Upload model weights (#1)

- 665c57f6bf309ab0c6c8ad36c1636003e39461a01fed026c4d15e9102569a26a (a407ecf3bb50400c3c4461630bf0695ba7808b2a)
- be8b4937c1bfc88b56db629a463476b7536d9d4d3aaabbecc35d2a5b3122888e (e9aa06ac119668bacc16f973d69bc5f64bb7dc61)
- 452405d7d8fdfe26ab37a2fa12e9ef68a29301522a480454f9e7a246d035d853 (81db3e47887f2a8a62fa1e3cef28549fdb373d6c)
- c48845a850c2477a927ec6b5e5250712e9dc830b2be3212a4d8a5744ab9cf4fb (d3e88d269b9f2855584c1f8203eeb3326e80d730)
- 548af04d6a0e9e71664c78ae1aefe059db13a45e486fc4b69da28bef2ec20b83 (6c41d3e4c9da6cb8c24004bddefdc728db6cbfc0)
- 697e146201075b5464230aa5e47a65fa0cfe72ee87a227e235d18b8853611732 (94b422232922ce09a29733349bd262130e50bdb4)
- fb8784094ed88003a553963004f6dce02b81c99f38cd1cc69626119b2c560027 (6f2ab5c5269308cf27201f531efb87f138855fba)
- de5fdeb3751e822508c7a196c6068cd0a02e1b8b02d98115941e3824951da7a0 (2c0611b03a2c90c634fd48e417f93337b8fa8083)
- 118e03bf1eb573c55e7b7c91c055d571070b1304c122078ab1a2ffa485551e40 (c54fb75632eae43248fdbc7fae0fe773cfef1245)
- 1766777620c29609882a25423b47f4e68224e176e8d932ca7990ac9a6bbefaf8 (09666fac9827983079cdc74d77ee2fe52c0b28e8)
- d263a450cce91c455be5969d9ebef56a5435c68b77c90cf7dc0c791892dd2b04 (f336fee41bcacdc6e946bbd638744c3d8ee0c30b)
- 4a64dbe415b708bea9a9f01347337777d36083d03772f6d93e55d78b9a18e397 (aef6b348ec5176ea64838aab4ced8021fd86611f)
- 1bf98c2041e92054e36a4ea4b306ed6278fdafda3712c1aca257229cac33fe7a (c0bf1051289d11c10937da1b53b9e580abb1945f)
- 5736d166c789c7eacf601d59c7404c19cf34428cc4f756ecddd814612c1e07e5 (aa7fb6024a102ebf54a255c6bbb630b28ec4fca8)
- bb3671e7c8b13080b067122f560eb18eb5a766d4774fd6912126204e073494e7 (709dbb5799cc95176d85fe512b7e0c799d8f11b3)
- 87b0fe450b2755e693b688fbbc60ede0183cce9fc8f9a33419e679a95390be6d (d7e85ed83f50239546ceee208a2ce2b1bae7d256)
- db7927a10de47cf143d724591c19f2da534f6ad6b569cf26aec56c3d1801acb8 (7c139879c13870c54a9e4e71e10680e28bf3895f)
- 214a57c22f20bbbf6ff47f485a797a8791fb730ef12705be9004c69d49f7dd76 (d20414e7cb855b9b63996f414745122613d2c588)
- 49b07f53df33292a2f9edfb2f62d2d53bce6504c19016343cb2416129ca3ecdf (37fd747b7b2aafa856099c9daabd63c663a09ae1)
- 685d046d54a0222561e52ca53acc68d562768135123c240f23960e9e733a39ff (c4d71ef1f1a1d6be94f6988cf9c63e588d916c18)
- 292f65bbe0579ca018340375de730c6e82b5df6a88db89035ea4e3a66eb2ed0f (aea1da4c29335844afcfee367cd3b4af35ec4d3a)
- a7bdbc3936f9b3a9dfe2bf3f5c9668c0111f187fc3c60381597f5fb8ddc6fe00 (b5a534b373a9ad810087b8a8d9e46e25fb7d898d)

README.md

@@ -1,3 +1,39 @@
 ---
-license: llama2
+base_model: []
+tags:
+- mergekit
+- merge
+
 ---
+# mr-103b-v2.0.3
+
+This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
+
+## Merge Details
+### Merge Method
+
+This model was merged using the passthrough merge method.
+
+### Models Merged
+
+The following models were included in the merge:
+* /home/llm/mergequant/models/mr-v2.0.3-wizardblend
+
+### Configuration
+
+The following YAML configuration was used to produce this model:
+
+```yaml
+slices:
+  - sources:
+      - model: /home/llm/mergequant/models/mr-v2.0.3-wizardblend
+        layer_range: [0, 40] # 40
+  - sources:
+      - model: /home/llm/mergequant/models/mr-v2.0.3-wizardblend
+        layer_range: [20, 60] # 40
+  - sources:
+      - model: /home/llm/mergequant/models/mr-v2.0.3-wizardblend
+        layer_range: [40, 80] # 40
+merge_method: passthrough
+dtype: float16
+```

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "midnight-rose-103b-v2.0.3",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 8192,
+  "initializer_range": 0.02,
+  "intermediate_size": 28672,
+  "max_position_embeddings": 4096,
+  "model_type": "llama",
+  "num_attention_heads": 64,
+  "num_hidden_layers": 120,
+  "num_key_value_heads": 8,
+  "pad_token_id": 0,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.36.2",
+  "use_cache": true,
+  "vocab_size": 32000
+}

mergekit_config.yml

@@ -0,0 +1,12 @@
+slices:
+  - sources:
+      - model: /home/llm/mergequant/models/mr-v2.0.3-wizardblend
+        layer_range: [0, 40] # 40
+  - sources:
+      - model: /home/llm/mergequant/models/mr-v2.0.3-wizardblend
+        layer_range: [20, 60] # 40
+  - sources:
+      - model: /home/llm/mergequant/models/mr-v2.0.3-wizardblend
+        layer_range: [40, 80] # 40
+merge_method: passthrough
+dtype: float16

model-00001-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44a1c814d789f83a33f6b59cd68d4afa92b3a5c8ad2e09fca971927a7610fa0a
+size 9986824152

model-00002-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ef5dcb72c1326461a7d5d4b78c58356f17c61d212adb99bcf47b2185fb2e478
+size 9663879040

model-00003-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75c7faf9a67acb895f7d2dc490ce3034dd8cf796edebee51b6dc9ffdd4296da0
+size 9949074880

model-00004-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a90212b68dd64ceeb3e1c018d8b46ac185431163d6195cdc86b647790d15d1f9
+size 9965852920

model-00005-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce287d4a131aad71e69a6d7f375fb7c9944a89f1fcb4980257b1eb9c135515c2
+size 9613597984

model-00006-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b05848b1661174c214c9285c005c959160bf152823e2d17574a2e3987e39f44b
+size 9798096896

model-00007-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e812483b2a0554373a1dd81afe5d6f703b85199ed76da1ec2099e88ca6ffa19
+size 9798063904

model-00008-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ff740eed0b07a69d5e46799e24b027bdb4875573038b2efa4d54bdbabd30390
+size 9798096912

model-00009-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e97132976c9bd94f7a32079986f02c83848e66a5b38535a91b02ac63efb4552e
+size 9563181496

model-00010-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18fa6a6c8c88eec8b4cd9b21d8302304c121a691d32f0181e7a894357c959c2a
+size 9663878360

model-00011-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae59c4e720568b39c00d31cbaf709b7519c76d0d7c6b3682eb846487cbbba915
+size 9663812376

model-00012-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4728b2be7bd061ffd7ffe4ee15cce94d3bdf8ed5f60c234eadda5b7b0c3c116c
+size 9798096920

model-00013-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f3c40be1b9d97673b22690435e590164428df06686f9095ac6b4f7d40ab6ee7
+size 9965870104

model-00014-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:849c2f7c41f6fc5c1769981bf5431b4a5b1fef60d807fffecaf4de5edc28567b
+size 9932314048

model-00015-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64b14163a3e8df2cb0add4f59cf388534d228fdcecb8cd66452abc61df4ee519
+size 9898793800

model-00016-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:046bfe4e964fd5f9f7019004b7da222be464b4c431555a5a04db78d2551c9a52
+size 9798096896

model-00017-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c84fd898ba6ebb05d7dd2fd57a85cadf6b2833414f19c071762340a6e717603
+size 9982646080

model-00018-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56be7568d484ed01fba4de5cde6c274380841d7342442f0dcad0821174366ba8
+size 9663812376

model-00019-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66819b0c3eaef3c298562a4f717bb900c7ded4fb3660cc1e40bdbb35c79330dc
+size 9781303128

model-00020-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc6a7fbddc499bb70784c5e90cfdfa53df88e39dae077652e64e79468f97afd7
+size 9798096960

model-00021-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd622010df6d4ca62ce2fa1689072dd129f1c83995efc98ed108440d5efc4922
+size 9852622880

model-00022-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c762c6ccf60c3f9c3a659aa38be17fc427532cdc6de16cf24dd813e2b9e9c321
+size 469762192

