Add files using upload-large-folder tool

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,42 @@
+---
+license: other
+library_name: transformers
+pipeline_tag: text-generation
+tags:
+  - computer-use
+  - tool-use
+  - reinforcement-learning
+---
+
+# Computer-9a
+
+Computer-9a is the step-60 policy from an online terminal reinforcement-learning experiment. It is a full-weight bfloat16 causal language model exported from the exact retained FSDP checkpoint used for evaluation.
+
+The training environment presented a `bash` tool through textual `<tool name="bash">...</tool>` calls and returned textual `<tool_result name="bash">...</tool_result>` observations. The model was optimized on automatically verified terminal tasks while regularizing against a conversational parent policy.
+
+## Prompt format
+
+The experiment used an explicit transcript header and role markers:
+
+```text
+**User:** ...
+
+**Model C:** ...
+
+**Environment:** <tool_result name="bash">...</tool_result>
+```
+
+Tool execution is not built into the weights. Applications must parse tool calls, execute them in an appropriately isolated sandbox, append the tool result, and generate the next model turn.
+
+## Safety and limitations
+
+This is a research checkpoint. It can emit shell commands and should only be connected to a strongly isolated, least-privilege environment. It may make destructive requests, hallucinate command effects, contradict visible tool traces, or fail exact multi-stage tasks. Do not give it credentials, unrestricted network access, or access to valuable files.
+
+The model often produces terse answers. Its terminal-task performance and fixed-probe performance are noisy, and this checkpoint should not be interpreted as a generally reliable computer-use agent.
+
+## Checkpoint identity
+
+- Training step: 60
+- Export dtype: bfloat16
+- Source checkpoint: `retained-checkpoint-60`
+- Export format: sharded safetensors

config.json

@@ -0,0 +1,35 @@
+{
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 128000,
+  "eos_token_id": 128001,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 8192,
+  "initializer_range": 0.02,
+  "intermediate_size": 28672,
+  "max_position_embeddings": 131072,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 64,
+  "num_hidden_layers": 80,
+  "num_key_value_heads": 8,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": {
+    "factor": 8.0,
+    "high_freq_factor": 4.0,
+    "low_freq_factor": 1.0,
+    "original_max_position_embeddings": 8192,
+    "rope_type": "llama3"
+  },
+  "rope_theta": 500000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.55.2",
+  "use_cache": true,
+  "vocab_size": 128256
+}

generation_config.json

@@ -0,0 +1,9 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 128000,
+  "do_sample": true,
+  "eos_token_id": 128001,
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.55.2"
+}

model-00001-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6beae6ee3ae24858033026d668a4a4f3cb3c860690c23dd8255ac72ca4f9f7f4
+size 2403336800

model-00002-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9f5b63c7ecbe1b36bcad1329668bb578612068e6381c6ef3e7a0213f8690c28
+size 2181072024

model-00003-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b46c1be5cbdc65db4b03554720042b94ef8386e73ab1d2aa4c5b8df389fe429
+size 2181072024

model-00005-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4fe034d480d2e22c9309fb80e723009bc4395cf0d424b6d341df5dd18dff83fa
+size 2181072024

model-00006-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e9c1eac70a5d6c4bcc74b46a5a731a9699697e4a3672c5d6106d151cb18318e
+size 2181072024

model-00008-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1da81a6badfb60ada87ef0c4e6a9c76781676521a27ae5f3067648b577661be9
+size 2181072024

model-00009-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed40a032a5c4cb8c1f7961d33081f5765b6a2ee003384bbec2af2427d04f2ecd
+size 2181072040

model-00011-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:41d8ac8951baa38b6dc9faa8035731b3e522de83a5324176ee48bd995ba03ef7
+size 2181072032

model-00012-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5264a2468c37ce17833c9e50bcd1b2f18f362ee8176c475d365580210b68904f
+size 2181072032

model-00014-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ca48dde8a8bc5f0dbfcacf374eb040ee1ff67c72249e5a6de8ca4535b99874e
+size 2181072032

model-00015-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e2de41dd63cf03de42c0282ddb59ff98315749bf775ba6ea4737c5a3b3399d98
+size 2181072032

model-00017-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98d674652894dbc407724a5e4f01e3fbba5106160376468ec43902be1ed20f88
+size 2181072032

model-00018-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:23a74988b3b14fbff9613047b46d81969cbf70c58930667ed7d3c7a33037801e
+size 2181072032

model-00020-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0191d0e5b45292bce8492926a455c6a16fadfa5d9b14f3e4612b97aa52a3b831
+size 2181072032

model-00021-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85af3bf6350960ff8e5a4f874738f252d6360038000c5c9194b4e33f7c2132d8
+size 2181072032

model-00023-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:204e642d72cbc7a9578f0351136bf95def360209249ce4be95350e0db5940a0b
+size 2181072032

model-00024-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5459434b50e252e519feb29bf6e08d94d6dacb49adc6aee154e857eb7f32fc4a
+size 2181072032

model-00026-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0ccbd39167079dae8ea4539428a17bc490334316c854efbe787643add7ff851
+size 2181072032

model-00027-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0828157f10ee09857dc7b5be8703cbc44812a706e1353b8731553e8be9161f87
+size 2181072032

model-00029-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:255c9f82ee3fa86ce9a59b4fec7d2a3fdbcae2da2923c6d880deebfd7ce29cd6
+size 2181072032

model-00030-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87d16cf5b62c015c04711445b46e69ee29e25a1e7697f6676230ec9c412848f5
+size 2181072032

model-00032-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b1cffe47284325fa87349dc14bcba5b9618746554d25a3f89fbfbbbb63463da
+size 2181072032

model-00033-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c849f2b7dfc80409859b7d6d11375a3da6997adf3f673d21850bb361f9c3ae4
+size 2181072032

model-00035-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:461ca21f4bd0ea295581753a4ff7eb2e400fc7109948f4148017cfcc966809e3
+size 2181072032

model-00036-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27bc0ed1e978216b331982690bab1a10bbab9b39345cf96a66f93e063a6ea26d
+size 2181072032

model-00038-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6488504a904d748f9cb777077a44a05657fae1bda06e27c2fd205b83d0f5d242
+size 2181072032

model-00039-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34e2a5ad5b075c53842ff66d01904d1d22cfc7f66fcfe5307d606bae53c8d6bb
+size 2181072032

model-00041-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e6192f8380117117b2bb2283a9f7ed21df65cc3a04ee71f3cea81698bf9da35
+size 2181072032

model-00042-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea1abfbf9a53fd1e162e79909c5ebe759eee0fce64ed97b618c647987d407815
+size 2181072032

model-00044-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ee665c969ce1a0f566ae0c625614f5cad828a77f618e09b22dfcdda7f3973f2
+size 2181072032

model-00045-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ab53c860f385e6c40624e32061e0672b6dd69d030d1362cdc5916659665263b
+size 2181072032

