetensors",
+    "model.layers.46.mlp.gate_proj.weight": "model-00024-of-00031.safetensors",
+    "model.layers.46.mlp.up_proj.weight": "model-00024-of-00031.safetensors",
+    "model.layers.46.post_attention_layernorm.weight": "model-00024-of-00031.safetensors",
+    "model.layers.46.self_attn.k_proj.weight": "model-00024-of-00031.safetensors",
+    "model.layers.46.self_attn.o_proj.weight": "model-00024-of-00031.safetensors",
+    "model.layers.46.self_attn.q_proj.weight": "model-00024-of-00031.safetensors",
+    "model.layers.46.self_attn.v_proj.weight": "model-00024-of-00031.safetensors",
+    "model.layers.47.input_layernorm.weight": "model-00025-of-00031.safetensors",
+    "model.layers.47.mlp.down_proj.weight": "model-00025-of-00031.safetensors",
+    "model.layers.47.mlp.gate_proj.weight": "model-00025-of-00031.safetensors",
+    "model.layers.47.mlp.up_proj.weight": "model-00025-of-00031.safetensors",
+    "model.layers.47.post_attention_layernorm.weight": "model-00025-of-00031.safetensors",
+    "model.layers.47.self_attn.k_proj.weight": "model-00024-of-00031.safetensors",
+    "model.layers.47.self_attn.o_proj.weight": "model-00024-of-00031.safetensors",
+    "model.layers.47.self_attn.q_proj.weight": "model-00024-of-00031.safetensors",
+    "model.layers.47.self_attn.v_proj.weight": "model-00024-of-00031.safetensors",
+    "model.layers.48.input_layernorm.weight": "model-00025-of-00031.safetensors",
+    "model.layers.48.mlp.down_proj.weight": "model-00025-of-00031.safetensors",
+    "model.layers.48.mlp.gate_proj.weight": "model-00025-of-00031.safetensors",
+    "model.layers.48.mlp.up_proj.weight": "model-00025-of-00031.safetensors",
+    "model.layers.48.post_attention_layernorm.weight": "model-00025-of-00031.safetensors",
+    "model.layers.48.self_attn.k_proj.weight": "model-00025-of-00031.safetensors",
+    "model.layers.48.self_attn.o_proj.weight": "model-00025-of-00031.safetensors",
+    "model.layers.48.self_attn.q_proj.weight": "model-00025-of-00031.safetensors",
+    "model.layers.48.self_attn.v_proj.weight": "model-00025-of-00031.safetensors",
+    "model.layers.49.input_layernorm.weight": "model-00026-of-00031.safetensors",
+    "model.layers.49.mlp.down_proj.weight": "model-00026-of-00031.safetensors",
+    "model.layers.49.mlp.gate_proj.weight": "model-00026-of-00031.safetensors",
+    "model.layers.49.mlp.up_proj.weight": "model-00026-of-00031.safetensors",
+    "model.layers.49.post_attention_layernorm.weight": "model-00026-of-00031.safetensors",
+    "model.layers.49.self_attn.k_proj.weight": "model-00025-of-00031.safetensors",
+    "model.layers.49.self_attn.o_proj.weight": "model-00025-of-00031.safetensors",
+    "model.layers.49.self_attn.q_proj.weight": "model-00025-of-00031.safetensors",
+    "model.layers.49.self_attn.v_proj.weight": "model-00025-of-00031.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00004-of-00031.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00004-of-00031.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00004-of-00031.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00004-of-00031.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00004-of-00031.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00003-of-00031.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00003-of-00031.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00003-of-00031.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00003-of-00031.safetensors",
+    "model.layers.50.input_layernorm.weight": "model-00026-of-00031.safetensors",
+    "model.layers.50.mlp.down_proj.weight": "model-00026-of-00031.safetensors",
+    "model.layers.50.mlp.gate_proj.weight": "model-00026-of-00031.safetensors",
+    "model.layers.50.mlp.up_proj.weight": "model-00026-of-00031.safetensors",
+    "model.layers.50.post_attention_layernorm.weight": "model-00026-of-00031.safetensors",
+    "model.layers.50.self_attn.k_proj.weight": "model-00026-of-00031.safetensors",
+    "model.layers.50.self_attn.o_proj.weight": "model-00026-of-00031.safetensors",
+    "model.layers.50.self_attn.q_proj.weight": "model-00026-of-00031.safetensors",
+    "model.layers.50.self_attn.v_proj.weight": "model-00026-of-00031.safetensors",
+    "model.layers.51.input_layernorm.weight": "model-00027-of-00031.safetensors",
+    "model.layers.51.mlp.down_proj.weight": "model-00027-of-00031.safetensors",
+    "model.layers.51.mlp.gate_proj.weight": "model-00027-of-00031.safetensors",