model.safetensors.index.json

@@ -0,0 +1 @@
+{"metadata": {"mergekit_version": "0.0.4"}, "weight_map": {"model.layers.9.self_attn.o_proj.weight": "model-00001-of-00022.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00022.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00022.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00022.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00022.safetensors", "model.layers.5.mlp.down_proj.weight": "model-00001-of-00022.safetensors", "model.layers.5.input_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00022.safetensors", "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00022.safetensors", "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00022.safetensors", "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00022.safetensors", "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00022.safetensors", "model.layers.4.mlp.down_proj.weight": "model-00001-of-00022.safetensors", "model.layers.4.mlp.up_proj.weight": "model-00001-of-00022.safetensors", "model.layers.4.input_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00022.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00022.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00022.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00022.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00022.safetensors", "model.layers.3.mlp.down_proj.weight": "model-00001-of-00022.safetensors", "model.layers.3.mlp.up_proj.weight": "model-00001-of-00022.safetensors", "model.layers.3.input_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00022.safetensors", "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00022.safetensors", "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00022.safetensors", "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00022.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00022.safetensors", "model.layers.2.mlp.down_proj.weight": "model-00001-of-00022.safetensors", "model.layers.2.mlp.up_proj.weight": "model-00001-of-00022.safetensors", "model.layers.2.input_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00022.safetensors", "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00022.safetensors", "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00022.safetensors", "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00022.safetensors", "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00022.safetensors", "model.layers.1.mlp.down_proj.weight": "model-00001-of-00022.safetensors", "model.layers.1.mlp.up_proj.weight": "model-00001-of-00022.safetensors", "model.layers.1.input_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00022.safetensors", "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00022.safetensors", "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00022.safetensors", "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00022.safetensors", "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00022.safetensors", "model.layers.0.mlp.down_proj.weight": "model-00001-of-00022.safetensors", "model.layers.0.mlp.up_proj.weight": "model-00001-of-00022.safetensors", "model.layers.0.input_layernorm.weight": "model-00001-of-00022.safetensors", "model.embed_tokens.weight": "model-00001-of-00022.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00022.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00022.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00022.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00022.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00022.safetensors", "model.layers.11.mlp.down_proj.weight": "model-00002-of-00022.safetensors", "model.layers.11.mlp.up_proj.weight": "model-00002-of-00022.safetensors", "model.layers.11.input_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00022.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00022.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00022.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00022.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00022.safetensors", "model.layers.10.mlp.down_proj.weight": "model-00002-of-00022.safetensors", "model.layers.10.mlp.up_proj.weight": "model-00002-of-00022.safetensors", "model.layers.10.input_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.8.self_attn.v_proj.weight": "model-00002-of-00022.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00002-of-00022.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00002-of-00022.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00022.safetensors", "model.layers.8.mlp.down_proj.weight": "model-00002-of-00022.safetensors", "model.layers.8.mlp.up_proj.weight": "model-00002-of-00022.safetensors", "model.layers.8.input_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00022.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00002-of-00022.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00002-of-00022.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00002-of-00022.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.7.mlp.gate_proj.weight": "model-00002-of-00022.safetensors", "model.layers.7.mlp.down_proj.weight": "model-00002-of-00022.safetensors", "model.layers.7.mlp.up_proj.weight": "model-00002-of-00022.safetensors", "model.layers.7.input_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00002-of-00022.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00002-of-00022.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00002-of-00022.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00002-of-00022.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.6.mlp.gate_proj.weight": "model-00002-of-00022.safetensors", "model.layers.6.mlp.down_proj.weight": "model-00002-of-00022.safetensors", "model.layers.6.mlp.up_proj.weight": "model-00002-of-00022.safetensors", "model.layers.6.input_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.5.self_attn.o_proj.weight": "model-00002-of-00022.safetensors", "model.layers.5.self_attn.v_proj.weight": "model-00002-of-00022.safetensors", "model.layers.5.self_attn.k_proj.weight": "model-00002-of-00022.safetensors", "model.layers.5.self_attn.q_proj.weight": "model-00002-of-00022.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.5.mlp.gate_proj.weight": "model-00002-of-00022.safetensors", "model.layers.5.mlp.up_proj.weight": "model-00002-of-00022.safetensors", "model.layers.16.mlp.gate_proj.weight": "model-00003-of-00022.safetensors", "model.layers.16.mlp.down_proj.weight": "model-00003-of-00022.safetensors", "model.layers.16.mlp.up_proj.weight": "model-00003-of-00022.safetensors", "model.layers.16.input_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00003-of-00022.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00003-of-00022.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00003-of-00022.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00003-of-00022.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.15.mlp.gate_proj.weight": "model-00003-of-00022.safetensors", "model.layers.15.mlp.down_proj.weight": "model-00003-of-00022.safetensors", "model.layers.15.mlp.up_proj.weight": "model-00003-of-00022.safetensors", "model.layers.15.input_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00003-of-00022.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00003-of-00022.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00003-of-00022.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00003-of-00022.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.14.mlp.gate_proj.weight": "model-00003-of-00022.safetensors", "model.layers.14.mlp.down_proj.weight": "model-00003-of-00022.safetensors", "model.layers.14.mlp.up_proj.weight": "model-00003-of-00022.safetensors", "model.layers.14.input_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00003-of-00022.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00003-of-00022.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00003-of-00022.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00003-of-00022.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.13.mlp.gate_proj.weight": "model-00003-of-00022.safetensors", "model.layers.13.mlp.down_proj.weight": "model-00003-of-00022.safetensors", "model.layers.13.mlp.up_proj.weight": "model-00003-of-00022.safetensors", "model.layers.13.input_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00003-of-00022.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00003-of-00022.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00003-of-00022.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00003-of-00022.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.12.mlp.gate_proj.weight": "model-00003-of-00022.safetensors", "model.layers.12.mlp.down_proj.weight": "model-00003-of-00022.safetensors", "model.layers.12.mlp.up_proj.weight": "model-00003-of-00022.safetensors", "model.layers.12.input_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00003-of-00022.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00003-of-00022.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.9.mlp.gate_proj.weight": "model-00003-of-00022.safetensors", "model.layers.9.mlp.down_proj.weight": "model-00003-of-00022.safetensors", "model.layers.9.mlp.up_proj.weight": "model-00003-of-00022.safetensors", "model.layers.9.input_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.48.self_attn.o_proj.weight": "model-00003-of-00022.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00004-of-00022.safetensors", "model.layers.48.self_attn.v_proj.weight": "model-00004-of-00022.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00004-of-00022.safetensors", "model.layers.48.self_attn.k_proj.weight": "model-00004-of-00022.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00004-of-00022.safetensors", "model.layers.48.self_attn.q_proj.weight": "model-00004-of-00022.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00004-of-00022.safetensors", "model.layers.42.mlp.gate_proj.weight": "model-00004-of-00022.safetensors", "model.layers.22.mlp.gate_proj.weight": "model-00004-of-00022.safetensors", "model.layers.42.mlp.down_proj.weight": "model-00004-of-00022.safetensors", "model.layers.22.mlp.down_proj.weight": "model-00004-of-00022.safetensors", "model.layers.42.input_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.22.input_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.41.self_attn.o_proj.weight": "model-00004-of-00022.safetensors", "model.layers.21.self_attn.o_proj.weight": "model-00004-of-00022.safetensors", "model.layers.41.self_attn.v_proj.weight": "model-00004-of-00022.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00004-of-00022.safetensors", "model.layers.41.self_attn.k_proj.weight": "model-00004-of-00022.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00004-of-00022.safetensors", "model.layers.41.self_attn.q_proj.weight": "model-00004-of-00022.