[WIP] Upload folder using huggingface_hub (multi-commit fd4b4972122e62602c40b41b4bca9ec9a29f0a9f657f94d1a1d4a33b9279b300)

.gitattributes

@@ -34,4 +34,3 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 aranea-tenebris.png filter=lfs diff=lfs merge=lfs -text
-imatrix.dat filter=lfs diff=lfs merge=lfs -text

README.md

@@ -10,7 +10,7 @@ tags:
 - mergekit
 - merge
 ---
-# aranea-tenebris-120b-v1.0
+# aranea-tenebris-116b-v1.0
 **aka Netrve/Miqu-PlayMaid-70B-v0.1 + ShinojiResearch/Senku-70B**  
 Model merge for uncensored creative writing and rp  
 
@@ -27,14 +27,6 @@ This model outperformed a number of other popular 70B+ models and merges in both
 - Recommended prompt format: Alpaca
 - Layers: 137
 
-
-### Quantization
-  
-llama.cpp [imatrix.dat](./imatrix.dat)  
-exllamav2 [measurement.json](./measurement.json)  
-
-Will upload a few quants when bandwidth permits.
-
 ### Testing
 
 Two different writing styles were considered for each testing scenario:
@@ -45,7 +37,7 @@ Tests assumed a mature audience, but a range of scenarios were constructed.
 Thematic inconsistancy or bias in character behaviour was penalized heavily.  
 
 Models showing the following were penalized during manual comparison:
-- Consistently short responses.
+- Consistently short responses
 - Laziness or readily gave up on solving a character problem.
 - Overly malleable, where characters could not hold opinions or beliefs.
 - Passiveness or an inability to drive the narrative.
@@ -61,3 +53,7 @@ Models showing the following were penalized during manual comparison:
 - 8 layer interleave stride with a 16 layer interleave width consistantly outperformed alternative combinations for creativity whilst remaining reasonably coherant.
 - Regular stride intervals are not optimal. In particular offsetting the first or last set of base models offets often improved metrics.
 - Goliath-120B is still a good standard for coherancy below 4096 context. A few miqu-1 merges are comparable, but testing found a small amount coherancy could be sacrificed for notable creativity improvements.
+
+### Quantizations
+
+Exllamav2 quants will be available when bandwidth permits.

config.json

@@ -1,29 +0,0 @@
-{
-  "_name_or_path": "Miqu-PlayMaid-70B-v0.1",
-  "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": 32764,
-  "model_type": "llama",
-  "num_attention_heads": 64,
-  "num_hidden_layers": 138,
-  "num_key_value_heads": 8,
-  "pad_token_id": 0,
-  "pretraining_tp": 1,
-  "rms_norm_eps": 1e-05,
-  "rope_scaling": null,
-  "rope_theta": 1000000,
-  "tie_word_embeddings": false,
-  "torch_dtype": "float16",
-  "transformers_version": "4.40.0.dev0",
-  "use_cache": false,
-  "vocab_size": 32000
-}

imatrix.dat

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:d01fad66e09e09ee5cff5bf080fdc60c96ac80659f62da9919bc30a8c22ff1e5
-size 42991202

mergekit_config.yml

@@ -1,33 +0,0 @@
-merge_method: linear
-parameters:
-  weight: 1.0
-slices:
-  - sources:
-      - model: Miqu-PlayMaid-70B-v0.1
-        layer_range: [0, 17]
-  - sources:
-      - model: Senku-70B-Full
-        layer_range: [10, 24]
-  - sources:
-      - model: Miqu-PlayMaid-70B-v0.1
-        layer_range: [17, 32]
-  - sources:
-      - model: Senku-70B-Full
-        layer_range: [24, 40]
-  - sources:
-      - model: Miqu-PlayMaid-70B-v0.1
-        layer_range: [32, 48]
-  - sources:
-      - model: Senku-70B-Full
-        layer_range: [40, 56]
-  - sources:
-      - model: Miqu-PlayMaid-70B-v0.1
-        layer_range: [49, 63]
-  - sources:
-      - model: Senku-70B-Full
-        layer_range: [56, 70]
-  - sources:
-      - model: Miqu-PlayMaid-70B-v0.1
-        layer_range: [64, 80]
-dtype: float16
-tokenizer_source: model:Miqu-PlayMaid-70B-v0.1

model-00001-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:f8a17293e30974f82c450405cc182e20de23611640a5288d1be933f88d4178a5
-size 9550600160

model-00002-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:96308aaa4842c07d081a7467887671a5f7401e6c237aff0f8a9fd419915604d9
-size 9965868688

model-00003-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:aa378c18f7c9460255deb623bd37db08550cb4c99efbaf576459afa13715af22
-size 9630258400

model-00004-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:e8b0208a7116afe3d4a8a5b7acb4a820318580b1f741a4ae7447278e94e90c0a
-size 9965852920

model-00005-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:86f17d548595560057ba5df502f13d37d48df10e91ce7f0cc2376e7f4c1d49a6
-size 9965918912

model-00006-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:77811c11e236d64ad6c4a77191ec53f81ec1dcd39d44a47fcfd624db44246bdb
-size 9965868720

model-00007-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:36d6acb9639301fa3ba7da1dd06f98c487bdbcf6110eff7404c5dbb99e6c8b2d
-size 9932314048

model-00008-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:ebffde08011b21a5467ef7a295c4a60918ea38264b85569fc6b3578a062618a6
-size 9965803424

model-00009-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:d75c9c5bb0638761ea2bc64f73b09d48f56257aa115c2d9e0e24c3e0c8f6f731
-size 9932314752

model-00010-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:e0ebfe7f1b8b70724a3c8de259832476a91b869b802b49d33928adbef10c6316
-size 9965902416

model-00011-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:e8f68cd61a60b6abd420186ca6d9977b5ea22ce3d917d80e71ca8651022fa33c
-size 9831650872

model-00012-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:254d9494ebc0f30714bc97b8a6c045711f7191fe1d00bbd31a09ba6989a04e7f
-size 9932281048

model-00013-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:c0fbded74e005fd22d379f1157af7d03993e2736fca81db641a0bfc6fb7e7202
-size 9965819928

model-00014-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:693d62771cc61d422bbc127e95a30288f47036a3f0df18d708d7e1d747c7efd1
-size 9932331248

model-00015-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:70f02a19b6fe7b34039c5415ab280b7c600ef120c21635f2e49958c580e8e138
-size 9999440584

model-00016-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:9b02b7fd9d8b577bd47ddb1aa844b6c8efd8eb386d1e72f20c7432ccacc5e0a5
-size 9630340192