model-00047-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29b4b064615e89d34b85ac001d70ec63a3ff618750c58ff2178430a45fa7d26c
+size 2181072032

model-00048-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8b93eb6c52d5d9b6041758211a546a34fdfcd37b83a107394f5318db75f66de
+size 2181072032

model-00050-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:65edd1c86a38ebfacc35b349212e2c8e1d5968874684e3a74adff10c7d1aa0e6
+size 2181072032

model-00051-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:720ea7ef2437a413fe7bf4b73e2e36647d1baad15987fcd7ae5287123b8a40fd
+size 2181072032

model-00053-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e661f913d1ec70401a1581dd7bb4b53f175433d1b339e41eb17bdc44e044964
+size 2181072032

model-00054-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab4f56db08f575a67efe3c140d8509cc75e4991bf78140c011f1301fbe498d7c
+size 2181072032

model-00056-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5fab12a99da87185a5ecd35b86be361c8dac73758ddd41f750b73d06f452662
+size 2181072032

model-00057-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:861ea14ed236709a0f5dbbe048fb2e2749ca94ac9de08605ade0927da2fdb224
+size 2181072032

model-00059-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8568be70792cde01340c5067eea3158af273e42e8739a61d49c4e49c323e9d31
+size 2181072032

model-00060-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6ae97a2143d6752d367848e86de0e666733449e99db8c98341a759668abe54d
+size 2181072032

model-00061-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b555afe13e3d2dec223a49a00af0b10a6c9413022ace31d9baa279c641c95d47
+size 2181121504

model-00062-of-00062.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7ffd327d6d3a7e988b3af0e73c6ba40ae57f62d10f215338cef962918f78d2d
+size 2101346432