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00004-of-00022.safetensors", "model.layers.41.post_attention_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.41.mlp.gate_proj.weight": "model-00004-of-00022.safetensors", "model.layers.21.mlp.gate_proj.weight": "model-00004-of-00022.safetensors", "model.layers.41.mlp.down_proj.weight": "model-00004-of-00022.safetensors", "model.layers.21.mlp.down_proj.weight": "model-00004-of-00022.safetensors", "model.layers.41.mlp.up_proj.weight": "model-00004-of-00022.safetensors", "model.layers.21.mlp.up_proj.weight": "model-00004-of-00022.safetensors", "model.layers.41.input_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.21.input_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.40.self_attn.o_proj.weight": "model-00004-of-00022.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00004-of-00022.safetensors", "model.layers.40.self_attn.v_proj.weight": "model-00004-of-00022.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00004-of-00022.safetensors", "model.layers.40.self_attn.k_proj.weight": "model-00004-of-00022.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00004-of-00022.safetensors", "model.layers.40.self_attn.q_proj.weight": "model-00004-of-00022.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00004-of-00022.safetensors", "model.layers.40.post_attention_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.40.mlp.gate_proj.weight": "model-00004-of-00022.safetensors", "model.layers.20.mlp.gate_proj.weight": "model-00004-of-00022.safetensors", "model.layers.40.mlp.down_proj.weight": "model-00004-of-00022.safetensors", "model.layers.20.mlp.down_proj.weight": "model-00004-of-00022.safetensors", "model.layers.40.mlp.up_proj.weight": "model-00004-of-00022.safetensors", "model.layers.20.mlp.up_proj.weight": "model-00004-of-00022.safetensors", "model.layers.40.input_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.20.input_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00004-of-00022.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00004-of-00022.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00004-of-00022.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00004-of-00022.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.19.mlp.gate_proj.weight": "model-00004-of-00022.safetensors", "model.layers.19.mlp.down_proj.weight": "model-00005-of-00022.safetensors", "model.layers.19.mlp.up_proj.weight": "model-00005-of-00022.safetensors", "model.layers.19.input_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00005-of-00022.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.18.mlp.gate_proj.weight": "model-00005-of-00022.safetensors", "model.layers.18.mlp.down_proj.weight": "model-00005-of-00022.safetensors", "model.layers.18.mlp.up_proj.weight": "model-00005-of-00022.safetensors", "model.layers.18.input_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00005-of-00022.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.17.mlp.gate_proj.weight": "model-00005-of-00022.safetensors", "model.layers.17.mlp.down_proj.weight": "model-00005-of-00022.safetensors", "model.layers.17.mlp.up_proj.weight": "model-00005-of-00022.safetensors", "model.layers.17.input_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.52.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.32.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.52.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.32.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.52.self_attn.k_proj.weight": "model-00005-of-00022.safetensors", "model.layers.32.self_attn.k_proj.weight": "model-00005-of-00022.safetensors", "model.layers.52.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.32.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.52.post_attention_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.32.post_attention_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.48.input_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.28.input_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.47.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.47.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.47.self_attn.k_proj.weight": "model-00005-of-00022.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00005-of-00022.safetensors", "model.layers.47.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.47.post_attention_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.47.mlp.gate_proj.weight": "model-00005-of-00022.safetensors", "model.layers.27.mlp.gate_proj.weight": "model-00005-of-00022.safetensors", "model.layers.47.mlp.down_proj.weight": "model-00005-of-00022.safetensors", "model.layers.27.mlp.down_proj.weight": "model-00005-of-00022.safetensors", "model.layers.47.mlp.up_proj.weight": "model-00005-of-00022.safetensors", "model.layers.27.mlp.up_proj.weight": "model-00005-of-00022.safetensors", "model.layers.47.input_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.27.input_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.46.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.46.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.46.self_attn.k_proj.weight": "model-00005-of-00022.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00005-of-00022.safetensors", "model.layers.46.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.46.post_attention_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.46.mlp.gate_proj.weight": "model-00005-of-00022.safetensors", "model.layers.26.mlp.gate_proj.weight": "model-00006-of-00022.safetensors", "model.layers.46.mlp.down_proj.weight": "model-00006-of-00022.safetensors", "model.layers.26.mlp.down_proj.weight": "model-00006-of-00022.safetensors", "model.layers.46.mlp.up_proj.weight": "model-00006-of-00022.safetensors", "model.layers.26.mlp.up_proj.weight": "model-00006-of-00022.safetensors", "model.layers.46.input_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.26.input_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.45.self_attn.o_proj.weight": "model-00006-of-00022.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00006-of-00022.safetensors", "model.layers.45.self_attn.v_proj.weight": "model-00006-of-00022.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00006-of-00022.safetensors", "model.layers.45.self_attn.k_proj.weight": "model-00006-of-00022.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00006-of-00022.safetensors", "model.layers.45.self_attn.q_proj.weight": "model-00006-of-00022.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00006-of-00022.safetensors", "model.layers.45.post_attention_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.45.mlp.gate_proj.weight": "model-00006-of-00022.safetensors", "model.layers.25.mlp.gate_proj.weight": "model-00006-of-00022.safetensors", "model.layers.45.mlp.down_proj.weight": "model-00006-of-00022.safetensors", "model.layers.25.mlp.down_proj.weight": "model-00006-of-00022.safetensors", "model.layers.45.mlp.up_proj.weight": "model-00006-of-00022.safetensors", "model.layers.25.mlp.up_proj.weight": "model-00006-of-00022.safetensors", "model.layers.45.input_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.25.input_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.44.self_attn.o_proj.weight": "model-00006-of-00022.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00006-of-00022.safetensors", "model.layers.44.self_attn.v_proj.weight": "model-00006-of-00022.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00006-of-00022.safetensors", "model.layers.44.self_attn.k_proj.weight": "model-00006-of-00022.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00006-of-00022.safetensors", "model.layers.44.self_attn.q_proj.weight": "model-00006-of-00022.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00006-of-00022.safetensors", "model.layers.44.post_attention_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.44.mlp.gate_proj.weight": "model-00006-of-00022.safetensors", "model.layers.24.mlp.gate_proj.weight": "model-00006-of-00022.safetensors", "model.layers.44.mlp.down_proj.weight": "model-00006-of-00022.safetensors", "model.layers.24.mlp.down_proj.weight": "model-00006-of-00022.safetensors", "model.layers.44.mlp.up_proj.weight": "model-00006-of-00022.safetensors", "model.layers.24.mlp.up_proj.weight": "model-00006-of-00022.safetensors", "model.layers.44.input_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.24.input_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.43.self_attn.o_proj.weight": "model-00006-of-00022.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00006-of-00022.safetensors", "model.layers.43.self_attn.v_proj.weight": "model-00006-of-00022.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00006-of-00022.safetensors", "model.layers.43.self_attn.k_proj.weight": "model-00006-of-00022.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00006-of-00022.safetensors", "model.layers.43.self_attn.q_proj.weight": "model-00006-of-00022.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00006-of-00022.safetensors", "model.layers.43.post_attention_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.43.mlp.gate_proj.weight": "model-00007-of-00022.safetensors", "model.layers.23.mlp.gate_proj.weight": "model-00007-of-00022.safetensors", "model.layers.43.mlp.down_proj.weight": "model-00007-of-00022.safetensors", "model.layers.23.mlp.down_proj.weight": "model-00007-of-00022.safetensors", "model.layers.43.mlp.up_proj.weight": "model-00007-of-00022.safetensors", "model.layers.23.mlp.up_proj.weight": "model-00007-of-00022.safetensors", "model.layers.43.input_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.23.input_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.42.self_attn.o_proj.weight": "model-00007-of-00022.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00007-of-00022.safetensors", "model.layers.42.self_attn.v_proj.weight": "model-00007-of-00022.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00007-of-00022.safetensors", "model.layers.42.self_attn.k_proj.weight": "model-00007-of-00022.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00007-of-00022.safetensors", "model.layers.42.self_attn.q_proj.weight": "model-00007-of-00022.