model-00017-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:f9368a33e18dc28071956d62e98d6dc6514d42226f1e466338a8e08ae4734feb
-size 9630291416

model-00018-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:a1b877b50a66ed98bf95744a3f4fe374eda7f27481d24a7b837cb191f795bb48
-size 9630259136

model-00019-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:6e12858611510be68675c9a3fb1beee6cd00e5196cb77cc5e55bd8f671410570
-size 9798129960

model-00020-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:d9c4ed2d6e9cb62bf15b275b54a1ae94e903b284f2c5256a8d704345d74ca4d5
-size 9999423432

model-00021-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:b68db47db091cc95b70d0379c10fcd386bb39aacfba2cb7d6e77e90d244b034b
-size 9798096256

model-00022-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:137dff53efda2c074cef806ff5d3580587a3c96bfa635a36b827eb51c6ea671f
-size 9932281808

model-00023-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:5b249f0237842dec5beaf19d631af44be9b2610bd2a4dfc612bd3c21f24f3e58
-size 9798096968

model-00024-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:19a86033feabcd41aa284f20acf3f38d9613d93f673495d42d8fee06ee0b7904
-size 9886193328

model-00025-of-00025.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:88e468b32f0476143f0334dbd7b1f95c3fe70f444f64ad5424e3c59715ab3dd5
-size 604013032