model.safetensors.index.json

@@ -0,0 +1,730 @@
+{
+  "metadata": {
+    "total_size": 141107412992
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00062-of-00062.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00062.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00002-of-00062.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00002-of-00062.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00002-of-00062.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00002-of-00062.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00002-of-00062.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00062.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00062.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00062.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00062.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00003-of-00062.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00003-of-00062.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00002-of-00062.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00003-of-00062.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00003-of-00062.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00002-of-00062.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00002-of-00062.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00002-of-00062.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00002-of-00062.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00010-of-00062.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00010-of-00062.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00009-of-00062.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00009-of-00062.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00010-of-00062.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00009-of-00062.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00009-of-00062.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00009-of-00062.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00009-of-00062.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00010-of-00062.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00010-of-00062.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00010-of-00062.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00010-of-00062.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00010-of-00062.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00010-of-00062.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00010-of-00062.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00010-of-00062.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00010-of-00062.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00011-of-00062.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00011-of-00062.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00011-of-00062.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00011-of-00062.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00011-of-00062.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00010-of-00062.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00010-of-00062.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00010-of-00062.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00010-of-00062.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00012-of-00062.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00012-of-00062.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00011-of-00062.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00012-of-00062.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00012-of-00062.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00011-of-00062.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00011-of-00062.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00011-of-00062.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00011-of-00062.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00013-of-00062.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00013-of-00062.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00012-of-00062.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00012-of-00062.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00013-of-00062.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00012-of-00062.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00012-of-00062.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00012-of-00062.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00012-of-00062.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00013-of-00062.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00013-of-00062.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00013-of-00062.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00013-of-00062.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00013-of-00062.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00013-of-00062.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00013-of-00062.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00013-of-00062.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00013-of-00062.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00014-of-00062.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00014-of-00062.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00014-of-00062.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00014-of-00062.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00014-of-00062.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00013-of-00062.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00013-of-00062.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00013-of-00062.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00013-of-00062.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00015-of-00062.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00015-of-00062.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00014-of-00062.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00015-of-00062.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00015-of-00062.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00014-of-00062.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00014-of-00062.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00014-of-00062.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00014-of-00062.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00016-of-00062.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00016-of-00062.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00015-of-00062.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00015-of-00062.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00016-of-00062.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00015-of-00062.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00015-of-00062.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00015-of-00062.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00015-of-00062.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00016-of-00062.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00016-of-00062.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00016-of-00062.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00016-of-00062.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00016-of-00062.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00016-of-00062.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00016-of-00062.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00016-of-00062.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00016-of-00062.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00004-of-00062.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00004-of-00062.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00003-of-00062.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00003-of-00062.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00004-of-00062.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00003-of-00062.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00003-of-00062.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00003-of-00062.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00003-of-00062.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00017-of-00062.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00017-of-00062.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00017-of-00062.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00017-of-00062.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00017-of-00062.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00016-of-00062.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00016-of-00062.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00016-of-00062.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00016-of-00062.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00018-of-00062.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00018-of-00062.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00017-of-00062.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00018-of-00062.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00018-of-00062.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00017-of-00062.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00017-of-00062.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00017-of-00062.