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00007-of-00022.safetensors", "model.layers.42.post_attention_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.42.mlp.up_proj.weight": "model-00007-of-00022.safetensors", "model.layers.22.mlp.up_proj.weight": "model-00007-of-00022.safetensors", "model.layers.57.self_attn.o_proj.weight": "model-00007-of-00022.safetensors", "model.layers.37.self_attn.o_proj.weight": "model-00007-of-00022.safetensors", "model.layers.57.self_attn.v_proj.weight": "model-00007-of-00022.safetensors", "model.layers.37.self_attn.v_proj.weight": "model-00007-of-00022.safetensors", "model.layers.57.self_attn.k_proj.weight": "model-00007-of-00022.safetensors", "model.layers.37.self_attn.k_proj.weight": "model-00007-of-00022.safetensors", "model.layers.57.self_attn.q_proj.weight": "model-00007-of-00022.safetensors", "model.layers.37.self_attn.q_proj.weight": "model-00007-of-00022.safetensors", "model.layers.54.input_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.34.input_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.53.self_attn.o_proj.weight": "model-00007-of-00022.safetensors", "model.layers.33.self_attn.o_proj.weight": "model-00007-of-00022.safetensors", "model.layers.53.self_attn.v_proj.weight": "model-00007-of-00022.safetensors", "model.layers.33.self_attn.v_proj.weight": "model-00007-of-00022.safetensors", "model.layers.53.self_attn.k_proj.weight": "model-00007-of-00022.safetensors", "model.layers.33.self_attn.k_proj.weight": "model-00007-of-00022.safetensors", "model.layers.53.self_attn.q_proj.weight": "model-00007-of-00022.safetensors", "model.layers.33.self_attn.q_proj.weight": "model-00007-of-00022.safetensors", "model.layers.53.post_attention_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.33.post_attention_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.53.mlp.gate_proj.weight": "model-00007-of-00022.safetensors", "model.layers.33.mlp.gate_proj.weight": "model-00007-of-00022.safetensors", "model.layers.53.mlp.down_proj.weight": "model-00007-of-00022.safetensors", "model.layers.33.mlp.down_proj.weight": "model-00007-of-00022.safetensors", "model.layers.53.mlp.up_proj.weight": "model-00007-of-00022.safetensors", "model.layers.33.mlp.up_proj.weight": "model-00007-of-00022.safetensors", "model.layers.53.input_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.33.input_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.52.mlp.gate_proj.weight": "model-00007-of-00022.safetensors", "model.layers.32.mlp.gate_proj.weight": "model-00007-of-00022.safetensors", "model.layers.52.mlp.down_proj.weight": "model-00007-of-00022.safetensors", "model.layers.32.mlp.down_proj.weight": "model-00008-of-00022.safetensors", "model.layers.52.mlp.up_proj.weight": "model-00008-of-00022.safetensors", "model.layers.32.mlp.up_proj.weight": "model-00008-of-00022.safetensors", "model.layers.52.input_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.32.input_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.51.self_attn.o_proj.weight": "model-00008-of-00022.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00008-of-00022.safetensors", "model.layers.51.self_attn.v_proj.weight": "model-00008-of-00022.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00008-of-00022.safetensors", "model.layers.51.self_attn.k_proj.weight": "model-00008-of-00022.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00008-of-00022.safetensors", "model.layers.51.self_attn.q_proj.weight": "model-00008-of-00022.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00008-of-00022.safetensors", "model.layers.51.post_attention_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.51.mlp.gate_proj.weight": "model-00008-of-00022.safetensors", "model.layers.31.mlp.gate_proj.weight": "model-00008-of-00022.safetensors", "model.layers.51.mlp.down_proj.weight": "model-00008-of-00022.safetensors", "model.layers.31.mlp.down_proj.weight": "model-00008-of-00022.safetensors", "model.layers.51.mlp.up_proj.weight": "model-00008-of-00022.safetensors", "model.layers.31.mlp.up_proj.weight": "model-00008-of-00022.safetensors", "model.layers.51.input_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.31.input_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.50.self_attn.o_proj.weight": "model-00008-of-00022.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00008-of-00022.safetensors", "model.layers.50.self_attn.v_proj.weight": "model-00008-of-00022.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00008-of-00022.safetensors", "model.layers.50.self_attn.k_proj.weight": "model-00008-of-00022.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00008-of-00022.safetensors", "model.layers.50.self_attn.q_proj.weight": "model-00008-of-00022.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00008-of-00022.safetensors", "model.layers.50.post_attention_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.50.mlp.gate_proj.weight": "model-00008-of-00022.safetensors", "model.layers.30.mlp.gate_proj.weight": "model-00008-of-00022.safetensors", "model.layers.50.mlp.down_proj.weight": "model-00008-of-00022.safetensors", "model.layers.30.mlp.down_proj.weight": "model-00008-of-00022.safetensors", "model.layers.50.mlp.up_proj.weight": "model-00008-of-00022.safetensors", "model.layers.30.mlp.up_proj.weight": "model-00008-of-00022.safetensors", "model.layers.50.input_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.30.input_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.49.self_attn.o_proj.weight": "model-00008-of-00022.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00008-of-00022.safetensors", "model.layers.49.self_attn.v_proj.weight": "model-00008-of-00022.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00008-of-00022.safetensors", "model.layers.49.self_attn.k_proj.weight": "model-00008-of-00022.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00008-of-00022.safetensors", "model.layers.49.self_attn.q_proj.weight": "model-00008-of-00022.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00008-of-00022.safetensors", "model.layers.49.post_attention_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.49.mlp.gate_proj.weight": "model-00008-of-00022.safetensors", "model.layers.29.mlp.gate_proj.weight": "model-00008-of-00022.safetensors", "model.layers.49.mlp.down_proj.weight": "model-00009-of-00022.safetensors", "model.layers.29.mlp.down_proj.weight": "model-00009-of-00022.safetensors", "model.layers.49.mlp.up_proj.weight": "model-00009-of-00022.safetensors", "model.layers.29.mlp.up_proj.weight": "model-00009-of-00022.safetensors", "model.layers.49.input_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.29.input_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.48.post_attention_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.48.mlp.gate_proj.weight": "model-00009-of-00022.safetensors", "model.layers.28.mlp.gate_proj.weight": "model-00009-of-00022.safetensors", "model.layers.48.mlp.down_proj.weight": "model-00009-of-00022.safetensors", "model.layers.28.mlp.down_proj.weight": "model-00009-of-00022.safetensors", "model.layers.48.mlp.up_proj.weight": "model-00009-of-00022.safetensors", "model.layers.28.mlp.up_proj.weight": "model-00009-of-00022.safetensors", "model.layers.59.self_attn.k_proj.weight": "model-00009-of-00022.safetensors", "model.layers.39.self_attn.k_proj.weight": "model-00009-of-00022.safetensors", "model.layers.59.post_attention_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.39.post_attention_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.59.mlp.gate_proj.weight": "model-00009-of-00022.safetensors", "model.layers.39.mlp.gate_proj.weight": "model-00009-of-00022.safetensors", "model.layers.59.mlp.down_proj.weight": "model-00009-of-00022.safetensors", "model.layers.39.mlp.down_proj.weight": "model-00009-of-00022.safetensors", "model.layers.59.mlp.up_proj.weight": "model-00009-of-00022.safetensors", "model.layers.39.mlp.up_proj.weight": "model-00009-of-00022.safetensors", "model.layers.59.input_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.39.input_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.58.self_attn.o_proj.weight": "model-00009-of-00022.safetensors", "model.layers.38.self_attn.o_proj.weight": "model-00009-of-00022.safetensors", "model.layers.58.self_attn.v_proj.weight": "model-00009-of-00022.safetensors", "model.layers.38.self_attn.v_proj.weight": "model-00009-of-00022.safetensors", "model.layers.58.self_attn.k_proj.weight": "model-00009-of-00022.safetensors", "model.layers.38.self_attn.k_proj.weight": "model-00009-of-00022.safetensors", "model.layers.58.self_attn.q_proj.weight": "model-00009-of-00022.safetensors", "model.layers.38.self_attn.q_proj.weight": "model-00009-of-00022.safetensors", "model.layers.58.post_attention_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.38.post_attention_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.58.mlp.gate_proj.weight": "model-00009-of-00022.safetensors", "model.layers.38.mlp.gate_proj.weight": "model-00009-of-00022.safetensors", "model.layers.58.mlp.down_proj.weight": "model-00009-of-00022.safetensors", "model.layers.38.mlp.down_proj.weight": "model-00010-of-00022.safetensors", "model.layers.58.mlp.up_proj.weight": "model-00010-of-00022.safetensors", "model.layers.38.mlp.up_proj.weight": "model-00010-of-00022.safetensors", "model.layers.58.input_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.38.input_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.57.post_attention_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.37.post_attention_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.57.mlp.gate_proj.weight": "model-00010-of-00022.safetensors", "model.layers.37.mlp.gate_proj.weight": "model-00010-of-00022.safetensors", "model.layers.57.mlp.down_proj.weight": "model-00010-of-00022.safetensors", "model.layers.37.mlp.down_proj.weight": "model-00010-of-00022.safetensors", "model.layers.57.mlp.up_proj.weight": "model-00010-of-00022.safetensors", "model.layers.37.mlp.up_proj.weight": "model-00010-of-00022.safetensors", "model.layers.57.input_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.37.input_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.56.self_attn.o_proj.weight": "model-00010-of-00022.safetensors", "model.layers.36.self_attn.o_proj.weight": "model-00010-of-00022.safetensors", "model.layers.56.self_attn.v_proj.weight": "model-00010-of-00022.safetensors", "model.layers.36.self_attn.v_proj.weight": "model-00010-of-00022.safetensors", "model.layers.56.self_attn.k_proj.weight": "model-00010-of-00022.safetensors", "model.layers.36.self_attn.k_proj.weight": "model-00010-of-00022.safetensors", "model.layers.56.self_attn.q_proj.weight": "model-00010-of-00022.safetensors", "model.layers.36.self_attn.q_proj.weight": "model-00010-of-00022.safetensors", "model.layers.56.post_attention_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.36.post_attention_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.56.mlp.gate_proj.weight": "model-00010-of-00022.safetensors", "model.layers.36.mlp.gate_proj.weight": "model-00010-of-00022.safetensors", "model.layers.56.mlp.down_proj.weight": "model-00010-of-00022.safetensors", "model.layers.36.mlp.down_proj.weight": "model-00010-of-00022.safetensors", "model.layers.56.mlp.up_proj.weight": "model-00010-of-00022.safetensors", "model.layers.36.mlp.up_proj.weight": "model-00010-of-00022.safetensors", "model.layers.56.input_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