model.safetensors.index.json

@@ -1 +0,0 @@
-{"metadata": {"mergekit_version": "0.0.4.2", "total_size": 237209206784}, "weight_map": {"model.layers.2.mlp.gate_proj.weight": "model-00001-of-00025.safetensors", "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00025.safetensors", "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00025.safetensors", "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00025.safetensors", "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00025.safetensors", "model.layers.1.mlp.down_proj.weight": "model-00001-of-00025.safetensors", "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00025.safetensors", "model.layers.1.mlp.up_proj.weight": "model-00001-of-00025.safetensors", "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00025.safetensors", "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00025.safetensors", "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00025.safetensors", "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00025.safetensors", "model.layers.1.input_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.0.mlp.down_proj.weight": "model-00001-of-00025.safetensors", "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00025.safetensors", "model.layers.0.mlp.up_proj.weight": "model-00001-of-00025.safetensors", "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00025.safetensors", "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00025.safetensors", "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00025.safetensors", "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00025.safetensors", "model.layers.0.input_layernorm.weight": "model-00001-of-00025.safetensors", "model.embed_tokens.weight": "model-00001-of-00025.safetensors", "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00025.safetensors", "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00025.safetensors", "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00025.safetensors", "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00025.safetensors", "model.layers.4.mlp.down_proj.weight": "model-00001-of-00025.safetensors", "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00025.safetensors", "model.layers.4.mlp.up_proj.weight": "model-00001-of-00025.safetensors", "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00025.safetensors", "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00025.safetensors", "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00025.safetensors", "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00025.safetensors", "model.layers.4.input_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.3.mlp.down_proj.weight": "model-00001-of-00025.safetensors", "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00025.safetensors", "model.layers.3.mlp.up_proj.weight": "model-00001-of-00025.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00025.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00025.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00025.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00025.safetensors", "model.layers.3.input_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.2.mlp.down_proj.weight": "model-00001-of-00025.safetensors", "model.layers.2.mlp.up_proj.weight": "model-00001-of-00025.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.2.input_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00025.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00025.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00025.safetensors", "model.layers.7.mlp.down_proj.weight": "model-00002-of-00025.safetensors", "model.layers.7.mlp.gate_proj.weight": "model-00002-of-00025.safetensors", "model.layers.7.mlp.up_proj.weight": "model-00002-of-00025.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00025.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00002-of-00025.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00002-of-00025.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00002-of-00025.safetensors", "model.layers.7.input_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.6.mlp.down_proj.weight": "model-00002-of-00025.safetensors", "model.layers.6.mlp.gate_proj.weight": "model-00002-of-00025.safetensors", "model.layers.6.mlp.up_proj.weight": "model-00002-of-00025.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00002-of-00025.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00002-of-00025.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00002-of-00025.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00002-of-00025.safetensors", "model.layers.6.input_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.5.mlp.down_proj.weight": "model-00002-of-00025.safetensors", "model.layers.5.mlp.gate_proj.weight": "model-00002-of-00025.safetensors", "model.layers.5.mlp.up_proj.weight": "model-00002-of-00025.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.5.input_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.17.mlp.down_proj.weight": "model-00002-of-00025.safetensors", "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00025.safetensors", "model.layers.17.mlp.up_proj.weight": "model-00002-of-00025.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00025.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00025.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00025.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00025.safetensors", "model.layers.17.input_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.10.mlp.down_proj.weight": "model-00002-of-00025.safetensors", "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00025.safetensors", "model.layers.10.mlp.up_proj.weight": "model-00002-of-00025.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00025.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00025.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00025.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00025.safetensors", "model.layers.10.input_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.9.mlp.down_proj.weight": "model-00002-of-00025.safetensors", "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00025.safetensors", "model.layers.9.mlp.up_proj.weight": "model-00002-of-00025.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00025.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00025.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00025.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00025.safetensors", "model.layers.9.input_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.8.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.8.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.8.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00003-of-00025.safetensors", "model.layers.8.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.20.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.20.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00003-of-00025.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00003-of-00025.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00003-of-00025.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00003-of-00025.safetensors", "model.layers.19.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.19.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.19.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00003-of-00025.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00003-of-00025.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00003-of-00025.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00003-of-00025.safetensors", "model.layers.19.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.18.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.18.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.18.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00003-of-00025.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00003-of-00025.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00003-of-00025.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00003-of-00025.safetensors", "model.layers.18.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.13.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.13.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00003-of-00025.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00003-of-00025.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00003-of-00025.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00003-of-00025.safetensors", "model.layers.12.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.12.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.12.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00003-of-00025.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00003-of-00025.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00003-of-00025.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00003-of-00025.safetensors", "model.layers.12.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.11.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.11.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.11.mlp.up_proj.weight": "model-00004-of-00025.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.11.input_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.23.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.22.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.22.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.22.mlp.up_proj.weight": "model-00004-of-00025.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.22.input_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.21.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.21.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.21.mlp.up_proj.weight": "model-00004-of-00025.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.21.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.21.input_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.20.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.20.input_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.16.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.15.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.15.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.15.mlp.up_proj.weight": "model-00004-of-00025.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.15.input_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.14.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.14.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.14.mlp.up_proj.weight": "model-00004-of-00025.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.14.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.13.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.13.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.33.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.33.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.33.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.33.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.32.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.32.mlp.gate_proj.weight": "model-00005-of-00025.safetensors", "model.layers.32.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.32.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.32.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.32.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.32.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.32.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.32.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.31.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.31.mlp.gate_proj.weight": "model-00005-of-00025.safetensors", "model.layers.31.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.31.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.25.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.25.mlp.gate_proj.weight": "model-00005-of-00025.safetensors", "model.layers.25.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.25.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.24.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.24.mlp.gate_proj.weight": "model-00005-of-00025.safetensors", "model.layers.24.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.24.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.23.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.23.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.23.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.16.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.16.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.16.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.36.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.36.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.36.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.35.mlp.down_proj.weight": "model-00006-of-00025.safetensors", "model.layers.35.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.35.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.35.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.35.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.35.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.35.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.35.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.35.input_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.34.mlp.down_proj.weight": "model-00006-of-00025.safetensors", "model.layers.34.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.34.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.34.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.34.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.34.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.34.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.34.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.34.input_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.33.mlp.down_proj.weight": "model-00006-of-00025.safetensors", "model.layers.33.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.33.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.33.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.33.input_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.28.mlp.down_proj.weight": "model-00006-of-00025.safetensors", "model.layers.28.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.28.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.28.input_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.27.mlp.down_proj.weight": "model-00006-of-00025.safetensors", "model.layers.27.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.27.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.27.input_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.26.mlp.down_proj.weight": "model-00006-of-00025.safetensors", "model.layers.26.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.26.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.26.input_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.46.mlp.down_proj.weight": "model-00007-of-00025.safetensors", "model.layers.46.