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00017-of-00062.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00019-of-00062.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00019-of-00062.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00018-of-00062.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00018-of-00062.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00019-of-00062.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00018-of-00062.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00018-of-00062.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00018-of-00062.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00018-of-00062.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00019-of-00062.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00019-of-00062.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00019-of-00062.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00019-of-00062.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00019-of-00062.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00019-of-00062.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00019-of-00062.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00019-of-00062.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00019-of-00062.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00020-of-00062.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00020-of-00062.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00020-of-00062.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00020-of-00062.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00020-of-00062.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00019-of-00062.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00019-of-00062.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00019-of-00062.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00019-of-00062.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00021-of-00062.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00021-of-00062.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00020-of-00062.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00021-of-00062.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00021-of-00062.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00020-of-00062.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00020-of-00062.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00020-of-00062.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00020-of-00062.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00022-of-00062.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00022-of-00062.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00021-of-00062.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00021-of-00062.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00022-of-00062.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00021-of-00062.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00021-of-00062.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00021-of-00062.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00021-of-00062.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00022-of-00062.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00022-of-00062.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00022-of-00062.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00022-of-00062.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00022-of-00062.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00022-of-00062.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00022-of-00062.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00022-of-00062.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00022-of-00062.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00023-of-00062.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00023-of-00062.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00023-of-00062.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00023-of-00062.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00023-of-00062.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00022-of-00062.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00022-of-00062.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00022-of-00062.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00022-of-00062.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00024-of-00062.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00024-of-00062.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00023-of-00062.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00024-of-00062.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00024-of-00062.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00023-of-00062.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00023-of-00062.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00023-of-00062.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00023-of-00062.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00004-of-00062.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00004-of-00062.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00004-of-00062.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00004-of-00062.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00004-of-00062.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00004-of-00062.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00004-of-00062.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00004-of-00062.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00004-of-00062.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00025-of-00062.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00025-of-00062.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00024-of-00062.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00024-of-00062.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00025-of-00062.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00024-of-00062.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00024-of-00062.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00024-of-00062.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00024-of-00062.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00025-of-00062.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00025-of-00062.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00025-of-00062.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00025-of-00062.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00025-of-00062.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00025-of-00062.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00025-of-00062.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00025-of-00062.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00025-of-00062.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00026-of-00062.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-00026-of-00062.safetensors",
+    "model.layers.32.mlp.gate_proj.weight": "model-00026-of-00062.safetensors",
+    "model.layers.32.mlp.up_proj.weight": "model-00026-of-00062.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00026-of-00062.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00025-of-00062.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00025-of-00062.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00025-of-00062.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00025-of-00062.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00027-of-00062.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-00027-of-00062.safetensors",
+    "model.layers.33.mlp.gate_proj.weight": "model-00026-of-00062.safetensors",
+    "model.layers.33.mlp.up_proj.weight": "model-00027-of-00062.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00027-of-00062.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00026-of-00062.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00026-of-00062.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00026-of-00062.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00026-of-00062.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00028-of-00062.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-00028-of-00062.safetensors",
+    "model.layers.34.mlp.gate_proj.weight": "model-00027-of-00062.safetensors",
+    "model.layers.34.mlp.up_proj.weight": "model-00027-of-00062.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00028-of-00062.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00027-of-00062.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00027-of-00062.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00027-of-00062.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00027-of-00062.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00028-of-00062.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-00028-of-00062.safetensors",
+    "model.layers.35.mlp.gate_proj.weight": "model-00028-of-00062.safetensors",
+    "model.layers.35.mlp.up_proj.weight": "model-00028-of-00062.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00028-of-00062.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00028-of-00062.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00028-of-00062.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00028-of-00062.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00028-of-00062.safetensors",