.36.input_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.55.self_attn.o_proj.weight": "model-00010-of-00022.safetensors", "model.layers.35.self_attn.o_proj.weight": "model-00010-of-00022.safetensors", "model.layers.55.self_attn.v_proj.weight": "model-00010-of-00022.safetensors", "model.layers.35.self_attn.v_proj.weight": "model-00010-of-00022.safetensors", "model.layers.55.self_attn.k_proj.weight": "model-00010-of-00022.safetensors", "model.layers.35.self_attn.k_proj.weight": "model-00010-of-00022.safetensors", "model.layers.55.self_attn.q_proj.weight": "model-00010-of-00022.safetensors", "model.layers.35.self_attn.q_proj.weight": "model-00010-of-00022.safetensors", "model.layers.55.post_attention_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.35.post_attention_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.55.mlp.gate_proj.weight": "model-00010-of-00022.safetensors", "model.layers.35.mlp.gate_proj.weight": "model-00010-of-00022.safetensors", "model.layers.55.mlp.down_proj.weight": "model-00010-of-00022.safetensors", "model.layers.35.mlp.down_proj.weight": "model-00011-of-00022.safetensors", "model.layers.55.mlp.up_proj.weight": "model-00011-of-00022.safetensors", "model.layers.35.mlp.up_proj.weight": "model-00011-of-00022.safetensors", "model.layers.55.input_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.35.input_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.54.self_attn.o_proj.weight": "model-00011-of-00022.safetensors", "model.layers.34.self_attn.o_proj.weight": "model-00011-of-00022.safetensors", "model.layers.54.self_attn.v_proj.weight": "model-00011-of-00022.safetensors", "model.layers.34.self_attn.v_proj.weight": "model-00011-of-00022.safetensors", "model.layers.54.self_attn.k_proj.weight": "model-00011-of-00022.safetensors", "model.layers.34.self_attn.k_proj.weight": "model-00011-of-00022.safetensors", "model.layers.54.self_attn.q_proj.weight": "model-00011-of-00022.safetensors", "model.layers.34.self_attn.q_proj.weight": "model-00011-of-00022.safetensors", "model.layers.54.post_attention_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.34.post_attention_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.54.mlp.gate_proj.weight": "model-00011-of-00022.safetensors", "model.layers.34.mlp.gate_proj.weight": "model-00011-of-00022.safetensors", "model.layers.54.mlp.down_proj.weight": "model-00011-of-00022.safetensors", "model.layers.34.mlp.down_proj.weight": "model-00011-of-00022.safetensors", "model.layers.54.mlp.up_proj.weight": "model-00011-of-00022.safetensors", "model.layers.34.mlp.up_proj.weight": "model-00011-of-00022.safetensors", "model.layers.85.mlp.gate_proj.weight": "model-00011-of-00022.safetensors", "model.layers.65.mlp.gate_proj.weight": "model-00011-of-00022.safetensors", "model.layers.85.mlp.down_proj.weight": "model-00011-of-00022.safetensors", "model.layers.65.mlp.down_proj.weight": "model-00011-of-00022.safetensors", "model.layers.85.input_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.65.input_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.84.self_attn.o_proj.weight": "model-00011-of-00022.safetensors", "model.layers.64.self_attn.o_proj.weight": "model-00011-of-00022.safetensors", "model.layers.84.self_attn.v_proj.weight": "model-00011-of-00022.safetensors", "model.layers.64.self_attn.v_proj.weight": "model-00011-of-00022.safetensors", "model.layers.84.self_attn.k_proj.weight": "model-00011-of-00022.safetensors", "model.layers.64.self_attn.k_proj.weight": "model-00011-of-00022.safetensors", "model.layers.84.self_attn.q_proj.weight": "model-00011-of-00022.safetensors", "model.layers.64.self_attn.q_proj.weight": "model-00011-of-00022.safetensors", "model.layers.84.post_attention_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.64.post_attention_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.84.mlp.gate_proj.weight": "model-00011-of-00022.safetensors", "model.layers.64.mlp.gate_proj.weight": "model-00011-of-00022.safetensors", "model.layers.84.mlp.down_proj.weight": "model-00011-of-00022.safetensors", "model.layers.64.mlp.down_proj.weight": "model-00011-of-00022.safetensors", "model.layers.84.mlp.up_proj.weight": "model-00011-of-00022.safetensors", "model.layers.64.mlp.up_proj.weight": "model-00012-of-00022.safetensors", "model.layers.84.input_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.64.input_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.83.self_attn.o_proj.weight": "model-00012-of-00022.safetensors", "model.layers.63.self_attn.o_proj.weight": "model-00012-of-00022.safetensors", "model.layers.83.self_attn.v_proj.weight": "model-00012-of-00022.safetensors", "model.layers.63.self_attn.v_proj.weight": "model-00012-of-00022.safetensors", "model.layers.83.self_attn.k_proj.weight": "model-00012-of-00022.safetensors", "model.layers.63.self_attn.k_proj.weight": "model-00012-of-00022.safetensors", "model.layers.83.self_attn.q_proj.weight": "model-00012-of-00022.safetensors", "model.layers.63.self_attn.q_proj.weight": "model-00012-of-00022.safetensors", "model.layers.83.post_attention_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.63.post_attention_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.83.mlp.gate_proj.weight": "model-00012-of-00022.safetensors", "model.layers.63.mlp.gate_proj.weight": "model-00012-of-00022.safetensors", "model.layers.83.mlp.down_proj.weight": "model-00012-of-00022.safetensors", "model.layers.63.mlp.down_proj.weight": "model-00012-of-00022.safetensors", "model.layers.83.mlp.up_proj.weight": "model-00012-of-00022.safetensors", "model.layers.63.mlp.up_proj.weight": "model-00012-of-00022.safetensors", "model.layers.83.input_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.63.input_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.82.self_attn.o_proj.weight": "model-00012-of-00022.safetensors", "model.layers.62.self_attn.o_proj.weight": "model-00012-of-00022.safetensors", "model.layers.82.self_attn.v_proj.weight": "model-00012-of-00022.safetensors", "model.layers.62.self_attn.v_proj.weight": "model-00012-of-00022.safetensors", "model.layers.82.self_attn.k_proj.weight": "model-00012-of-00022.safetensors", "model.layers.62.self_attn.k_proj.weight": "model-00012-of-00022.safetensors", "model.layers.82.self_attn.q_proj.weight": "model-00012-of-00022.safetensors", "model.layers.62.self_attn.q_proj.weight": "model-00012-of-00022.safetensors", "model.layers.82.post_attention_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.62.post_attention_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.82.mlp.gate_proj.weight": "model-00012-of-00022.safetensors", "model.layers.62.mlp.gate_proj.weight": "model-00012-of-00022.safetensors", "model.layers.82.mlp.down_proj.weight": "model-00012-of-00022.safetensors", "model.layers.62.mlp.down_proj.weight": "model-00012-of-00022.safetensors", "model.layers.82.mlp.up_proj.weight": "model-00012-of-00022.safetensors", "model.layers.62.mlp.up_proj.weight": "model-00012-of-00022.safetensors", "model.layers.82.input_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.62.input_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.81.self_attn.o_proj.weight": "model-00012-of-00022.safetensors", "model.layers.61.self_attn.o_proj.weight": "model-00012-of-00022.safetensors", "model.layers.81.self_attn.v_proj.weight": "model-00012-of-00022.safetensors", "model.layers.61.self_attn.v_proj.weight": "model-00012-of-00022.safetensors", "model.layers.81.self_attn.k_proj.weight": "model-00012-of-00022.safetensors", "model.layers.61.self_attn.k_proj.weight": "model-00012-of-00022.safetensors", "model.layers.81.self_attn.q_proj.weight": "model-00012-of-00022.safetensors", "model.layers.61.self_attn.q_proj.weight": "model-00012-of-00022.safetensors", "model.layers.81.post_attention_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.61.post_attention_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.81.mlp.gate_proj.weight": "model-00012-of-00022.safetensors", "model.layers.61.mlp.gate_proj.weight": "model-00012-of-00022.safetensors", "model.layers.81.mlp.down_proj.weight": "model-00012-of-00022.safetensors", "model.layers.61.mlp.down_proj.weight": "model-00012-of-00022.safetensors", "model.layers.81.mlp.up_proj.weight": "model-00013-of-00022.safetensors", "model.layers.61.mlp.up_proj.weight": "model-00013-of-00022.safetensors", "model.layers.81.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.61.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.80.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.60.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.80.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.60.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.80.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.60.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.80.self_attn.q_proj.weight": "model-00013-of-00022.safetensors", "model.layers.60.self_attn.q_proj.weight": "model-00013-of-00022.safetensors", "model.layers.80.post_attention_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.60.post_attention_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.80.mlp.gate_proj.weight": "model-00013-of-00022.safetensors", "model.layers.60.mlp.gate_proj.weight": "model-00013-of-00022.safetensors", "model.layers.80.mlp.down_proj.weight": "model-00013-of-00022.safetensors", "model.layers.60.mlp.down_proj.weight": "model-00013-of-00022.safetensors", "model.layers.80.mlp.up_proj.weight": "model-00013-of-00022.safetensors", "model.layers.60.mlp.up_proj.weight": "model-00013-of-00022.safetensors", "model.layers.80.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.60.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.59.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.39.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.59.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.39.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.59.self_attn.q_proj.weight": "model-00013-of-00022.safetensors", "model.layers.39.self_attn.q_proj.weight": "model-00013-of-00022.safetensors", "model.layers.96.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.76.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.96.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.76.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.96.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.76.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.91.mlp.down_proj.weight": "model-00013-of-00022.safetensors", "model.layers.71.mlp.down_proj.weight": "model-00013-of-00022.safetensors", "model.layers.91.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.71.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.90.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.70.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.90.