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.46.mlp.up_proj.weight": "model-00007-of-00025.safetensors", "model.layers.46.post_attention_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.46.self_attn.o_proj.weight": "model-00007-of-00025.safetensors", "model.layers.46.self_attn.v_proj.weight": "model-00007-of-00025.safetensors", "model.layers.46.self_attn.k_proj.weight": "model-00007-of-00025.safetensors", "model.layers.46.self_attn.q_proj.weight": "model-00007-of-00025.safetensors", "model.layers.46.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.38.mlp.down_proj.weight": "model-00007-of-00025.safetensors", "model.layers.38.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.38.mlp.up_proj.weight": "model-00007-of-00025.safetensors", "model.layers.38.post_attention_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.38.self_attn.o_proj.weight": "model-00007-of-00025.safetensors", "model.layers.38.self_attn.v_proj.weight": "model-00007-of-00025.safetensors", "model.layers.38.self_attn.k_proj.weight": "model-00007-of-00025.safetensors", "model.layers.38.self_attn.q_proj.weight": "model-00007-of-00025.safetensors", "model.layers.38.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.37.mlp.down_proj.weight": "model-00007-of-00025.safetensors", "model.layers.37.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.37.mlp.up_proj.weight": "model-00007-of-00025.safetensors", "model.layers.37.post_attention_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.37.self_attn.o_proj.weight": "model-00007-of-00025.safetensors", "model.layers.37.self_attn.v_proj.weight": "model-00007-of-00025.safetensors", "model.layers.37.self_attn.k_proj.weight": "model-00007-of-00025.safetensors", "model.layers.37.self_attn.q_proj.weight": "model-00007-of-00025.safetensors", "model.layers.37.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.36.mlp.down_proj.weight": "model-00007-of-00025.safetensors", "model.layers.36.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.36.mlp.up_proj.weight": "model-00007-of-00025.safetensors", "model.layers.36.post_attention_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.36.self_attn.o_proj.weight": "model-00007-of-00025.safetensors", "model.layers.36.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.30.mlp.down_proj.weight": "model-00007-of-00025.safetensors", "model.layers.30.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.30.mlp.up_proj.weight": "model-00007-of-00025.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00007-of-00025.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00007-of-00025.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00007-of-00025.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00007-of-00025.safetensors", "model.layers.30.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.29.mlp.down_proj.weight": "model-00007-of-00025.safetensors", "model.layers.29.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.29.mlp.up_proj.weight": "model-00007-of-00025.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00007-of-00025.safetensors", "model.layers.29.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.49.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.49.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.49.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.49.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.49.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.49.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.48.mlp.down_proj.weight": "model-00008-of-00025.safetensors", "model.layers.48.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.48.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.48.post_attention_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.48.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.48.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.48.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.48.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.48.input_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.47.mlp.down_proj.weight": "model-00008-of-00025.safetensors", "model.layers.47.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.47.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.47.post_attention_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.47.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.47.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.47.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.47.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.47.input_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.41.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.41.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.41.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.41.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.41.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.41.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.40.mlp.down_proj.weight": "model-00008-of-00025.safetensors", "model.layers.40.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.40.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.40.post_attention_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.40.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.40.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.40.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.40.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.40.input_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.39.mlp.down_proj.weight": "model-00008-of-00025.safetensors", "model.layers.39.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.39.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.39.post_attention_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.39.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.39.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.39.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.39.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.39.input_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.52.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.52.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.52.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.52.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.52.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.51.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.51.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.51.mlp.up_proj.weight": "model-00009-of-00025.safetensors", "model.layers.51.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.51.self_attn.o_proj.weight": "model-00009-of-00025.safetensors", "model.layers.51.self_attn.v_proj.weight": "model-00009-of-00025.safetensors", "model.layers.51.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.51.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.51.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.50.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.50.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.50.mlp.up_proj.weight": "model-00009-of-00025.safetensors", "model.layers.50.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.50.self_attn.o_proj.weight": "model-00009-of-00025.safetensors", "model.layers.50.self_attn.v_proj.weight": "model-00009-of-00025.safetensors", "model.layers.50.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.50.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.50.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.49.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.49.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.49.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.44.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.44.self_attn.o_proj.weight": "model-00009-of-00025.safetensors", "model.layers.44.self_attn.v_proj.weight": "model-00009-of-00025.safetensors", "model.layers.44.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.44.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.43.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.43.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.43.mlp.up_proj.weight": "model-00009-of-00025.safetensors", "model.layers.43.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.43.self_attn.o_proj.weight": "model-00009-of-00025.safetensors", "model.layers.43.self_attn.v_proj.weight": "model-00009-of-00025.safetensors", "model.layers.43.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.43.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.43.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.42.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.42.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.42.mlp.up_proj.weight": "model-00009-of-00025.safetensors", "model.layers.42.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.42.self_attn.o_proj.weight": "model-00009-of-00025.safetensors", "model.layers.42.self_attn.v_proj.weight": "model-00009-of-00025.safetensors", "model.layers.42.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.42.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.42.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.41.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.41.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.41.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.63.self_attn.o_proj.weight": "model-00009-of-00025.safetensors", "model.layers.63.self_attn.v_proj.weight": "model-00009-of-00025.safetensors", "model.layers.63.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.63.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.62.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.62.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.62.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.62.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.62.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.62.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.62.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.62.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.62.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.55.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.55.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.55.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.55.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.54.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.54.mlp.gate_proj.weight": "model-00010-of-00025.safetensors", "model.layers.54.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.54.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.54.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.54.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.54.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.54.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.54.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.53.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.53.mlp.gate_proj.weight": "model-00010-of-00025.safetensors", "model.layers.53.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.53.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.53.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.53.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.53.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.53.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.53.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.52.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.52.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.52.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.52.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.45.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.45.mlp.gate_proj.weight": "model-00010-of-00025.safetensors", "model.layers.45.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.45.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.45.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.45.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.45.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.45.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.45.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.44.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.44.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.44.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.44.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.66.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.66.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.66.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.65.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.65.mlp.gate_proj.weight": "model-00010-of-00025.safetensors", "model.layers.65.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.65.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.65.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.65.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.65.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.65.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.65.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.64.mlp.down_proj.weight": "model-00011-of-00025.safetensors", "model.layers.64.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.64.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.64.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.64.self_attn.o_proj.weight": "model-00011-of-00025.safetensors", "model.layers.64.self_attn.v_proj.weight": "model-00011-of-00025.safetensors", "model.layers.64.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.64.self_attn.q_proj.weight": "model-00011-of-00025.safetensors", "model.layers.64.input_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.63.mlp.down_proj.weight": "model-00011-of-00025.safetensors", "model.layers.63.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.63.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.63.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.63.input_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.58.self_attn.v_proj.weight": "model-00011-of-00025.safetensors", "model.layers.58.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.58.self_attn.q_proj.weight": "model-00011-of-00025.safetensors", "model.layers.57.mlp.down_proj.weight": "model-00011-of-00025.safetensors", "model.layers.57.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.57.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.57.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.57.self_attn.o_proj.weight": "model-00011-of-00025.safetensors", "model.layers.57.self_attn.v_proj.weight": "model-00011-of-00025.safetensors", "model.layers.57.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.57.self_attn.q_proj.weight": "model-00011-of-00025.safetensors", "model.layers.57.input_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.56.mlp.down_proj.weight": "model-00011-of-00025.safetensors", "model.layers.56.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.56.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.56.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.56.self_attn.o_proj.weight": "model-00011-of-00025.safetensors", "model.layers.56.self_attn.v_proj.weight": "model-00011-of-00025.safetensors", "model.layers.56.