+    "model.layers.36.input_layernorm.weight": "model-00029-of-00062.safetensors",
+    "model.layers.36.mlp.down_proj.weight": "model-00029-of-00062.safetensors",
+    "model.layers.36.mlp.gate_proj.weight": "model-00029-of-00062.safetensors",
+    "model.layers.36.mlp.up_proj.weight": "model-00029-of-00062.safetensors",
+    "model.layers.36.post_attention_layernorm.weight": "model-00029-of-00062.safetensors",
+    "model.layers.36.self_attn.k_proj.weight": "model-00028-of-00062.safetensors",
+    "model.layers.36.self_attn.o_proj.weight": "model-00028-of-00062.safetensors",
+    "model.layers.36.self_attn.q_proj.weight": "model-00028-of-00062.safetensors",
+    "model.layers.36.self_attn.v_proj.weight": "model-00028-of-00062.safetensors",
+    "model.layers.37.input_layernorm.weight": "model-00030-of-00062.safetensors",
+    "model.layers.37.mlp.down_proj.weight": "model-00030-of-00062.safetensors",
+    "model.layers.37.mlp.gate_proj.weight": "model-00029-of-00062.safetensors",
+    "model.layers.37.mlp.up_proj.weight": "model-00030-of-00062.safetensors",
+    "model.layers.37.post_attention_layernorm.weight": "model-00030-of-00062.safetensors",
+    "model.layers.37.self_attn.k_proj.weight": "model-00029-of-00062.safetensors",
+    "model.layers.37.self_attn.o_proj.weight": "model-00029-of-00062.safetensors",
+    "model.layers.37.self_attn.q_proj.weight": "model-00029-of-00062.safetensors",
+    "model.layers.37.self_attn.v_proj.weight": "model-00029-of-00062.safetensors",
+    "model.layers.38.input_layernorm.weight": "model-00031-of-00062.safetensors",
+    "model.layers.38.mlp.down_proj.weight": "model-00031-of-00062.safetensors",
+    "model.layers.38.mlp.gate_proj.weight": "model-00030-of-00062.safetensors",
+    "model.layers.38.mlp.up_proj.weight": "model-00030-of-00062.safetensors",
+    "model.layers.38.post_attention_layernorm.weight": "model-00031-of-00062.safetensors",
+    "model.layers.38.self_attn.k_proj.weight": "model-00030-of-00062.safetensors",
+    "model.layers.38.self_attn.o_proj.weight": "model-00030-of-00062.safetensors",
+    "model.layers.38.self_attn.q_proj.weight": "model-00030-of-00062.safetensors",
+    "model.layers.38.self_attn.v_proj.weight": "model-00030-of-00062.safetensors",
+    "model.layers.39.input_layernorm.weight": "model-00031-of-00062.safetensors",
+    "model.layers.39.mlp.down_proj.weight": "model-00031-of-00062.safetensors",
+    "model.layers.39.mlp.gate_proj.weight": "model-00031-of-00062.safetensors",
+    "model.layers.39.mlp.up_proj.weight": "model-00031-of-00062.safetensors",
+    "model.layers.39.post_attention_layernorm.weight": "model-00031-of-00062.safetensors",
+    "model.layers.39.self_attn.k_proj.weight": "model-00031-of-00062.safetensors",
+    "model.layers.39.self_attn.o_proj.weight": "model-00031-of-00062.safetensors",
+    "model.layers.39.self_attn.q_proj.weight": "model-00031-of-00062.safetensors",
+    "model.layers.39.self_attn.v_proj.weight": "model-00031-of-00062.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00005-of-00062.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00005-of-00062.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00005-of-00062.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00005-of-00062.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00005-of-00062.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00004-of-00062.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00004-of-00062.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00004-of-00062.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00004-of-00062.safetensors",
+    "model.layers.40.input_layernorm.weight": "model-00032-of-00062.safetensors",
+    "model.layers.40.mlp.down_proj.weight": "model-00032-of-00062.safetensors",
+    "model.layers.40.mlp.gate_proj.weight": "model-00032-of-00062.safetensors",
+    "model.layers.40.mlp.up_proj.weight": "model-00032-of-00062.safetensors",
+    "model.layers.40.post_attention_layernorm.weight": "model-00032-of-00062.safetensors",
+    "model.layers.40.self_attn.k_proj.weight": "model-00031-of-00062.safetensors",
+    "model.layers.40.self_attn.o_proj.weight": "model-00031-of-00062.safetensors",
+    "model.layers.40.self_attn.q_proj.weight": "model-00031-of-00062.safetensors",
+    "model.layers.40.self_attn.v_proj.weight": "model-00031-of-00062.safetensors",
+    "model.layers.41.input_layernorm.weight": "model-00033-of-00062.safetensors",
+    "model.layers.41.mlp.down_proj.weight": "model-00033-of-00062.safetensors",
+    "model.layers.41.mlp.gate_proj.weight": "model-00032-of-00062.safetensors",
+    "model.layers.41.mlp.up_proj.weight": "model-00033-of-00062.safetensors",
+    "model.layers.41.post_attention_layernorm.weight": "model-00033-of-00062.safetensors",
+    "model.layers.41.self_attn.k_proj.weight": "model-00032-of-00062.safetensors",
+    "model.layers.41.self_attn.o_proj.weight": "model-00032-of-00062.safetensors",
+    "model.layers.41.self_attn.q_proj.weight": "model-00032-of-00062.safetensors",
+    "model.layers.41.self_attn.v_proj.weight": "model-00032-of-00062.safetensors",
+    "model.layers.42.input_layernorm.weight": "model-00034-of-00062.safetensors",
+    "model.layers.42.mlp.down_proj.weight": "model-00034-of-00062.safetensors",
+    "model.layers.42.mlp.gate_proj.weight": "model-00033-of-00062.safetensors",
+    "model.layers.42.mlp.up_proj.weight": "model-00033-of-00062.safetensors",
+    "model.layers.42.post_attention_layernorm.weight": "model-00034-of-00062.safetensors",
+    "model.layers.42.self_attn.k_proj.weight": "model-00033-of-00062.safetensors",
+    "model.layers.42.self_attn.o_proj.weight": "model-00033-of-00062.safetensors",
+    "model.layers.42.self_attn.q_proj.weight": "model-00033-of-00062.safetensors",
+    "model.layers.42.self_attn.v_proj.weight": "model-00033-of-00062.safetensors",
+    "model.layers.43.input_layernorm.weight": "model-00034-of-00062.safetensors",
+    "model.layers.43.mlp.down_proj.weight": "model-00034-of-00062.safetensors",
+    "model.layers.43.mlp.gate_proj.weight": "model-00034-of-00062.safetensors",
+    "model.layers.43.mlp.up_proj.weight": "model-00034-of-00062.safetensors",
+    "model.layers.43.post_attention_layernorm.weight": "model-00034-of-00062.safetensors",
+    "model.layers.43.self_attn.k_proj.weight": "model-00034-of-00062.safetensors",
+    "model.layers.43.self_attn.o_proj.weight": "model-00034-of-00062.safetensors",
+    "model.layers.43.self_attn.q_proj.weight": "model-00034-of-00062.safetensors",
+    "model.layers.43.self_attn.v_proj.weight": "model-00034-of-00062.safetensors",
+    "model.layers.44.input_layernorm.weight": "model-00035-of-00062.safetensors",
+    "model.layers.44.mlp.down_proj.weight": "model-00035-of-00062.safetensors",
+    "model.layers.44.mlp.gate_proj.weight": "model-00035-of-00062.safetensors",
+    "model.layers.44.mlp.up_proj.weight": "model-00035-of-00062.safetensors",
+    "model.layers.44.post_attention_layernorm.weight": "model-00035-of-00062.safetensors",
+    "model.layers.44.self_attn.k_proj.weight": "model-00034-of-00062.safetensors",
+    "model.layers.44.self_attn.o_proj.weight": "model-00034-of-00062.safetensors",
+    "model.layers.44.self_attn.q_proj.weight": "model-00034-of-00062.safetensors",
+    "model.layers.44.self_attn.v_proj.weight": "model-00034-of-00062.safetensors",
+    "model.layers.45.input_layernorm.weight": "model-00036-of-00062.safetensors",
+    "model.layers.45.mlp.down_proj.weight": "model-00036-of-00062.safetensors",
+    "model.layers.45.mlp.gate_proj.weight": "model-00035-of-00062.safetensors",
+    "model.layers.45.mlp.up_proj.weight": "model-00036-of-00062.safetensors",
+    "model.layers.45.post_attention_layernorm.weight": "model-00036-of-00062.safetensors",
+    "model.layers.45.self_attn.k_proj.weight": "model-00035-of-00062.safetensors",
+    "model.layers.45.self_attn.o_proj.weight": "model-00035-of-00062.safetensors",
+    "model.layers.45.self_attn.q_proj.weight": "model-00035-of-00062.safetensors",
+    "model.layers.45.self_attn.v_proj.weight": "model-00035-of-00062.safetensors",
+    "model.layers.46.input_layernorm.weight": "model-00037-of-00062.safetensors",
+    "model.layers.46.mlp.down_proj.weight": "model-00037-of-00062.safetensors",
+    "model.layers.46.mlp.gate_proj.weight": "model-00036-of-00062.safetensors",
+    "model.layers.46.mlp.up_proj.weight": "model-00036-of-00062.safetensors",
+    "model.layers.46.post_attention_layernorm.weight": "model-00037-of-00062.safetensors",
+    "model.layers.46.self_attn.k_proj.weight": "model-00036-of-00062.safetensors",
+    "model.layers.46.self_attn.o_proj.weight": "model-00036-of-00062.safetensors",
+    "model.layers.46.self_attn.q_proj.weight": "model-00036-of-00062.safetensors",
+    "model.layers.46.self_attn.v_proj.weight": "model-00036-of-00062.safetensors",
+    "model.layers.47.input_layernorm.weight": "model-00037-of-00062.safetensors",
+    "model.layers.47.mlp.down_proj.weight": "model-00037-of-00062.safetensors",
+    "model.layers.47.mlp.gate_proj.weight": "model-00037-of-00062.safetensors",
+    "model.layers.47.mlp.up_proj.weight": "model-00037-of-00062.safetensors",