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.70.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.90.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.70.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.90.self_attn.q_proj.weight": "model-00013-of-00022.safetensors", "model.layers.70.self_attn.q_proj.weight": "model-00013-of-00022.safetensors", "model.layers.90.post_attention_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.70.post_attention_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.90.mlp.gate_proj.weight": "model-00013-of-00022.safetensors", "model.layers.70.mlp.gate_proj.weight": "model-00013-of-00022.safetensors", "model.layers.90.mlp.down_proj.weight": "model-00013-of-00022.safetensors", "model.layers.70.mlp.down_proj.weight": "model-00013-of-00022.safetensors", "model.layers.90.mlp.up_proj.weight": "model-00013-of-00022.safetensors", "model.layers.70.mlp.up_proj.weight": "model-00013-of-00022.safetensors", "model.layers.90.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.70.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.89.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.69.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.89.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.69.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.89.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.69.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.89.self_attn.q_proj.weight": "model-00014-of-00022.safetensors", "model.layers.69.self_attn.q_proj.weight": "model-00014-of-00022.safetensors", "model.layers.89.post_attention_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.69.post_attention_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.89.mlp.gate_proj.weight": "model-00014-of-00022.safetensors", "model.layers.69.mlp.gate_proj.weight": "model-00014-of-00022.safetensors", "model.layers.89.mlp.down_proj.weight": "model-00014-of-00022.safetensors", "model.layers.69.mlp.down_proj.weight": "model-00014-of-00022.safetensors", "model.layers.89.mlp.up_proj.weight": "model-00014-of-00022.safetensors", "model.layers.69.mlp.up_proj.weight": "model-00014-of-00022.safetensors", "model.layers.89.input_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.69.input_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.88.self_attn.o_proj.weight": "model-00014-of-00022.safetensors", "model.layers.68.self_attn.o_proj.weight": "model-00014-of-00022.safetensors", "model.layers.88.self_attn.v_proj.weight": "model-00014-of-00022.safetensors", "model.layers.68.self_attn.v_proj.weight": "model-00014-of-00022.safetensors", "model.layers.88.self_attn.k_proj.weight": "model-00014-of-00022.safetensors", "model.layers.68.self_attn.k_proj.weight": "model-00014-of-00022.safetensors", "model.layers.88.self_attn.q_proj.weight": "model-00014-of-00022.safetensors", "model.layers.68.self_attn.q_proj.weight": "model-00014-of-00022.safetensors", "model.layers.88.post_attention_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.68.post_attention_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.88.mlp.gate_proj.weight": "model-00014-of-00022.safetensors", "model.layers.68.mlp.gate_proj.weight": "model-00014-of-00022.safetensors", "model.layers.88.mlp.down_proj.weight": "model-00014-of-00022.safetensors", "model.layers.68.mlp.down_proj.weight": "model-00014-of-00022.safetensors", "model.layers.88.mlp.up_proj.weight": "model-00014-of-00022.safetensors", "model.layers.68.mlp.up_proj.weight": "model-00014-of-00022.safetensors", "model.layers.88.input_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.68.input_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.87.self_attn.o_proj.weight": "model-00014-of-00022.safetensors", "model.layers.67.self_attn.o_proj.weight": "model-00014-of-00022.safetensors", "model.layers.87.self_attn.v_proj.weight": "model-00014-of-00022.safetensors", "model.layers.67.self_attn.v_proj.weight": "model-00014-of-00022.safetensors", "model.layers.87.self_attn.k_proj.weight": "model-00014-of-00022.safetensors", "model.layers.67.self_attn.k_proj.weight": "model-00014-of-00022.safetensors", "model.layers.87.self_attn.q_proj.weight": "model-00014-of-00022.safetensors", "model.layers.67.self_attn.q_proj.weight": "model-00014-of-00022.safetensors", "model.layers.87.post_attention_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.67.post_attention_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.87.mlp.gate_proj.weight": "model-00014-of-00022.safetensors", "model.layers.67.mlp.gate_proj.weight": "model-00014-of-00022.safetensors", "model.layers.87.mlp.down_proj.weight": "model-00014-of-00022.safetensors", "model.layers.67.mlp.down_proj.weight": "model-00014-of-00022.safetensors", "model.layers.87.mlp.up_proj.weight": "model-00014-of-00022.safetensors", "model.layers.67.mlp.up_proj.weight": "model-00014-of-00022.safetensors", "model.layers.87.input_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.67.input_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.86.self_attn.o_proj.weight": "model-00015-of-00022.safetensors", "model.layers.66.self_attn.o_proj.weight": "model-00015-of-00022.safetensors", "model.layers.86.self_attn.v_proj.weight": "model-00015-of-00022.safetensors", "model.layers.66.self_attn.v_proj.weight": "model-00015-of-00022.safetensors", "model.layers.86.self_attn.k_proj.weight": "model-00015-of-00022.safetensors", "model.layers.66.self_attn.k_proj.weight": "model-00015-of-00022.safetensors", "model.layers.86.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.66.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.86.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.66.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.86.mlp.gate_proj.weight": "model-00015-of-00022.safetensors", "model.layers.66.mlp.gate_proj.weight": "model-00015-of-00022.safetensors", "model.layers.86.mlp.down_proj.weight": "model-00015-of-00022.safetensors", "model.layers.66.mlp.down_proj.weight": "model-00015-of-00022.safetensors", "model.layers.86.mlp.up_proj.weight": "model-00015-of-00022.safetensors", "model.layers.66.mlp.up_proj.weight": "model-00015-of-00022.safetensors", "model.layers.86.input_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.66.input_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.85.self_attn.o_proj.weight": "model-00015-of-00022.safetensors", "model.layers.65.self_attn.o_proj.weight": "model-00015-of-00022.safetensors", "model.layers.85.self_attn.v_proj.weight": "model-00015-of-00022.safetensors", "model.layers.65.self_attn.v_proj.weight": "model-00015-of-00022.safetensors", "model.layers.85.self_attn.k_proj.weight": "model-00015-of-00022.safetensors", "model.layers.65.self_attn.k_proj.weight": "model-00015-of-00022.safetensors", "model.layers.85.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.65.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.85.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.65.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.85.mlp.up_proj.weight": "model-00015-of-00022.safetensors", "model.layers.65.mlp.up_proj.weight": "model-00015-of-00022.safetensors", "model.layers.101.self_attn.o_proj.weight": "model-00015-of-00022.safetensors", "model.layers.101.self_attn.v_proj.weight": "model-00015-of-00022.safetensors", "model.layers.101.self_attn.k_proj.weight": "model-00015-of-00022.safetensors", "model.layers.101.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.97.input_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.77.input_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.96.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.76.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.96.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.76.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.96.mlp.gate_proj.weight": "model-00015-of-00022.safetensors", "model.layers.76.mlp.gate_proj.weight": "model-00015-of-00022.safetensors", "model.layers.96.mlp.down_proj.weight": "model-00015-of-00022.safetensors", "model.layers.76.mlp.down_proj.weight": "model-00015-of-00022.safetensors", "model.layers.96.mlp.up_proj.weight": "model-00015-of-00022.safetensors", "model.layers.76.mlp.up_proj.weight": "model-00015-of-00022.safetensors", "model.layers.96.input_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.76.input_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.95.self_attn.o_proj.weight": "model-00015-of-00022.safetensors", "model.layers.75.self_attn.o_proj.weight": "model-00015-of-00022.safetensors", "model.layers.95.self_attn.v_proj.weight": "model-00015-of-00022.safetensors", "model.layers.75.self_attn.v_proj.weight": "model-00015-of-00022.safetensors", "model.layers.95.self_attn.k_proj.weight": "model-00015-of-00022.safetensors", "model.layers.75.self_attn.k_proj.weight": "model-00015-of-00022.safetensors", "model.layers.95.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.75.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.95.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.75.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.95.mlp.gate_proj.weight": "model-00015-of-00022.safetensors", "model.layers.75.mlp.gate_proj.weight": "model-00015-of-00022.safetensors", "model.layers.95.mlp.down_proj.weight": "model-00016-of-00022.safetensors", "model.layers.75.mlp.down_proj.weight": "model-00016-of-00022.safetensors", "model.layers.95.mlp.up_proj.weight": "model-00016-of-00022.safetensors", "model.layers.75.mlp.up_proj.weight": "model-00016-of-00022.safetensors", "model.layers.95.input_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.75.input_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.94.self_attn.o_proj.weight": "model-00016-of-00022.safetensors", "model.layers.74.self_attn.o_proj.weight": "model-00016-of-00022.safetensors", "model.layers.94.self_attn.v_proj.weight": "model-00016-of-00022.safetensors", "model.layers.74.self_attn.v_proj.weight": "model-00016-of-00022.safetensors", "model.layers.94.self_attn.k_proj.weight": "model-00016-of-00022.safetensors", "model.layers.74.self_attn.k_proj.weight": "model-00016-of-00022.safetensors", "model.layers.94.self_attn.q_proj.weight": "model-00016-of-00022.safetensors", "model.layers.74.self_attn.q_proj.weight": "model-00016-of-00022.safetensors", "model.layers.94.post_attention_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.74.post_attention_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.94.mlp.gate_proj.weight": "model-00016-of-00022.safetensors", "model.layers.74.mlp.gate_proj.weight": "model-00016-of-00022.safetensors", "model.layers.94.mlp.down_proj.weight": "model-00016-of-00022.safetensors", "model.layers.74.mlp.down_proj.weight": "model-00016-of-00022.safetensors", "model.layers.94.mlp.up_proj.weight": "model-00016-of-00022.safetensors", "model