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.56.self_attn.q_proj.weight": "model-00011-of-00025.safetensors", "model.layers.56.input_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.55.mlp.down_proj.weight": "model-00011-of-00025.safetensors", "model.layers.55.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.55.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.55.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.55.input_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.68.mlp.down_proj.weight": "model-00011-of-00025.safetensors", "model.layers.68.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.68.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.68.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.68.self_attn.o_proj.weight": "model-00011-of-00025.safetensors", "model.layers.68.self_attn.v_proj.weight": "model-00011-of-00025.safetensors", "model.layers.68.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.68.self_attn.q_proj.weight": "model-00011-of-00025.safetensors", "model.layers.68.input_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.67.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.67.mlp.gate_proj.weight": "model-00012-of-00025.safetensors", "model.layers.67.mlp.up_proj.weight": "model-00012-of-00025.safetensors", "model.layers.67.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.67.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.67.self_attn.v_proj.weight": "model-00012-of-00025.safetensors", "model.layers.67.self_attn.k_proj.weight": "model-00012-of-00025.safetensors", "model.layers.67.self_attn.q_proj.weight": "model-00012-of-00025.safetensors", "model.layers.67.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.66.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.66.mlp.gate_proj.weight": "model-00012-of-00025.safetensors", "model.layers.66.mlp.up_proj.weight": "model-00012-of-00025.safetensors", "model.layers.66.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.66.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.66.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.60.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.60.mlp.gate_proj.weight": "model-00012-of-00025.safetensors", "model.layers.60.mlp.up_proj.weight": "model-00012-of-00025.safetensors", "model.layers.60.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.60.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.60.self_attn.v_proj.weight": "model-00012-of-00025.safetensors", "model.layers.60.self_attn.k_proj.weight": "model-00012-of-00025.safetensors", "model.layers.60.self_attn.q_proj.weight": "model-00012-of-00025.safetensors", "model.layers.60.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.59.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.59.mlp.gate_proj.weight": "model-00012-of-00025.safetensors", "model.layers.59.mlp.up_proj.weight": "model-00012-of-00025.safetensors", "model.layers.59.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.59.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.59.self_attn.v_proj.weight": "model-00012-of-00025.safetensors", "model.layers.59.self_attn.k_proj.weight": "model-00012-of-00025.safetensors", "model.layers.59.self_attn.q_proj.weight": "model-00012-of-00025.safetensors", "model.layers.59.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.58.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.58.mlp.gate_proj.weight": "model-00012-of-00025.safetensors", "model.layers.58.mlp.up_proj.weight": "model-00012-of-00025.safetensors", "model.layers.58.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.58.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.58.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.79.mlp.gate_proj.weight": "model-00012-of-00025.safetensors", "model.layers.79.mlp.up_proj.weight": "model-00012-of-00025.safetensors", "model.layers.79.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.79.self_attn.v_proj.weight": "model-00012-of-00025.safetensors", "model.layers.79.self_attn.k_proj.weight": "model-00012-of-00025.safetensors", "model.layers.79.self_attn.q_proj.weight": "model-00012-of-00025.safetensors", "model.layers.78.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.78.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.78.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.78.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.78.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.78.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.78.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.78.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.78.input_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.71.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.71.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.71.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.71.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.71.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.71.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.70.mlp.down_proj.weight": "model-00013-of-00025.safetensors", "model.layers.70.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.70.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.70.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.70.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.70.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.70.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.70.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.70.input_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.69.mlp.down_proj.weight": "model-00013-of-00025.safetensors", "model.layers.69.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.69.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.69.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.69.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.69.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.69.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.69.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.69.input_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.61.mlp.down_proj.weight": "model-00013-of-00025.safetensors", "model.layers.61.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.61.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.61.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.61.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.61.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.61.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.61.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.61.input_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.82.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.82.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.82.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.82.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.82.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.81.mlp.down_proj.weight": "model-00013-of-00025.safetensors", "model.layers.81.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.81.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.81.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.81.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.81.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.81.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.81.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.81.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.80.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.80.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.80.mlp.up_proj.weight": "model-00014-of-00025.safetensors", "model.layers.80.post_attention_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.80.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.80.self_attn.v_proj.weight": "model-00014-of-00025.safetensors", "model.layers.80.self_attn.k_proj.weight": "model-00014-of-00025.safetensors", "model.layers.80.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.80.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.79.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.79.post_attention_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.79.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.74.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.74.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.74.self_attn.v_proj.weight": "model-00014-of-00025.safetensors", "model.layers.74.self_attn.k_proj.weight": "model-00014-of-00025.safetensors", "model.layers.74.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.73.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.73.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.73.mlp.up_proj.weight": "model-00014-of-00025.safetensors", "model.layers.73.post_attention_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.73.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.73.self_attn.v_proj.weight": "model-00014-of-00025.safetensors", "model.layers.73.self_attn.k_proj.weight": "model-00014-of-00025.safetensors", "model.layers.73.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.73.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.72.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.72.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.72.mlp.up_proj.weight": "model-00014-of-00025.safetensors", "model.layers.72.post_attention_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.72.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.72.self_attn.v_proj.weight": "model-00014-of-00025.safetensors", "model.layers.72.self_attn.k_proj.weight": "model-00014-of-00025.safetensors", "model.layers.72.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.72.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.71.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.71.post_attention_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.71.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.85.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.85.self_attn.v_proj.weight": "model-00014-of-00025.safetensors", "model.layers.85.self_attn.k_proj.weight": "model-00014-of-00025.safetensors", "model.layers.85.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.84.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.84.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.84.mlp.up_proj.weight": "model-00014-of-00025.safetensors", "model.layers.84.post_attention_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.84.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.84.self_attn.v_proj.weight": "model-00014-of-00025.safetensors", "model.layers.84.self_attn.k_proj.weight": "model-00014-of-00025.safetensors", "model.layers.84.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.84.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.83.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.83.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.83.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.83.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.83.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.83.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.83.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.83.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.83.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.82.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.82.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.82.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.82.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.77.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.77.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.77.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.77.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.76.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.76.mlp.gate_proj.weight": "model-00015-of-00025.safetensors", "model.layers.76.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.76.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.76.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.76.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.76.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.76.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.76.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.75.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.75.mlp.gate_proj.weight": "model-00015-of-00025.safetensors", "model.layers.75.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.75.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.75.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.75.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.75.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.75.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.75.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.74.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.74.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.74.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.74.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.95.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.95.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.95.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.94.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.94.mlp.gate_proj.weight": "model-00015-of-00025.safetensors", "model.layers.94.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.94.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.94.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.94.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.94.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.94.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.94.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.88.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.88.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.88.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.87.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.87.mlp.gate_proj.weight": "model-00015-of-00025.safetensors", "model.layers.87.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.87.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.87.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.87.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.87.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.87.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.87.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.86.mlp.down_proj.weight": "model-00016-of-00025.safetensors", "model.layers.86.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.86.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.86.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.86.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.86.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.86.self_attn.k_proj.weight": "model-00016-of-00025.safetensors", "model.layers.86.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.86.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.85.mlp.down_proj.weight": "model-00016-of-00025.safetensors", "model.layers.85.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.85.