+    "model.layers.47.post_attention_layernorm.weight": "model-00037-of-00062.safetensors",
+    "model.layers.47.self_attn.k_proj.weight": "model-00037-of-00062.safetensors",
+    "model.layers.47.self_attn.o_proj.weight": "model-00037-of-00062.safetensors",
+    "model.layers.47.self_attn.q_proj.weight": "model-00037-of-00062.safetensors",
+    "model.layers.47.self_attn.v_proj.weight": "model-00037-of-00062.safetensors",
+    "model.layers.48.input_layernorm.weight": "model-00038-of-00062.safetensors",
+    "model.layers.48.mlp.down_proj.weight": "model-00038-of-00062.safetensors",
+    "model.layers.48.mlp.gate_proj.weight": "model-00038-of-00062.safetensors",
+    "model.layers.48.mlp.up_proj.weight": "model-00038-of-00062.safetensors",
+    "model.layers.48.post_attention_layernorm.weight": "model-00038-of-00062.safetensors",
+    "model.layers.48.self_attn.k_proj.weight": "model-00037-of-00062.safetensors",
+    "model.layers.48.self_attn.o_proj.weight": "model-00037-of-00062.safetensors",
+    "model.layers.48.self_attn.q_proj.weight": "model-00037-of-00062.safetensors",
+    "model.layers.48.self_attn.v_proj.weight": "model-00037-of-00062.safetensors",
+    "model.layers.49.input_layernorm.weight": "model-00039-of-00062.safetensors",
+    "model.layers.49.mlp.down_proj.weight": "model-00039-of-00062.safetensors",
+    "model.layers.49.mlp.gate_proj.weight": "model-00038-of-00062.safetensors",
+    "model.layers.49.mlp.up_proj.weight": "model-00039-of-00062.safetensors",
+    "model.layers.49.post_attention_layernorm.weight": "model-00039-of-00062.safetensors",
+    "model.layers.49.self_attn.k_proj.weight": "model-00038-of-00062.safetensors",
+    "model.layers.49.self_attn.o_proj.weight": "model-00038-of-00062.safetensors",
+    "model.layers.49.self_attn.q_proj.weight": "model-00038-of-00062.safetensors",
+    "model.layers.49.self_attn.v_proj.weight": "model-00038-of-00062.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00006-of-00062.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00006-of-00062.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00005-of-00062.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00006-of-00062.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00006-of-00062.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00005-of-00062.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00005-of-00062.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00005-of-00062.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00005-of-00062.safetensors",
+    "model.layers.50.input_layernorm.weight": "model-00040-of-00062.safetensors",
+    "model.layers.50.mlp.down_proj.weight": "model-00040-of-00062.safetensors",
+    "model.layers.50.mlp.gate_proj.weight": "model-00039-of-00062.safetensors",
+    "model.layers.50.mlp.up_proj.weight": "model-00039-of-00062.safetensors",
+    "model.layers.50.post_attention_layernorm.weight": "model-00040-of-00062.safetensors",
+    "model.layers.50.self_attn.k_proj.weight": "model-00039-of-00062.safetensors",
+    "model.layers.50.self_attn.o_proj.weight": "model-00039-of-00062.safetensors",
+    "model.layers.50.self_attn.q_proj.weight": "model-00039-of-00062.safetensors",
+    "model.layers.50.self_attn.v_proj.weight": "model-00039-of-00062.safetensors",
+    "model.layers.51.input_layernorm.weight": "model-00040-of-00062.safetensors",
+    "model.layers.51.mlp.down_proj.weight": "model-00040-of-00062.safetensors",
+    "model.layers.51.mlp.gate_proj.weight": "model-00040-of-00062.safetensors",
+    "model.layers.51.mlp.up_proj.weight": "model-00040-of-00062.safetensors",
+    "model.layers.51.post_attention_layernorm.weight": "model-00040-of-00062.safetensors",
+    "model.layers.51.self_attn.k_proj.weight": "model-00040-of-00062.safetensors",
+    "model.layers.51.self_attn.o_proj.weight": "model-00040-of-00062.safetensors",
+    "model.layers.51.self_attn.q_proj.weight": "model-00040-of-00062.safetensors",
+    "model.layers.51.self_attn.v_proj.weight": "model-00040-of-00062.safetensors",
+    "model.layers.52.input_layernorm.weight": "model-00041-of-00062.safetensors",
+    "model.layers.52.mlp.down_proj.weight": "model-00041-of-00062.safetensors",
+    "model.layers.52.mlp.gate_proj.weight": "model-00041-of-00062.safetensors",
+    "model.layers.52.mlp.up_proj.weight": "model-00041-of-00062.safetensors",
+    "model.layers.52.post_attention_layernorm.weight": "model-00041-of-00062.safetensors",
+    "model.layers.52.self_attn.k_proj.weight": "model-00040-of-00062.safetensors",
+    "model.layers.52.self_attn.o_proj.weight": "model-00040-of-00062.safetensors",
+    "model.layers.52.self_attn.q_proj.weight": "model-00040-of-00062.safetensors",
+    "model.layers.52.self_attn.v_proj.weight": "model-00040-of-00062.safetensors",
+    "model.layers.53.input_layernorm.weight": "model-00042-of-00062.safetensors",
+    "model.layers.53.mlp.down_proj.weight": "model-00042-of-00062.safetensors",
+    "model.layers.53.mlp.gate_proj.weight": "model-00041-of-00062.safetensors",
+    "model.layers.53.mlp.up_proj.weight": "model-00042-of-00062.safetensors",
+    "model.layers.53.post_attention_layernorm.weight": "model-00042-of-00062.safetensors",
+    "model.layers.53.self_attn.k_proj.weight": "model-00041-of-00062.safetensors",
+    "model.layers.53.self_attn.o_proj.weight": "model-00041-of-00062.safetensors",
+    "model.layers.53.self_attn.q_proj.weight": "model-00041-of-00062.safetensors",
+    "model.layers.53.self_attn.v_proj.weight": "model-00041-of-00062.safetensors",
+    "model.layers.54.input_layernorm.weight": "model-00043-of-00062.safetensors",
+    "model.layers.54.mlp.down_proj.weight": "model-00043-of-00062.safetensors",
+    "model.layers.54.mlp.gate_proj.weight": "model-00042-of-00062.safetensors",
+    "model.layers.54.mlp.up_proj.weight": "model-00042-of-00062.safetensors",
+    "model.layers.54.post_attention_layernorm.weight": "model-00043-of-00062.safetensors",
+    "model.layers.54.self_attn.k_proj.weight": "model-00042-of-00062.safetensors",
+    "model.layers.54.self_attn.o_proj.weight": "model-00042-of-00062.safetensors",
+    "model.layers.54.self_attn.q_proj.weight": "model-00042-of-00062.safetensors",
+    "model.layers.54.self_attn.v_proj.weight": "model-00042-of-00062.safetensors",
+    "model.layers.55.input_layernorm.weight": "model-00043-of-00062.safetensors",
+    "model.layers.55.mlp.down_proj.weight": "model-00043-of-00062.safetensors",
+    "model.layers.55.mlp.gate_proj.weight": "model-00043-of-00062.safetensors",
+    "model.layers.55.mlp.up_proj.weight": "model-00043-of-00062.safetensors",
+    "model.layers.55.post_attention_layernorm.weight": "model-00043-of-00062.safetensors",
+    "model.layers.55.self_attn.k_proj.weight": "model-00043-of-00062.safetensors",
+    "model.layers.55.self_attn.o_proj.weight": "model-00043-of-00062.safetensors",
+    "model.layers.55.self_attn.q_proj.weight": "model-00043-of-00062.safetensors",
+    "model.layers.55.self_attn.v_proj.weight": "model-00043-of-00062.safetensors",
+    "model.layers.56.input_layernorm.weight": "model-00044-of-00062.safetensors",
+    "model.layers.56.mlp.down_proj.weight": "model-00044-of-00062.safetensors",
+    "model.layers.56.mlp.gate_proj.weight": "model-00044-of-00062.safetensors",
+    "model.layers.56.mlp.up_proj.weight": "model-00044-of-00062.safetensors",
+    "model.layers.56.post_attention_layernorm.weight": "model-00044-of-00062.safetensors",
+    "model.layers.56.self_attn.k_proj.weight": "model-00043-of-00062.safetensors",
+    "model.layers.56.self_attn.o_proj.weight": "model-00043-of-00062.safetensors",
+    "model.layers.56.self_attn.q_proj.weight": "model-00043-of-00062.safetensors",
+    "model.layers.56.self_attn.v_proj.weight": "model-00043-of-00062.safetensors",
+    "model.layers.57.input_layernorm.weight": "model-00045-of-00062.safetensors",
+    "model.layers.57.mlp.down_proj.weight": "model-00045-of-00062.safetensors",
+    "model.layers.57.mlp.gate_proj.weight": "model-00044-of-00062.safetensors",
+    "model.layers.57.mlp.up_proj.weight": "model-00045-of-00062.safetensors",
+    "model.layers.57.post_attention_layernorm.weight": "model-00045-of-00062.safetensors",
+    "model.layers.57.self_attn.k_proj.weight": "model-00044-of-00062.safetensors",
+    "model.layers.57.self_attn.o_proj.weight": "model-00044-of-00062.safetensors",
+    "model.layers.57.self_attn.q_proj.weight": "model-00044-of-00062.safetensors",
+    "model.layers.57.self_attn.v_proj.weight": "model-00044-of-00062.safetensors",
+    "model.layers.58.input_layernorm.weight": "model-00046-of-00062.safetensors",
+    "model.layers.58.mlp.down_proj.weight": "model-00046-of-00062.safetensors",
+    "model.layers.58.mlp.gate_proj.weight": "model-00045-of-00062.safetensors",
+    "model.layers.58.mlp.up_proj.weight": "model-00045-of-00062.safetensors",
+    "model.layers.58.post_attention_layernorm.weight": "model-00046-of-00062.safetensors",
+    "model.layers.58.self_attn.k_proj.weight": "model-00045-of-00062.safetensors",
+    "model.layers.58.self_attn.o_proj.weight": "model-00045-of-00062.safetensors",
+    "model.layers.58.self_attn.q_proj.weight": "model-00045-of-00062.safetensors",
+    "model.layers.58.self_attn.v_proj.weight": "model-00045-of-00062.safetensors",