+    "model.layers.51.mlp.up_proj.weight": "model-00027-of-00031.safetensors",
+    "model.layers.51.post_attention_layernorm.weight": "model-00027-of-00031.safetensors",
+    "model.layers.51.self_attn.k_proj.weight": "model-00026-of-00031.safetensors",
+    "model.layers.51.self_attn.o_proj.weight": "model-00026-of-00031.safetensors",
+    "model.layers.51.self_attn.q_proj.weight": "model-00026-of-00031.safetensors",
+    "model.layers.51.self_attn.v_proj.weight": "model-00026-of-00031.safetensors",
+    "model.layers.52.input_layernorm.weight": "model-00027-of-00031.safetensors",
+    "model.layers.52.mlp.down_proj.weight": "model-00027-of-00031.safetensors",
+    "model.layers.52.mlp.gate_proj.weight": "model-00027-of-00031.safetensors",
+    "model.layers.52.mlp.up_proj.weight": "model-00027-of-00031.safetensors",
+    "model.layers.52.post_attention_layernorm.weight": "model-00027-of-00031.safetensors",
+    "model.layers.52.self_attn.k_proj.weight": "model-00027-of-00031.safetensors",
+    "model.layers.52.self_attn.o_proj.weight": "model-00027-of-00031.safetensors",
+    "model.layers.52.self_attn.q_proj.weight": "model-00027-of-00031.safetensors",
+    "model.layers.52.self_attn.v_proj.weight": "model-00027-of-00031.safetensors",
+    "model.layers.53.input_layernorm.weight": "model-00028-of-00031.safetensors",
+    "model.layers.53.mlp.down_proj.weight": "model-00028-of-00031.safetensors",
+    "model.layers.53.mlp.gate_proj.weight": "model-00028-of-00031.safetensors",
+    "model.layers.53.mlp.up_proj.weight": "model-00028-of-00031.safetensors",
+    "model.layers.53.post_attention_layernorm.weight": "model-00028-of-00031.safetensors",
+    "model.layers.53.self_attn.k_proj.weight": "model-00027-of-00031.safetensors",
+    "model.layers.53.self_attn.o_proj.weight": "model-00027-of-00031.safetensors",
+    "model.layers.53.self_attn.q_proj.weight": "model-00027-of-00031.safetensors",
+    "model.layers.53.self_attn.v_proj.weight": "model-00027-of-00031.safetensors",
+    "model.layers.54.input_layernorm.weight": "model-00028-of-00031.safetensors",
+    "model.layers.54.mlp.down_proj.weight": "model-00028-of-00031.safetensors",
+    "model.layers.54.mlp.gate_proj.weight": "model-00028-of-00031.safetensors",
+    "model.layers.54.mlp.up_proj.weight": "model-00028-of-00031.safetensors",
+    "model.layers.54.post_attention_layernorm.weight": "model-00028-of-00031.safetensors",
+    "model.layers.54.self_attn.k_proj.weight": "model-00028-of-00031.safetensors",
+    "model.layers.54.self_attn.o_proj.weight": "model-00028-of-00031.safetensors",
+    "model.layers.54.self_attn.q_proj.weight": "model-00028-of-00031.safetensors",
+    "model.layers.54.self_attn.v_proj.weight": "model-00028-of-00031.safetensors",
+    "model.layers.55.input_layernorm.weight": "model-00029-of-00031.safetensors",
+    "model.layers.55.mlp.down_proj.weight": "model-00029-of-00031.safetensors",
+    "model.layers.55.mlp.gate_proj.weight": "model-00029-of-00031.safetensors",
+    "model.layers.55.mlp.up_proj.weight": "model-00029-of-00031.safetensors",
+    "model.layers.55.post_attention_layernorm.weight": "model-00029-of-00031.safetensors",
+    "model.layers.55.self_attn.k_proj.weight": "model-00028-of-00031.safetensors",
+    "model.layers.55.self_attn.o_proj.weight": "model-00028-of-00031.safetensors",
+    "model.layers.55.self_attn.q_proj.weight": "model-00028-of-00031.safetensors",
+    "model.layers.55.self_attn.v_proj.weight": "model-00028-of-00031.safetensors",
+    "model.layers.56.input_layernorm.weight": "model-00029-of-00031.safetensors",
+    "model.layers.56.mlp.down_proj.weight": "model-00029-of-00031.safetensors",
+    "model.layers.56.mlp.gate_proj.weight": "model-00029-of-00031.safetensors",
+    "model.layers.56.mlp.up_proj.weight": "model-00029-of-00031.safetensors",
+    "model.layers.56.post_attention_layernorm.weight": "model-00029-of-00031.safetensors",
+    "model.layers.56.self_attn.k_proj.weight": "model-00029-of-00031.safetensors",
+    "model.layers.56.self_attn.o_proj.weight": "model-00029-of-00031.safetensors",
+    "model.layers.56.self_attn.q_proj.weight": "model-00029-of-00031.safetensors",
+    "model.layers.56.self_attn.v_proj.weight": "model-00029-of-00031.safetensors",
+    "model.layers.57.input_layernorm.weight": "model-00030-of-00031.safetensors",
+    "model.layers.57.mlp.down_proj.weight": "model-00030-of-00031.safetensors",
+    "model.layers.57.mlp.gate_proj.weight": "model-00030-of-00031.safetensors",