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.85.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.85.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.77.mlp.down_proj.weight": "model-00016-of-00025.safetensors", "model.layers.77.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.77.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.77.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.77.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.97.mlp.down_proj.weight": "model-00016-of-00025.safetensors", "model.layers.97.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.97.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.97.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.97.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.97.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.97.self_attn.k_proj.weight": "model-00016-of-00025.safetensors", "model.layers.97.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.97.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.96.mlp.down_proj.weight": "model-00016-of-00025.safetensors", "model.layers.96.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.96.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.96.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.96.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.96.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.96.self_attn.k_proj.weight": "model-00016-of-00025.safetensors", "model.layers.96.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.96.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.95.mlp.down_proj.weight": "model-00016-of-00025.safetensors", "model.layers.95.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.95.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.95.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.95.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.95.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.90.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.90.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.90.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.90.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.90.self_attn.o_proj.weight": "model-00017-of-00025.safetensors", "model.layers.90.self_attn.v_proj.weight": "model-00017-of-00025.safetensors", "model.layers.90.self_attn.k_proj.weight": "model-00017-of-00025.safetensors", "model.layers.90.self_attn.q_proj.weight": "model-00017-of-00025.safetensors", "model.layers.90.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.89.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.89.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.89.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.89.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.89.self_attn.o_proj.weight": "model-00017-of-00025.safetensors", "model.layers.89.self_attn.v_proj.weight": "model-00017-of-00025.safetensors", "model.layers.89.self_attn.k_proj.weight": "model-00017-of-00025.safetensors", "model.layers.89.self_attn.q_proj.weight": "model-00017-of-00025.safetensors", "model.layers.89.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.88.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.88.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.88.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.88.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.88.self_attn.o_proj.weight": "model-00017-of-00025.safetensors", "model.layers.88.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.100.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.100.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.100.self_attn.o_proj.weight": "model-00017-of-00025.safetensors", "model.layers.100.self_attn.v_proj.weight": "model-00017-of-00025.safetensors", "model.layers.100.self_attn.k_proj.weight": "model-00017-of-00025.safetensors", "model.layers.100.self_attn.q_proj.weight": "model-00017-of-00025.safetensors", "model.layers.99.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.99.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.99.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.99.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.99.self_attn.o_proj.weight": "model-00017-of-00025.safetensors", "model.layers.99.self_attn.v_proj.weight": "model-00017-of-00025.safetensors", "model.layers.99.self_attn.k_proj.weight": "model-00017-of-00025.safetensors", "model.layers.99.self_attn.q_proj.weight": "model-00017-of-00025.safetensors", "model.layers.99.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.98.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.98.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.98.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.98.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.98.self_attn.o_proj.weight": "model-00017-of-00025.safetensors", "model.layers.98.self_attn.v_proj.weight": "model-00017-of-00025.safetensors", "model.layers.98.self_attn.k_proj.weight": "model-00017-of-00025.safetensors", "model.layers.98.self_attn.q_proj.weight": "model-00017-of-00025.safetensors", "model.layers.98.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.93.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.93.mlp.up_proj.weight": "model-00018-of-00025.safetensors", "model.layers.93.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.93.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.93.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.93.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.92.mlp.down_proj.weight": "model-00018-of-00025.safetensors", "model.layers.92.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.92.mlp.up_proj.weight": "model-00018-of-00025.safetensors", "model.layers.92.post_attention_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.92.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.92.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.92.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.92.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.92.input_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.91.mlp.down_proj.weight": "model-00018-of-00025.safetensors", "model.layers.91.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.91.mlp.up_proj.weight": "model-00018-of-00025.safetensors", "model.layers.91.post_attention_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.91.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.91.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.91.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.91.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.91.input_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.110.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.110.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.110.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.110.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.110.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.109.mlp.down_proj.weight": "model-00018-of-00025.safetensors", "model.layers.109.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.109.mlp.up_proj.weight": "model-00018-of-00025.safetensors", "model.layers.109.post_attention_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.109.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.109.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.109.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.109.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.109.input_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.108.mlp.down_proj.weight": "model-00018-of-00025.safetensors", "model.layers.108.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.108.mlp.up_proj.weight": "model-00018-of-00025.safetensors", "model.layers.108.post_attention_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.108.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.108.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.108.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.108.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.108.input_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.103.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.103.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.103.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.103.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.103.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.102.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.102.mlp.gate_proj.weight": "model-00019-of-00025.safetensors", "model.layers.102.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.102.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.102.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.102.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.102.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.102.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.102.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.101.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.101.mlp.gate_proj.weight": "model-00019-of-00025.safetensors", "model.layers.101.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.101.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.101.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.101.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.101.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.101.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.101.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.100.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.100.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.100.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.93.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.93.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.93.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.113.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.113.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.113.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.113.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.112.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.112.mlp.gate_proj.weight": "model-00019-of-00025.safetensors", "model.layers.112.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.112.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.112.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.112.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.112.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.112.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.112.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.111.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.111.mlp.gate_proj.weight": "model-00019-of-00025.safetensors", "model.layers.111.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.111.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.111.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.111.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.111.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.111.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.111.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.110.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.110.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.110.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.110.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.106.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.106.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.106.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.106.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.105.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.105.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.105.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.105.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.105.self_attn.o_proj.weight": "model-00020-of-00025.safetensors", "model.layers.105.self_attn.v_proj.weight": "model-00020-of-00025.safetensors", "model.layers.105.self_attn.k_proj.weight": "model-00020-of-00025.safetensors", "model.layers.105.self_attn.q_proj.weight": "model-00020-of-00025.safetensors", "model.layers.105.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.104.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.104.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.104.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.104.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.104.self_attn.o_proj.weight": "model-00020-of-00025.safetensors", "model.layers.104.self_attn.v_proj.weight": "model-00020-of-00025.safetensors", "model.layers.104.self_attn.k_proj.weight": "model-00020-of-00025.safetensors", "model.layers.104.self_attn.q_proj.weight": "model-00020-of-00025.safetensors", "model.layers.104.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.103.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.103.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.103.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.103.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.122.self_attn.v_proj.weight": "model-00020-of-00025.safetensors", "model.layers.122.self_attn.k_proj.weight": "model-00020-of-00025.safetensors", "model.layers.122.self_attn.q_proj.weight": "model-00020-of-00025.safetensors", "model.layers.116.self_attn.v_proj.weight": "model-00020-of-00025.safetensors", "model.layers.116.self_attn.k_proj.weight": "model-00020-of-00025.safetensors", "model.layers.116.self_attn.q_proj.weight": "model-00020-of-00025.safetensors", "model.layers.115.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.115.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.115.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.115.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.115.self_attn.o_proj.weight": "model-00020-of-00025.safetensors", "model.layers.115.self_attn.v_proj.weight": "model-00020-of-00025.safetensors", "model.layers.115.self_attn.k_proj.weight": "model-00020-of-00025.safetensors", "model.layers.115.self_attn.q_proj.weight": "model-00020-of-00025.safetensors", "model.layers.115.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.114.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.114.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.114.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.114.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.114.self_attn.o_proj.weight": "model-00020-of-00025.safetensors", "model.layers.114.self_attn.v_proj.weight": "model-00020-of-00025.safetensors", "model.layers.114.self_attn.k_proj.weight": "model-00020-of-00025.safetensors", "model.layers.114.self_attn.q_proj.weight": "model-00020-of-00025.safetensors", "model.layers.114.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.113.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.113.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.113.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.113.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.113.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.107.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.107.