+    "model.layers.59.input_layernorm.weight": "model-00046-of-00062.safetensors",
+    "model.layers.59.mlp.down_proj.weight": "model-00046-of-00062.safetensors",
+    "model.layers.59.mlp.gate_proj.weight": "model-00046-of-00062.safetensors",
+    "model.layers.59.mlp.up_proj.weight": "model-00046-of-00062.safetensors",
+    "model.layers.59.post_attention_layernorm.weight": "model-00046-of-00062.safetensors",
+    "model.layers.59.self_attn.k_proj.weight": "model-00046-of-00062.safetensors",
+    "model.layers.59.self_attn.o_proj.weight": "model-00046-of-00062.safetensors",
+    "model.layers.59.self_attn.q_proj.weight": "model-00046-of-00062.safetensors",
+    "model.layers.59.self_attn.v_proj.weight": "model-00046-of-00062.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00007-of-00062.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00007-of-00062.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00006-of-00062.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00006-of-00062.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00007-of-00062.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00006-of-00062.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00006-of-00062.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00006-of-00062.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00006-of-00062.safetensors",
+    "model.layers.60.input_layernorm.weight": "model-00047-of-00062.safetensors",
+    "model.layers.60.mlp.down_proj.weight": "model-00047-of-00062.safetensors",
+    "model.layers.60.mlp.gate_proj.weight": "model-00047-of-00062.safetensors",
+    "model.layers.60.mlp.up_proj.weight": "model-00047-of-00062.safetensors",
+    "model.layers.60.post_attention_layernorm.weight": "model-00047-of-00062.safetensors",
+    "model.layers.60.self_attn.k_proj.weight": "model-00046-of-00062.safetensors",
+    "model.layers.60.self_attn.o_proj.weight": "model-00046-of-00062.safetensors",
+    "model.layers.60.self_attn.q_proj.weight": "model-00046-of-00062.safetensors",
+    "model.layers.60.self_attn.v_proj.weight": "model-00046-of-00062.safetensors",
+    "model.layers.61.input_layernorm.weight": "model-00048-of-00062.safetensors",
+    "model.layers.61.mlp.down_proj.weight": "model-00048-of-00062.safetensors",
+    "model.layers.61.mlp.gate_proj.weight": "model-00047-of-00062.safetensors",
+    "model.layers.61.mlp.up_proj.weight": "model-00048-of-00062.safetensors",
+    "model.layers.61.post_attention_layernorm.weight": "model-00048-of-00062.safetensors",
+    "model.layers.61.self_attn.k_proj.weight": "model-00047-of-00062.safetensors",
+    "model.layers.61.self_attn.o_proj.weight": "model-00047-of-00062.safetensors",
+    "model.layers.61.self_attn.q_proj.weight": "model-00047-of-00062.safetensors",
+    "model.layers.61.self_attn.v_proj.weight": "model-00047-of-00062.safetensors",
+    "model.layers.62.input_layernorm.weight": "model-00049-of-00062.safetensors",
+    "model.layers.62.mlp.down_proj.weight": "model-00049-of-00062.safetensors",
+    "model.layers.62.mlp.gate_proj.weight": "model-00048-of-00062.safetensors",
+    "model.layers.62.mlp.up_proj.weight": "model-00048-of-00062.safetensors",
+    "model.layers.62.post_attention_layernorm.weight": "model-00049-of-00062.safetensors",
+    "model.layers.62.self_attn.k_proj.weight": "model-00048-of-00062.safetensors",
+    "model.layers.62.self_attn.o_proj.weight": "model-00048-of-00062.safetensors",
+    "model.layers.62.self_attn.q_proj.weight": "model-00048-of-00062.safetensors",
+    "model.layers.62.self_attn.v_proj.weight": "model-00048-of-00062.safetensors",
+    "model.layers.63.input_layernorm.weight": "model-00049-of-00062.safetensors",
+    "model.layers.63.mlp.down_proj.weight": "model-00049-of-00062.safetensors",
+    "model.layers.63.mlp.gate_proj.weight": "model-00049-of-00062.safetensors",
+    "model.layers.63.mlp.up_proj.weight": "model-00049-of-00062.safetensors",
+    "model.layers.63.post_attention_layernorm.weight": "model-00049-of-00062.safetensors",
+    "model.layers.63.self_attn.k_proj.weight": "model-00049-of-00062.safetensors",
+    "model.layers.63.self_attn.o_proj.weight": "model-00049-of-00062.safetensors",
+    "model.layers.63.self_attn.q_proj.weight": "model-00049-of-00062.safetensors",
+    "model.layers.63.self_attn.v_proj.weight": "model-00049-of-00062.safetensors",
+    "model.layers.64.input_layernorm.weight": "model-00050-of-00062.safetensors",
+    "model.layers.64.mlp.down_proj.weight": "model-00050-of-00062.safetensors",
+    "model.layers.64.mlp.gate_proj.weight": "model-00050-of-00062.safetensors",
+    "model.layers.64.mlp.up_proj.weight": "model-00050-of-00062.safetensors",
+    "model.layers.64.post_attention_layernorm.weight": "model-00050-of-00062.safetensors",
+    "model.layers.64.self_attn.k_proj.weight": "model-00049-of-00062.safetensors",
+    "model.layers.64.self_attn.o_proj.weight": "model-00049-of-00062.safetensors",
+    "model.layers.64.self_attn.q_proj.weight": "model-00049-of-00062.safetensors",
+    "model.layers.64.self_attn.v_proj.weight": "model-00049-of-00062.safetensors",
+    "model.layers.65.input_layernorm.weight": "model-00051-of-00062.safetensors",
+    "model.layers.65.mlp.down_proj.weight": "model-00051-of-00062.safetensors",
+    "model.layers.65.mlp.gate_proj.weight": "model-00050-of-00062.safetensors",
+    "model.layers.65.mlp.up_proj.weight": "model-00051-of-00062.safetensors",
+    "model.layers.65.post_attention_layernorm.weight": "model-00051-of-00062.safetensors",
+    "model.layers.65.self_attn.k_proj.weight": "model-00050-of-00062.safetensors",
+    "model.layers.65.self_attn.o_proj.weight": "model-00050-of-00062.safetensors",
+    "model.layers.65.self_attn.q_proj.weight": "model-00050-of-00062.safetensors",
+    "model.layers.65.self_attn.v_proj.weight": "model-00050-of-00062.safetensors",
+    "model.layers.66.input_layernorm.weight": "model-00052-of-00062.safetensors",
+    "model.layers.66.mlp.down_proj.weight": "model-00052-of-00062.safetensors",
+    "model.layers.66.mlp.gate_proj.weight": "model-00051-of-00062.safetensors",
+    "model.layers.66.mlp.up_proj.weight": "model-00051-of-00062.safetensors",
+    "model.layers.66.post_attention_layernorm.weight": "model-00052-of-00062.safetensors",
+    "model.layers.66.self_attn.k_proj.weight": "model-00051-of-00062.safetensors",
+    "model.layers.66.self_attn.o_proj.weight": "model-00051-of-00062.safetensors",
+    "model.layers.66.self_attn.q_proj.weight": "model-00051-of-00062.safetensors",
+    "model.layers.66.self_attn.v_proj.weight": "model-00051-of-00062.safetensors",
+    "model.layers.67.input_layernorm.weight": "model-00052-of-00062.safetensors",
+    "model.layers.67.mlp.down_proj.weight": "model-00052-of-00062.safetensors",
+    "model.layers.67.mlp.gate_proj.weight": "model-00052-of-00062.safetensors",
+    "model.layers.67.mlp.up_proj.weight": "model-00052-of-00062.safetensors",
+    "model.layers.67.post_attention_layernorm.weight": "model-00052-of-00062.safetensors",
+    "model.layers.67.self_attn.k_proj.weight": "model-00052-of-00062.safetensors",
+    "model.layers.67.self_attn.o_proj.weight": "model-00052-of-00062.safetensors",
+    "model.layers.67.self_attn.q_proj.weight": "model-00052-of-00062.safetensors",
+    "model.layers.67.self_attn.v_proj.weight": "model-00052-of-00062.safetensors",
+    "model.layers.68.input_layernorm.weight": "model-00053-of-00062.safetensors",
+    "model.layers.68.mlp.down_proj.weight": "model-00053-of-00062.safetensors",
+    "model.layers.68.mlp.gate_proj.weight": "model-00053-of-00062.safetensors",
+    "model.layers.68.mlp.up_proj.weight": "model-00053-of-00062.safetensors",
+    "model.layers.68.post_attention_layernorm.weight": "model-00053-of-00062.safetensors",
+    "model.layers.68.self_attn.k_proj.weight": "model-00052-of-00062.safetensors",
+    "model.layers.68.self_attn.o_proj.weight": "model-00052-of-00062.safetensors",
+    "model.layers.68.self_attn.q_proj.weight": "model-00052-of-00062.safetensors",
+    "model.layers.68.self_attn.v_proj.weight": "model-00052-of-00062.safetensors",
+    "model.layers.69.input_layernorm.weight": "model-00054-of-00062.safetensors",
+    "model.layers.69.mlp.down_proj.weight": "model-00054-of-00062.safetensors",
+    "model.layers.69.mlp.gate_proj.weight": "model-00053-of-00062.safetensors",
+    "model.layers.69.mlp.up_proj.weight": "model-00054-of-00062.safetensors",
+    "model.layers.69.post_attention_layernorm.weight": "model-00054-of-00062.safetensors",
+    "model.layers.69.self_attn.k_proj.weight": "model-00053-of-00062.safetensors",
+    "model.layers.69.self_attn.o_proj.weight": "model-00053-of-00062.safetensors",
+    "model.layers.69.self_attn.q_proj.weight": "model-00053-of-00062.safetensors",
+    "model.layers.69.self_attn.v_proj.weight": "model-00053-of-00062.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00007-of-00062.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00007-of-00062.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00007-of-00062.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00007-of-00062.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00007-of-00062.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00007-of-00062.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00007-of-00062.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00007-of-00062.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00007-of-00062.safetensors",