+    "model.layers.57.mlp.up_proj.weight": "model-00030-of-00031.safetensors",
+    "model.layers.57.post_attention_layernorm.weight": "model-00030-of-00031.safetensors",
+    "model.layers.57.self_attn.k_proj.weight": "model-00029-of-00031.safetensors",
+    "model.layers.57.self_attn.o_proj.weight": "model-00029-of-00031.safetensors",
+    "model.layers.57.self_attn.q_proj.weight": "model-00029-of-00031.safetensors",
+    "model.layers.57.self_attn.v_proj.weight": "model-00029-of-00031.safetensors",
+    "model.layers.58.input_layernorm.weight": "model-00030-of-00031.safetensors",
+    "model.layers.58.mlp.down_proj.weight": "model-00030-of-00031.safetensors",
+    "model.layers.58.mlp.gate_proj.weight": "model-00030-of-00031.safetensors",
+    "model.layers.58.mlp.up_proj.weight": "model-00030-of-00031.safetensors",
+    "model.layers.58.post_attention_layernorm.weight": "model-00030-of-00031.safetensors",
+    "model.layers.58.self_attn.k_proj.weight": "model-00030-of-00031.safetensors",
+    "model.layers.58.self_attn.o_proj.weight": "model-00030-of-00031.safetensors",
+    "model.layers.58.self_attn.q_proj.weight": "model-00030-of-00031.safetensors",
+    "model.layers.58.self_attn.v_proj.weight": "model-00030-of-00031.safetensors",
+    "model.layers.59.input_layernorm.weight": "model-00031-of-00031.safetensors",
+    "model.layers.59.mlp.down_proj.weight": "model-00031-of-00031.safetensors",
+    "model.layers.59.mlp.gate_proj.weight": "model-00031-of-00031.safetensors",
+    "model.layers.59.mlp.up_proj.weight": "model-00031-of-00031.safetensors",
+    "model.layers.59.post_attention_layernorm.weight": "model-00031-of-00031.safetensors",
+    "model.layers.59.self_attn.k_proj.weight": "model-00030-of-00031.safetensors",
+    "model.layers.59.self_attn.o_proj.weight": "model-00030-of-00031.safetensors",
+    "model.layers.59.self_attn.q_proj.weight": "model-00030-of-00031.safetensors",
+    "model.layers.59.self_attn.v_proj.weight": "model-00030-of-00031.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00004-of-00031.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00004-of-00031.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00004-of-00031.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00004-of-00031.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00004-of-00031.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00004-of-00031.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00004-of-00031.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00004-of-00031.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00004-of-00031.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00005-of-00031.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00005-of-00031.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00005-of-00031.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00005-of-00031.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00005-of-00031.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00004-of-00031.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00004-of-00031.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00004-of-00031.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00004-of-00031.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00005-of-00031.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00005-of-00031.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00005-of-00031.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00005-of-00031.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00005-of-00031.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00005-of-00031.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00005-of-00031.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00005-of-00031.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00005-of-00031.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00006-of-00031.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00006-of-00031.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00006-of-00031.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00006-of-00031.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00006-of-00031.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00005-of-00031.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00005-of-00031.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00005-of-00031.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00005-of-00031.safetensors",
+    "model.norm.weight": "model-00031-of-00031.safetensors"
+  }
+}