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.107.mlp.up_proj.weight": "model-00021-of-00025.safetensors", "model.layers.107.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.107.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.107.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.107.self_attn.k_proj.weight": "model-00021-of-00025.safetensors", "model.layers.107.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.107.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.106.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.106.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.106.mlp.up_proj.weight": "model-00021-of-00025.safetensors", "model.layers.106.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.106.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.124.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.124.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.124.mlp.up_proj.weight": "model-00021-of-00025.safetensors", "model.layers.124.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.124.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.124.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.124.self_attn.k_proj.weight": "model-00021-of-00025.safetensors", "model.layers.124.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.124.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.123.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.123.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.123.mlp.up_proj.weight": "model-00021-of-00025.safetensors", "model.layers.123.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.123.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.123.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.123.self_attn.k_proj.weight": "model-00021-of-00025.safetensors", "model.layers.123.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.123.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.122.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.122.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.122.mlp.up_proj.weight": "model-00021-of-00025.safetensors", "model.layers.122.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.122.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.122.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.118.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.118.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.118.mlp.up_proj.weight": "model-00021-of-00025.safetensors", "model.layers.118.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.118.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.118.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.118.self_attn.k_proj.weight": "model-00021-of-00025.safetensors", "model.layers.118.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.118.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.117.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.117.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.117.mlp.up_proj.weight": "model-00022-of-00025.safetensors", "model.layers.117.post_attention_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.117.self_attn.o_proj.weight": "model-00022-of-00025.safetensors", "model.layers.117.self_attn.v_proj.weight": "model-00022-of-00025.safetensors", "model.layers.117.self_attn.k_proj.weight": "model-00022-of-00025.safetensors", "model.layers.117.self_attn.q_proj.weight": "model-00022-of-00025.safetensors", "model.layers.117.input_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.116.mlp.down_proj.weight": "model-00022-of-00025.safetensors", "model.layers.116.mlp.gate_proj.weight": "model-00022-of-00025.safetensors", "model.layers.116.mlp.up_proj.weight": "model-00022-of-00025.safetensors", "model.layers.116.post_attention_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.116.self_attn.o_proj.weight": "model-00022-of-00025.safetensors", "model.layers.116.input_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.127.mlp.gate_proj.weight": "model-00022-of-00025.safetensors", "model.layers.127.mlp.up_proj.weight": "model-00022-of-00025.safetensors", "model.layers.127.self_attn.o_proj.weight": "model-00022-of-00025.safetensors", "model.layers.127.self_attn.v_proj.weight": "model-00022-of-00025.safetensors", "model.layers.127.self_attn.k_proj.weight": "model-00022-of-00025.safetensors", "model.layers.127.self_attn.q_proj.weight": "model-00022-of-00025.safetensors", "model.layers.126.mlp.down_proj.weight": "model-00022-of-00025.safetensors", "model.layers.126.mlp.gate_proj.weight": "model-00022-of-00025.safetensors", "model.layers.126.mlp.up_proj.weight": "model-00022-of-00025.safetensors", "model.layers.126.post_attention_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.126.self_attn.o_proj.weight": "model-00022-of-00025.safetensors", "model.layers.126.self_attn.v_proj.weight": "model-00022-of-00025.safetensors", "model.layers.126.self_attn.k_proj.weight": "model-00022-of-00025.safetensors", "model.layers.126.self_attn.q_proj.weight": "model-00022-of-00025.safetensors", "model.layers.126.input_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.125.mlp.down_proj.weight": "model-00022-of-00025.safetensors", "model.layers.125.mlp.gate_proj.weight": "model-00022-of-00025.safetensors", "model.layers.125.mlp.up_proj.weight": "model-00022-of-00025.safetensors", "model.layers.125.post_attention_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.125.self_attn.o_proj.weight": "model-00022-of-00025.safetensors", "model.layers.125.self_attn.v_proj.weight": "model-00022-of-00025.safetensors", "model.layers.125.self_attn.k_proj.weight": "model-00022-of-00025.safetensors", "model.layers.125.self_attn.q_proj.weight": "model-00022-of-00025.safetensors", "model.layers.125.input_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.121.mlp.gate_proj.weight": "model-00022-of-00025.safetensors", "model.layers.121.mlp.up_proj.weight": "model-00022-of-00025.safetensors", "model.layers.121.self_attn.o_proj.weight": "model-00022-of-00025.safetensors", "model.layers.121.self_attn.v_proj.weight": "model-00022-of-00025.safetensors", "model.layers.121.self_attn.k_proj.weight": "model-00022-of-00025.safetensors", "model.layers.121.self_attn.q_proj.weight": "model-00022-of-00025.safetensors", "model.layers.120.mlp.down_proj.weight": "model-00022-of-00025.safetensors", "model.layers.120.mlp.gate_proj.weight": "model-00022-of-00025.safetensors", "model.layers.120.mlp.up_proj.weight": "model-00022-of-00025.safetensors", "model.layers.120.post_attention_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.120.self_attn.o_proj.weight": "model-00022-of-00025.safetensors", "model.layers.120.self_attn.v_proj.weight": "model-00022-of-00025.safetensors", "model.layers.120.self_attn.k_proj.weight": "model-00022-of-00025.safetensors", "model.layers.120.self_attn.q_proj.weight": "model-00022-of-00025.safetensors", "model.layers.120.input_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.119.mlp.down_proj.weight": "model-00023-of-00025.safetensors", "model.layers.119.mlp.gate_proj.weight": "model-00023-of-00025.safetensors", "model.layers.119.mlp.up_proj.weight": "model-00023-of-00025.safetensors", "model.layers.119.post_attention_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.119.self_attn.o_proj.weight": "model-00023-of-00025.safetensors", "model.layers.119.self_attn.v_proj.weight": "model-00023-of-00025.safetensors", "model.layers.119.self_attn.k_proj.weight": "model-00023-of-00025.safetensors", "model.layers.119.self_attn.q_proj.weight": "model-00023-of-00025.safetensors", "model.layers.119.input_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.130.mlp.gate_proj.weight": "model-00023-of-00025.safetensors", "model.layers.130.self_attn.o_proj.weight": "model-00023-of-00025.safetensors", "model.layers.130.self_attn.v_proj.weight": "model-00023-of-00025.safetensors", "model.layers.130.self_attn.k_proj.weight": "model-00023-of-00025.safetensors", "model.layers.130.self_attn.q_proj.weight": "model-00023-of-00025.safetensors", "model.layers.129.mlp.down_proj.weight": "model-00023-of-00025.safetensors", "model.layers.129.mlp.gate_proj.weight": "model-00023-of-00025.safetensors", "model.layers.129.mlp.up_proj.weight": "model-00023-of-00025.safetensors", "model.layers.129.post_attention_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.129.self_attn.o_proj.weight": "model-00023-of-00025.safetensors", "model.layers.129.self_attn.v_proj.weight": "model-00023-of-00025.safetensors", "model.layers.129.self_attn.k_proj.weight": "model-00023-of-00025.safetensors", "model.layers.129.self_attn.q_proj.weight": "model-00023-of-00025.safetensors", "model.layers.129.input_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.128.mlp.down_proj.weight": "model-00023-of-00025.safetensors", "model.layers.128.mlp.gate_proj.weight": "model-00023-of-00025.safetensors", "model.layers.128.mlp.up_proj.weight": "model-00023-of-00025.safetensors", "model.layers.128.post_attention_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.128.self_attn.o_proj.weight": "model-00023-of-00025.safetensors", "model.layers.128.self_attn.v_proj.weight": "model-00023-of-00025.safetensors", "model.layers.128.self_attn.k_proj.weight": "model-00023-of-00025.safetensors", "model.layers.128.self_attn.q_proj.weight": "model-00023-of-00025.safetensors", "model.layers.128.input_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.127.mlp.down_proj.weight": "model-00023-of-00025.safetensors", "model.layers.127.post_attention_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.127.input_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.121.mlp.down_proj.weight": "model-00023-of-00025.safetensors", "model.layers.121.post_attention_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.121.input_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.133.self_attn.o_proj.weight": "model-00023-of-00025.safetensors", "model.layers.133.self_attn.v_proj.weight": "model-00023-of-00025.safetensors", "model.layers.133.self_attn.k_proj.weight": "model-00023-of-00025.safetensors", "model.layers.133.self_attn.q_proj.weight": "model-00023-of-00025.safetensors", "model.layers.132.mlp.down_proj.weight": "model-00023-of-00025.safetensors", "model.layers.132.mlp.gate_proj.weight": "model-00023-of-00025.safetensors", "model.layers.132.mlp.up_proj.weight": "model-00023-of-00025.safetensors", "model.layers.132.post_attention_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.132.self_attn.o_proj.weight": "model-00023-of-00025.safetensors", "model.layers.132.self_attn.v_proj.weight": "model-00023-of-00025.safetensors", "model.layers.132.self_attn.k_proj.weight": "model-00023-of-00025.safetensors", "model.layers.132.self_attn.q_proj.weight": "model-00023-of-00025.safetensors", "model.layers.132.input_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.131.mlp.down_proj.weight": "model-00023-of-00025.safetensors", "model.layers.131.mlp.gate_proj.weight": "model-00023-of-00025.safetensors", "model.layers.131.mlp.up_proj.weight": "model-00024-of-00025.safetensors", "model.layers.131.post_attention_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.131.self_attn.o_proj.weight": "model-00024-of-00025.safetensors", "model.layers.131.self_attn.v_proj.weight": "model-00024-of-00025.safetensors", "model.layers.131.self_attn.k_proj.weight": "model-00024-of-00025.safetensors", "model.layers.131.self_attn.q_proj.weight": "model-00024-of-00025.safetensors", "model.layers.131.input_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.130.mlp.down_proj.weight": "model-00024-of-00025.safetensors", "model.layers.130.mlp.up_proj.weight": "model-00024-of-00025.safetensors", "model.layers.130.post_attention_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.130.input_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.136.self_attn.v_proj.weight": "model-00024-of-00025.safetensors", "model.layers.136.self_attn.k_proj.weight": "model-00024-of-00025.safetensors", "model.layers.136.self_attn.q_proj.weight": "model-00024-of-00025.safetensors", "model.layers.135.mlp.down_proj.weight": "model-00024-of-00025.safetensors", "model.layers.135.mlp.gate_proj.weight": "model-00024-of-00025.safetensors", "model.layers.135.mlp.up_proj.weight": "model-00024-of-00025.safetensors", "model.layers.135.post_attention_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.135.self_attn.o_proj.weight": "model-00024-of-00025.safetensors", "model.layers.135.self_attn.v_proj.weight": "model-00024-of-00025.safetensors", "model.layers.135.self_attn.k_proj.weight": "model-00024-of-00025.safetensors", "model.layers.135.self_attn.q_proj.weight": "model-00024-of-00025.safetensors", "model.layers.135.input_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.134.mlp.down_proj.weight": "model-00024-of-00025.safetensors", "model.layers.134.mlp.gate_proj.weight": "model-00024-of-00025.safetensors", "model.layers.134.mlp.up_proj.weight": "model-00024-of-00025.safetensors", "model.layers.134.post_attention_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.134.self_attn.o_proj.weight": "model-00024-of-00025.safetensors", "model.layers.134.self_attn.v_proj.weight": "model-00024-of-00025.safetensors", "model.layers.134.self_attn.k_proj.weight": "model-00024-of-00025.safetensors", "model.layers.134.self_attn.q_proj.weight": "model-00024-of-00025.safetensors", "model.layers.134.input_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.133.mlp.down_proj.weight": "model-00024-of-00025.safetensors", "model.layers.133.mlp.gate_proj.weight": "model-00024-of-00025.safetensors", "model.layers.133.mlp.up_proj.weight": "model-00024-of-00025.safetensors", "model.layers.133.post_attention_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.133.input_layernorm.weight": "model-00024-of-00025.safetensors", "lm_head.weight": "model-00024-of-00025.safetensors", "model.norm.weight": "model-00024-of-00025.safetensors", "model.layers.137.mlp.down_proj.weight": "model-00024-of-00025.safetensors", "model.layers.137.mlp.gate_proj.weight": "model-00024-of-00025.safetensors", "model.layers.137.mlp.up_proj.weight": "model-00024-of-00025.safetensors", "model.layers.137.post_attention_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.137.self_attn.o_proj.weight": "model-00024-of-00025.safetensors", "model.layers.137.self_attn.v_proj.weight": "model-00024-of-00025.safetensors", "model.layers.137.self_attn.k_proj.weight": "model-00024-of-00025.safetensors", "model.layers.137.self_attn.q_proj.weight": "model-00024-of-00025.safetensors", "model.layers.137.input_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.136.mlp.down_proj.weight": "model-00024-of-00025.safetensors", "model.layers.136.mlp.gate_proj.weight": "model-00024-of-00025.safetensors", "model.layers.136.mlp.up_proj.weight": "model-00025-of-00025.safetensors", "model.layers.136.post_attention_layernorm.weight": "model-00025-of-00025.safetensors", "model.layers.136.self_attn.o_proj.weight": "model-00025-of-00025.safetensors", "model.layers.136.input_layernorm.weight": "model-00025-of-00025.safetensors"}}