+    "model.layers.70.input_layernorm.weight": "model-00055-of-00062.safetensors",
+    "model.layers.70.mlp.down_proj.weight": "model-00055-of-00062.safetensors",
+    "model.layers.70.mlp.gate_proj.weight": "model-00054-of-00062.safetensors",
+    "model.layers.70.mlp.up_proj.weight": "model-00054-of-00062.safetensors",
+    "model.layers.70.post_attention_layernorm.weight": "model-00055-of-00062.safetensors",
+    "model.layers.70.self_attn.k_proj.weight": "model-00054-of-00062.safetensors",
+    "model.layers.70.self_attn.o_proj.weight": "model-00054-of-00062.safetensors",
+    "model.layers.70.self_attn.q_proj.weight": "model-00054-of-00062.safetensors",
+    "model.layers.70.self_attn.v_proj.weight": "model-00054-of-00062.safetensors",
+    "model.layers.71.input_layernorm.weight": "model-00055-of-00062.safetensors",
+    "model.layers.71.mlp.down_proj.weight": "model-00055-of-00062.safetensors",
+    "model.layers.71.mlp.gate_proj.weight": "model-00055-of-00062.safetensors",
+    "model.layers.71.mlp.up_proj.weight": "model-00055-of-00062.safetensors",
+    "model.layers.71.post_attention_layernorm.weight": "model-00055-of-00062.safetensors",
+    "model.layers.71.self_attn.k_proj.weight": "model-00055-of-00062.safetensors",
+    "model.layers.71.self_attn.o_proj.weight": "model-00055-of-00062.safetensors",
+    "model.layers.71.self_attn.q_proj.weight": "model-00055-of-00062.safetensors",
+    "model.layers.71.self_attn.v_proj.weight": "model-00055-of-00062.safetensors",
+    "model.layers.72.input_layernorm.weight": "model-00056-of-00062.safetensors",
+    "model.layers.72.mlp.down_proj.weight": "model-00056-of-00062.safetensors",
+    "model.layers.72.mlp.gate_proj.weight": "model-00056-of-00062.safetensors",
+    "model.layers.72.mlp.up_proj.weight": "model-00056-of-00062.safetensors",
+    "model.layers.72.post_attention_layernorm.weight": "model-00056-of-00062.safetensors",
+    "model.layers.72.self_attn.k_proj.weight": "model-00055-of-00062.safetensors",
+    "model.layers.72.self_attn.o_proj.weight": "model-00055-of-00062.safetensors",
+    "model.layers.72.self_attn.q_proj.weight": "model-00055-of-00062.safetensors",
+    "model.layers.72.self_attn.v_proj.weight": "model-00055-of-00062.safetensors",
+    "model.layers.73.input_layernorm.weight": "model-00057-of-00062.safetensors",
+    "model.layers.73.mlp.down_proj.weight": "model-00057-of-00062.safetensors",
+    "model.layers.73.mlp.gate_proj.weight": "model-00056-of-00062.safetensors",
+    "model.layers.73.mlp.up_proj.weight": "model-00057-of-00062.safetensors",
+    "model.layers.73.post_attention_layernorm.weight": "model-00057-of-00062.safetensors",
+    "model.layers.73.self_attn.k_proj.weight": "model-00056-of-00062.safetensors",
+    "model.layers.73.self_attn.o_proj.weight": "model-00056-of-00062.safetensors",
+    "model.layers.73.self_attn.q_proj.weight": "model-00056-of-00062.safetensors",
+    "model.layers.73.self_attn.v_proj.weight": "model-00056-of-00062.safetensors",
+    "model.layers.74.input_layernorm.weight": "model-00058-of-00062.safetensors",
+    "model.layers.74.mlp.down_proj.weight": "model-00058-of-00062.safetensors",
+    "model.layers.74.mlp.gate_proj.weight": "model-00057-of-00062.safetensors",
+    "model.layers.74.mlp.up_proj.weight": "model-00057-of-00062.safetensors",
+    "model.layers.74.post_attention_layernorm.weight": "model-00058-of-00062.safetensors",
+    "model.layers.74.self_attn.k_proj.weight": "model-00057-of-00062.safetensors",
+    "model.layers.74.self_attn.o_proj.weight": "model-00057-of-00062.safetensors",
+    "model.layers.74.self_attn.q_proj.weight": "model-00057-of-00062.safetensors",
+    "model.layers.74.self_attn.v_proj.weight": "model-00057-of-00062.safetensors",
+    "model.layers.75.input_layernorm.weight": "model-00058-of-00062.safetensors",
+    "model.layers.75.mlp.down_proj.weight": "model-00058-of-00062.safetensors",
+    "model.layers.75.mlp.gate_proj.weight": "model-00058-of-00062.safetensors",
+    "model.layers.75.mlp.up_proj.weight": "model-00058-of-00062.safetensors",
+    "model.layers.75.post_attention_layernorm.weight": "model-00058-of-00062.safetensors",
+    "model.layers.75.self_attn.k_proj.weight": "model-00058-of-00062.safetensors",
+    "model.layers.75.self_attn.o_proj.weight": "model-00058-of-00062.safetensors",
+    "model.layers.75.self_attn.q_proj.weight": "model-00058-of-00062.safetensors",
+    "model.layers.75.self_attn.v_proj.weight": "model-00058-of-00062.safetensors",
+    "model.layers.76.input_layernorm.weight": "model-00059-of-00062.safetensors",
+    "model.layers.76.mlp.down_proj.weight": "model-00059-of-00062.safetensors",
+    "model.layers.76.mlp.gate_proj.weight": "model-00059-of-00062.safetensors",
+    "model.layers.76.mlp.up_proj.weight": "model-00059-of-00062.safetensors",
+    "model.layers.76.post_attention_layernorm.weight": "model-00059-of-00062.safetensors",
+    "model.layers.76.self_attn.k_proj.weight": "model-00058-of-00062.safetensors",
+    "model.layers.76.self_attn.o_proj.weight": "model-00058-of-00062.safetensors",
+    "model.layers.76.self_attn.q_proj.weight": "model-00058-of-00062.safetensors",
+    "model.layers.76.self_attn.v_proj.weight": "model-00058-of-00062.safetensors",
+    "model.layers.77.input_layernorm.weight": "model-00060-of-00062.safetensors",
+    "model.layers.77.mlp.down_proj.weight": "model-00060-of-00062.safetensors",
+    "model.layers.77.mlp.gate_proj.weight": "model-00059-of-00062.safetensors",
+    "model.layers.77.mlp.up_proj.weight": "model-00060-of-00062.safetensors",
+    "model.layers.77.post_attention_layernorm.weight": "model-00060-of-00062.safetensors",
+    "model.layers.77.self_attn.k_proj.weight": "model-00059-of-00062.safetensors",
+    "model.layers.77.self_attn.o_proj.weight": "model-00059-of-00062.safetensors",
+    "model.layers.77.self_attn.q_proj.weight": "model-00059-of-00062.safetensors",
+    "model.layers.77.self_attn.v_proj.weight": "model-00059-of-00062.safetensors",
+    "model.layers.78.input_layernorm.weight": "model-00061-of-00062.safetensors",
+    "model.layers.78.mlp.down_proj.weight": "model-00061-of-00062.safetensors",
+    "model.layers.78.mlp.gate_proj.weight": "model-00060-of-00062.safetensors",
+    "model.layers.78.mlp.up_proj.weight": "model-00060-of-00062.safetensors",
+    "model.layers.78.post_attention_layernorm.weight": "model-00061-of-00062.safetensors",
+    "model.layers.78.self_attn.k_proj.weight": "model-00060-of-00062.safetensors",
+    "model.layers.78.self_attn.o_proj.weight": "model-00060-of-00062.safetensors",
+    "model.layers.78.self_attn.q_proj.weight": "model-00060-of-00062.safetensors",
+    "model.layers.78.self_attn.v_proj.weight": "model-00060-of-00062.safetensors",
+    "model.layers.79.input_layernorm.weight": "model-00061-of-00062.safetensors",
+    "model.layers.79.mlp.down_proj.weight": "model-00061-of-00062.safetensors",
+    "model.layers.79.mlp.gate_proj.weight": "model-00061-of-00062.safetensors",
+    "model.layers.79.mlp.up_proj.weight": "model-00061-of-00062.safetensors",
+    "model.layers.79.post_attention_layernorm.weight": "model-00061-of-00062.safetensors",
+    "model.layers.79.self_attn.k_proj.weight": "model-00061-of-00062.safetensors",
+    "model.layers.79.self_attn.o_proj.weight": "model-00061-of-00062.safetensors",
+    "model.layers.79.self_attn.q_proj.weight": "model-00061-of-00062.safetensors",
+    "model.layers.79.self_attn.v_proj.weight": "model-00061-of-00062.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00008-of-00062.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00008-of-00062.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00008-of-00062.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00008-of-00062.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00008-of-00062.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00007-of-00062.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00007-of-00062.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00007-of-00062.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00007-of-00062.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00009-of-00062.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00009-of-00062.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00008-of-00062.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00009-of-00062.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00009-of-00062.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00008-of-00062.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00008-of-00062.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00008-of-00062.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00008-of-00062.safetensors",
+    "model.norm.weight": "model-00061-of-00062.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,23 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|end_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|finetune_right_pad_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920