.layers.74.mlp.up_proj.weight": "model-00016-of-00022.safetensors", "model.layers.94.input_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.74.input_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.93.self_attn.o_proj.weight": "model-00016-of-00022.safetensors", "model.layers.73.self_attn.o_proj.weight": "model-00016-of-00022.safetensors", "model.layers.93.self_attn.v_proj.weight": "model-00016-of-00022.safetensors", "model.layers.73.self_attn.v_proj.weight": "model-00016-of-00022.safetensors", "model.layers.93.self_attn.k_proj.weight": "model-00016-of-00022.safetensors", "model.layers.73.self_attn.k_proj.weight": "model-00016-of-00022.safetensors", "model.layers.93.self_attn.q_proj.weight": "model-00016-of-00022.safetensors", "model.layers.73.self_attn.q_proj.weight": "model-00016-of-00022.safetensors", "model.layers.93.post_attention_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.73.post_attention_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.93.mlp.gate_proj.weight": "model-00016-of-00022.safetensors", "model.layers.73.mlp.gate_proj.weight": "model-00016-of-00022.safetensors", "model.layers.93.mlp.down_proj.weight": "model-00016-of-00022.safetensors", "model.layers.73.mlp.down_proj.weight": "model-00016-of-00022.safetensors", "model.layers.93.mlp.up_proj.weight": "model-00016-of-00022.safetensors", "model.layers.73.mlp.up_proj.weight": "model-00016-of-00022.safetensors", "model.layers.93.input_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.73.input_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.92.self_attn.o_proj.weight": "model-00016-of-00022.safetensors", "model.layers.72.self_attn.o_proj.weight": "model-00016-of-00022.safetensors", "model.layers.92.self_attn.v_proj.weight": "model-00016-of-00022.safetensors", "model.layers.72.self_attn.v_proj.weight": "model-00016-of-00022.safetensors", "model.layers.92.self_attn.k_proj.weight": "model-00016-of-00022.safetensors", "model.layers.72.self_attn.k_proj.weight": "model-00016-of-00022.safetensors", "model.layers.92.self_attn.q_proj.weight": "model-00016-of-00022.safetensors", "model.layers.72.self_attn.q_proj.weight": "model-00016-of-00022.safetensors", "model.layers.92.post_attention_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.72.post_attention_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.92.mlp.gate_proj.weight": "model-00016-of-00022.safetensors", "model.layers.72.mlp.gate_proj.weight": "model-00017-of-00022.safetensors", "model.layers.92.mlp.down_proj.weight": "model-00017-of-00022.safetensors", "model.layers.72.mlp.down_proj.weight": "model-00017-of-00022.safetensors", "model.layers.92.mlp.up_proj.weight": "model-00017-of-00022.safetensors", "model.layers.72.mlp.up_proj.weight": "model-00017-of-00022.safetensors", "model.layers.92.input_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.72.input_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.91.self_attn.o_proj.weight": "model-00017-of-00022.safetensors", "model.layers.71.self_attn.o_proj.weight": "model-00017-of-00022.safetensors", "model.layers.91.self_attn.v_proj.weight": "model-00017-of-00022.safetensors", "model.layers.71.self_attn.v_proj.weight": "model-00017-of-00022.safetensors", "model.layers.91.self_attn.k_proj.weight": "model-00017-of-00022.safetensors", "model.layers.71.self_attn.k_proj.weight": "model-00017-of-00022.safetensors", "model.layers.91.self_attn.q_proj.weight": "model-00017-of-00022.safetensors", "model.layers.71.self_attn.q_proj.weight": "model-00017-of-00022.safetensors", "model.layers.91.post_attention_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.71.post_attention_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.91.mlp.gate_proj.weight": "model-00017-of-00022.safetensors", "model.layers.71.mlp.gate_proj.weight": "model-00017-of-00022.safetensors", "model.layers.91.mlp.up_proj.weight": "model-00017-of-00022.safetensors", "model.layers.71.mlp.up_proj.weight": "model-00017-of-00022.safetensors", "model.layers.102.self_attn.k_proj.weight": "model-00017-of-00022.safetensors", "model.layers.102.post_attention_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.102.mlp.gate_proj.weight": "model-00017-of-00022.safetensors", "model.layers.102.mlp.down_proj.weight": "model-00017-of-00022.safetensors", "model.layers.102.mlp.up_proj.weight": "model-00017-of-00022.safetensors", "model.layers.102.input_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.101.post_attention_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.101.mlp.gate_proj.weight": "model-00017-of-00022.safetensors", "model.layers.101.mlp.down_proj.weight": "model-00017-of-00022.safetensors", "model.layers.101.mlp.up_proj.weight": "model-00017-of-00022.safetensors", "model.layers.101.input_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.100.self_attn.o_proj.weight": "model-00017-of-00022.safetensors", "model.layers.100.self_attn.v_proj.weight": "model-00017-of-00022.safetensors", "model.layers.100.self_attn.k_proj.weight": "model-00017-of-00022.safetensors", "model.layers.100.self_attn.q_proj.weight": "model-00017-of-00022.safetensors", "model.layers.100.post_attention_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.100.mlp.gate_proj.weight": "model-00017-of-00022.safetensors", "model.layers.100.mlp.down_proj.weight": "model-00017-of-00022.safetensors", "model.layers.100.mlp.up_proj.weight": "model-00017-of-00022.safetensors", "model.layers.100.input_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.99.self_attn.o_proj.weight": "model-00017-of-00022.safetensors", "model.layers.79.self_attn.o_proj.weight": "model-00017-of-00022.safetensors", "model.layers.99.self_attn.v_proj.weight": "model-00017-of-00022.safetensors", "model.layers.79.self_attn.v_proj.weight": "model-00017-of-00022.safetensors", "model.layers.99.self_attn.k_proj.weight": "model-00017-of-00022.safetensors", "model.layers.79.self_attn.k_proj.weight": "model-00017-of-00022.safetensors", "model.layers.99.self_attn.q_proj.weight": "model-00017-of-00022.safetensors", "model.layers.79.self_attn.q_proj.weight": "model-00017-of-00022.safetensors", "model.layers.99.post_attention_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.79.post_attention_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.99.mlp.gate_proj.weight": "model-00018-of-00022.safetensors", "model.layers.79.mlp.gate_proj.weight": "model-00018-of-00022.safetensors", "model.layers.99.mlp.down_proj.weight": "model-00018-of-00022.safetensors", "model.layers.79.mlp.down_proj.weight": "model-00018-of-00022.safetensors", "model.layers.99.mlp.up_proj.weight": "model-00018-of-00022.safetensors", "model.layers.79.mlp.up_proj.weight": "model-00018-of-00022.safetensors", "model.layers.99.input_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.79.input_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.98.self_attn.o_proj.weight": "model-00018-of-00022.safetensors", "model.layers.78.self_attn.o_proj.weight": "model-00018-of-00022.safetensors", "model.layers.98.self_attn.v_proj.weight": "model-00018-of-00022.safetensors", "model.layers.78.self_attn.v_proj.weight": "model-00018-of-00022.safetensors", "model.layers.98.self_attn.k_proj.weight": "model-00018-of-00022.safetensors", "model.layers.78.self_attn.k_proj.weight": "model-00018-of-00022.safetensors", "model.layers.98.self_attn.q_proj.weight": "model-00018-of-00022.safetensors", "model.layers.78.self_attn.q_proj.weight": "model-00018-of-00022.safetensors", "model.layers.98.post_attention_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.78.post_attention_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.98.mlp.gate_proj.weight": "model-00018-of-00022.safetensors", "model.layers.78.mlp.gate_proj.weight": "model-00018-of-00022.safetensors", "model.layers.98.mlp.down_proj.weight": "model-00018-of-00022.safetensors", "model.layers.78.mlp.down_proj.weight": "model-00018-of-00022.safetensors", "model.layers.98.mlp.up_proj.weight": "model-00018-of-00022.safetensors", "model.layers.78.mlp.up_proj.weight": "model-00018-of-00022.safetensors", "model.layers.98.input_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.78.input_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.97.self_attn.o_proj.weight": "model-00018-of-00022.safetensors", "model.layers.77.self_attn.o_proj.weight": "model-00018-of-00022.safetensors", "model.layers.97.self_attn.v_proj.weight": "model-00018-of-00022.safetensors", "model.layers.77.self_attn.v_proj.weight": "model-00018-of-00022.safetensors", "model.layers.97.self_attn.k_proj.weight": "model-00018-of-00022.safetensors", "model.layers.77.self_attn.k_proj.weight": "model-00018-of-00022.safetensors", "model.layers.97.self_attn.q_proj.weight": "model-00018-of-00022.safetensors", "model.layers.77.self_attn.q_proj.weight": "model-00018-of-00022.safetensors", "model.layers.97.post_attention_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.77.post_attention_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.97.mlp.gate_proj.weight": "model-00018-of-00022.safetensors", "model.layers.77.mlp.gate_proj.weight": "model-00018-of-00022.safetensors", "model.layers.97.mlp.down_proj.weight": "model-00018-of-00022.safetensors", "model.layers.77.mlp.down_proj.weight": "model-00018-of-00022.safetensors", "model.layers.97.mlp.up_proj.weight": "model-00018-of-00022.safetensors", "model.layers.77.mlp.up_proj.weight": "model-00018-of-00022.safetensors", "model.layers.108.mlp.gate_proj.weight": "model-00019-of-00022.safetensors", "model.layers.108.mlp.down_proj.weight": "model-00019-of-00022.safetensors", "model.layers.108.input_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.107.self_attn.o_proj.weight": "model-00019-of-00022.safetensors", "model.layers.107.self_attn.v_proj.weight": "model-00019-of-00022.safetensors", "model.layers.107.self_attn.k_proj.weight": "model-00019-of-00022.safetensors", "model.layers.107.self_attn.q_proj.weight": "model-00019-of-00022.safetensors", "model.layers.107.post_attention_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.107.mlp.gate_proj.weight": "model-00019-of-00022.safetensors", "model.layers.107.mlp.down_proj.weight": "model-00019-of-00022.safetensors", "model.layers.107.mlp.up_proj.weight": "model-00019-of-00022.safetensors", "model.layers.107.input_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.106.self_attn.o_proj.weight": "model-00019-of-00022.safetensors", "model.layers.106.self_attn.v_proj.weight": "model-00019-of-00022.safetensors", "model.layers.106.self_attn.k_proj.weight": "model-00019-of-00022.safetensors", "model.layers.106.self_attn.q_proj.weight": "model-00019-of-00022.safetensors", "model.layers.106.post_attention_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.106.mlp.gate_proj.weight": "model-00019-of-00022.safetensors", "model.layers.106.mlp.down_proj.weight": "model-00019-of-00022.safetensors", "model.layers.106.mlp.up_proj.weight": "model-00019-of-00022.safetensors", "model.layers.106.input_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.105.self_attn.o_proj.weight": "model-00019-of-00022.safetensors", "model.layers.105.self_attn.v_proj.weight": "model-00019-of-00022.safetensors", "model.layers.105.self_attn.k_proj.weight": "model-00019-of-00022.safetensors", "model.layers.105.self_attn.q_proj.weight": "model-00019-of-00022.safetensors", "model.layers.105.post_attention_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.105.mlp.gate_proj.weight": "model-00019-of-00022.safetensors", "model.layers.105.mlp.down_proj.weight": "model-00019-of-00022.safetensors", "model.layers.105.mlp.up_proj.weight": "model-00019-of-00022.safetensors", "model.layers.105.input_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.104.self_attn.o_proj.weight": "model-00019-of-00022.safetensors", "model.layers.104.self_attn.v_proj.weight": "model-00019-of-00022.safetensors", "model.layers.104.self_attn.k_proj.weight": "model-00019-of-00022.safetensors", "model.layers.104.self_attn.q_proj.weight": "model-00019-of-00022.safetensors", "model.layers.104.post_attention_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.104.mlp.gate_proj.weight": "model-00019-of-00022.safetensors", "model.layers.104.mlp.down_proj.weight": "model-00019-of-00022.safetensors", "model.layers.104.mlp.up_proj.weight": "model-00019-of-00022.safetensors", "model.layers.104.input_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.103.self_attn.o_proj.weight": "model-00019-of-00022.safetensors", "model.layers.103.self_attn.v_proj.weight": "model-00019-of-00022.safetensors", "model.layers.103.self_attn.k_proj.weight": "model-00019-of-00022.safetensors", "model.layers.103.self_attn.q_proj.weight": "model-00019-of-00022.safetensors", "model.layers.103.post_attention_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.103.mlp.gate_proj.weight": "model-00019-of-00022.safetensors", "model.layers.103.mlp.down_proj.weight": "model-00019-of-00022.safetensors", "model.layers.103.mlp.up_proj.weight": "model-00019-of-00022.safetensors", "model.layers.103.input_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.102.self_attn.o_proj.weight": "model-00019-of-00022.safetensors", "model.layers.102.self_attn.v_proj.weight": "model-00019-of-00022.safetensors", "model.layers.102.self_attn.q_proj.weight": "model-00019-of-00022.safetensors", "model.layers.114.mlp.down_proj.weight": "model-00020-of-00022.safetensors", "model.layers.114.input_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.113.self_attn.o_proj.weight": "model-00020-of-00022.safetensors", "model.layers.113.self_attn.v_proj.weight": "model-00020-of-00022.safetensors", "model.layers.113.self_attn.k_proj.weight": "model-00020-of-00022.safetensors", "model.layers.113.self_attn.q_proj.weight": "model-00020-of-00022.safetensors", "model.layers.113.post_attention_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.113.mlp.gate_proj.weight": "model-00020-of-00022.safetensors", "model.layers.113.mlp.down_proj.weight": "model-00020-of-00022.safetensors", "model.layers.113.mlp.up_proj.weight": "model-00020-of-00022.safetensors", "model.layers.113.input_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.112.self_attn.o_proj.weight": "model-00020-of-00022.safetensors", "model.layers.112.self_attn.v_proj.weight": "model-00020-of-00022.safetensors", "model.layers.112.self_attn.k_proj.weight": "model-00020-of-00022.safetensors", "model.layers.112.self_attn.q_proj.weight": "model-00020-of-00022.safetensors", "model.layers.112.post_attention_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.112.mlp.gate_proj.weight": "model-00020-of-00022.safetensors", "model.layers.112.mlp.down_proj.weight": "model-00020-of-00022.safetensors", "model.layers.112.mlp.up_proj.weight": "model-00020-of-00022.safetensors", "model.layers.112.input_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.111.self_attn.o_proj.weight": "model-00020-of-00022.safetensors", "model.layers.111.self_attn.v_proj.weight": "model-00020-of-00022.safetensors", "model.layers.111.self_attn.k_proj.weight": "model-00020-of-00022.safetensors", "model.layers.111.self_attn.q_proj.weight": "model-00020-of-00022.safetensors", "model.layers.111.post_attention_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.111.mlp.gate_proj.weight": "model-00020-of-00022.safetensors", "model.layers.111.mlp.down_proj.weight": "model-00020-of-00022.safetensors", "model.layers.111.mlp.up_proj.weight": "model-00020-of-00022.safetensors", "model.layers.111.input_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.110.self_attn.o_proj.weight": "model-00020-of-00022.safetensors", "model.layers.110.self_attn.v_proj.weight": "model-00020-of-00022.safetensors", "model.layers.110.self_attn.k_proj.weight": "model-00020-of-00022.safetensors", "model.layers.110.self_attn.q_proj.weight": "model-00020-of-00022.safetensors", "model.layers.110.post_attention_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.110.mlp.gate_proj.weight": "model-00020-of-00022.safetensors", "model.layers.110.mlp.down_proj.weight": "model-00020-of-00022.safetensors", "model.layers.110.mlp.up_proj.weight": "model-00020-of-00022.safetensors", "model.layers.110.input_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.109.self_attn.o_proj.weight": "model-00020-of-00022.safetensors", "model.layers.109.self_attn.v_proj.weight": "model-00020-of-00022.safetensors", "model.layers.109.self_attn.k_proj.weight": "model-00020-of-00022.safetensors", "model.layers.109.self_attn.q_proj.weight": "model-00020-of-00022.safetensors", "model.layers.109.post_attention_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.109.mlp.gate_proj.weight": "model-00020-of-00022.safetensors", "model.layers.109.mlp.down_proj.weight": "model-00020-of-00022.safetensors", "model.layers.109.mlp.up_proj.weight": "model-00020-of-00022.safetensors", "model.layers.109.input_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.108.self_attn.o_proj.weight": "model-00020-of-00022.safetensors", "model.layers.108.self_attn.v_proj.weight": "model-00020-of-00022.safetensors", "model.layers.108.self_attn.k_proj.weight": "model-00020-of-00022.safetensors", "model.layers.108.self_attn.q_proj.weight": "model-00020-of-00022.safetensors", "model.layers.108.post_attention_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.108.mlp.up_proj.weight": "model-00020-of-00022.safetensors", "lm_head.weight": "model-00021-of-00022.safetensors", "model.norm.weight": "model-00021-of-00022.safetensors", "model.layers.119.self_attn.o_proj.weight": "model-00021-of-00022.safetensors", "model.layers.119.self_attn.v_proj.weight": "model-00021-of-00022.safetensors", "model.layers.119.self_attn.k_proj.weight": "model-00021-of-00022.safetensors", "model.layers.119.self_attn.q_proj.weight": "model-00021-of-00022.safetensors", "model.layers.119.post_attention_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.119.mlp.gate_proj.weight": "model-00021-of-00022.safetensors", "model.layers.119.mlp.down_proj.weight": "model-00021-of-00022.safetensors", "model.layers.119.input_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.118.self_attn.o_proj.weight": "model-00021-of-00022.safetensors", "model.layers.118.self_attn.v_proj.weight": "model-00021-of-00022.safetensors", "model.layers.118.self_attn.k_proj.weight": "model-00021-of-00022.safetensors", "model.layers.118.self_attn.q_proj.weight": "model-00021-of-00022.safetensors", "model.layers.118.post_attention_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.118.mlp.gate_proj.weight": "model-00021-of-00022.safetensors", "model.layers.118.mlp.down_proj.weight": "model-00021-of-00022.safetensors", "model.layers.118.mlp.up_proj.weight": "model-00021-of-00022.safetensors", "model.layers.118.input_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.117.self_attn.o_proj.weight": "model-00021-of-00022.safetensors", "model.layers.117.self_attn.v_proj.weight": "model-00021-of-00022.safetensors", "model.layers.117.self_attn.k_proj.weight": "model-00021-of-00022.safetensors", "model.layers.117.self_attn.q_proj.weight": "model-00021-of-00022.safetensors", "model.layers.117.post_attention_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.117.mlp.gate_proj.weight": "model-00021-of-00022.safetensors", "model.layers.117.mlp.down_proj.weight": "model-00021-of-00022.safetensors", "model.layers.117.mlp.up_proj.weight": "model-00021-of-00022.safetensors", "model.layers.117.input_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.116.self_attn.o_proj.weight": "model-00021-of-00022.safetensors", "model.layers.116.self_attn.v_proj.weight": "model-00021-of-00022.safetensors", "model.layers.116.self_attn.k_proj.weight": "model-00021-of-00022.safetensors", "model.layers.116.self_attn.q_proj.weight": "model-00021-of-00022.safetensors", "model.layers.116.post_attention_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.116.mlp.gate_proj.weight": "model-00021-of-00022.safetensors", "model.layers.116.mlp.down_proj.weight": "model-00021-of-00022.safetensors", "model.layers.116.mlp.up_proj.weight": "model-00021-of-00022.safetensors", "model.layers.116.input_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.115.self_attn.o_proj.weight": "model-00021-of-00022.safetensors", "model.layers.115.self_attn.v_proj.weight": "model-00021-of-00022.safetensors", "model.layers.115.self_attn.k_proj.weight": "model-00021-of-00022.safetensors", "model.layers.115.self_attn.q_proj.weight": "model-00021-of-00022.safetensors", "model.layers.115.post_attention_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.115.mlp.gate_proj.weight": "model-00021-of-00022.safetensors", "model.layers.115.mlp.down_proj.weight": "model-00021-of-00022.safetensors", "model.layers.115.mlp.up_proj.weight": "model-00021-of-00022.safetensors", "model.layers.115.input_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.114.self_attn.o_proj.weight": "model-00021-of-00022.safetensors", "model.layers.114.self_attn.v_proj.weight": "model-00021-of-00022.safetensors", "model.layers.114.self_attn.k_proj.weight": "model-00021-of-00022.safetensors", "model.layers.114.self_attn.q_proj.weight": "model-00021-of-00022.safetensors", "model.layers.114.post_attention_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.114.mlp.gate_proj.weight": "model-00021-of-00022.safetensors", "model.layers.114.mlp.up_proj.weight": "model-00021-of-00022.safetensors", "model.layers.119.mlp.up_proj.weight": "model-00022-of-00022.safetensors"}}

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

tokenizer_config.json

@@ -0,0 +1,41 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<unk>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}