special_tokens_map.json

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

tokenizer.model

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

tokenizer_config.json

@@ -1,46 +0,0 @@
-{
-  "add_bos_token": true,
-  "add_eos_token": false,
-  "add_prefix_space": true,
-  "added_tokens_decoder": {
-    "0": {
-      "content": "<unk>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "1": {
-      "content": "<s>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "2": {
-      "content": "</s>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    }
-  },
-  "bos_token": "<s>",
-  "chat_template": "{{ bos_token }}{% if messages[0]['from'] == 'system' %}{{ messages[0]['value'] + '\n\n' }}{% set loop_messages = messages[1:] %}{% else %}{{ 'Below are some instructions that describes some tasks. Write responses that appropriately completes each request.\n\n' }}{% set loop_messages = messages %}{% endif %}{% for message in loop_messages %}{% if message['from'] == 'human' %}{{ '### Instruction:\n' + message['value'] + '\n\n' }}{% elif message['from'] == 'gpt' %}{{ '### Response:\n' + message['value'] + eos_token + '\n\n' }}{% else %}{{ raise_exception('Only user and assistant roles are supported!') }}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '### Response:\n' }}{% endif %}",
-  "clean_up_tokenization_spaces": false,
-  "eos_token": "</s>",
-  "legacy": false,
-  "model_max_length": 32764,
-  "pad_token": "<unk>",
-  "padding_side": "right",
-  "sp_model_kwargs": {},
-  "spaces_between_special_tokens": false,
-  "tokenizer_class": "LlamaTokenizer",
-  "trust_remote_code": false,
-  "unk_token": "<unk>",
-  "use_default_system_prompt": false,
-  "use_fast": true
-}