jspr commited on
Commit
e50bcfb
0 Parent(s):

initial commit

Browse files
.gitattributes ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ *.7z filter=lfs diff=lfs merge=lfs -text
2
+ *.arrow filter=lfs diff=lfs merge=lfs -text
3
+ *.bin filter=lfs diff=lfs merge=lfs -text
4
+ *.bz2 filter=lfs diff=lfs merge=lfs -text
5
+ *.ckpt filter=lfs diff=lfs merge=lfs -text
6
+ *.ftz filter=lfs diff=lfs merge=lfs -text
7
+ *.gz filter=lfs diff=lfs merge=lfs -text
8
+ *.h5 filter=lfs diff=lfs merge=lfs -text
9
+ *.joblib filter=lfs diff=lfs merge=lfs -text
10
+ *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
+ *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
+ *.model filter=lfs diff=lfs merge=lfs -text
13
+ *.msgpack filter=lfs diff=lfs merge=lfs -text
14
+ *.npy filter=lfs diff=lfs merge=lfs -text
15
+ *.npz filter=lfs diff=lfs merge=lfs -text
16
+ *.onnx filter=lfs diff=lfs merge=lfs -text
17
+ *.ot filter=lfs diff=lfs merge=lfs -text
18
+ *.parquet filter=lfs diff=lfs merge=lfs -text
19
+ *.pb filter=lfs diff=lfs merge=lfs -text
20
+ *.pickle filter=lfs diff=lfs merge=lfs -text
21
+ *.pkl filter=lfs diff=lfs merge=lfs -text
22
+ *.pt filter=lfs diff=lfs merge=lfs -text
23
+ *.pth filter=lfs diff=lfs merge=lfs -text
24
+ *.rar filter=lfs diff=lfs merge=lfs -text
25
+ *.safetensors filter=lfs diff=lfs merge=lfs -text
26
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
+ *.tar.* filter=lfs diff=lfs merge=lfs -text
28
+ *.tar filter=lfs diff=lfs merge=lfs -text
29
+ *.tflite filter=lfs diff=lfs merge=lfs -text
30
+ *.tgz filter=lfs diff=lfs merge=lfs -text
31
+ *.wasm filter=lfs diff=lfs merge=lfs -text
32
+ *.xz filter=lfs diff=lfs merge=lfs -text
33
+ *.zip filter=lfs diff=lfs merge=lfs -text
34
+ *.zst filter=lfs diff=lfs merge=lfs -text
35
+ *tfevents* filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model:
3
+ - 152334H/miqu-1-70b-sf
4
+ - grimulkan/aurelian-v0.5-70b-rope8-32K-fp16
5
+ library_name: transformers
6
+ tags:
7
+ - mergekit
8
+ - merge
9
+ ---
10
+ # miqurelian-120b
11
+
12
+ This is v2.0 of a 120b frankenmerge created by interleaving layers of [miqu-1-70b-sf](https://huggingface.co/152334H/miqu-1-70b-sf) with [Aurelian](https://huggingface.co/grimulkan/aurelian-v0.5-70b-rope8-32K-fp16) using [mergekit](https://github.com/cg123/mergekit).
13
+
14
+ ## Model Details
15
+
16
+ - Max Context: 32768 tokens
17
+ - Layers: 140
18
+
19
+ ### Prompt template: Mistral
20
+
21
+ ```
22
+ <s>[INST] {prompt} [/INST]
23
+ ```
24
+
25
+ ## Merge Details
26
+
27
+ ### Merge Method
28
+
29
+ This model was merged using the [linear](https://arxiv.org/abs/2203.05482) merge method.
30
+
31
+ ### Models Merged
32
+
33
+ The following models were included in the merge:
34
+
35
+ - [152334H/miqu-1-70b-sf](https://huggingface.co/152334H/miqu-1-70b-sf)
36
+ - [grimulkan/aurelian-v0.5-70b-rope8-32K-fp16](https://huggingface.co/grimulkan/aurelian-v0.5-70b-rope8-32K-fp16)
37
+
38
+ ### Configuration
39
+
40
+ The following YAML configuration was used to produce this model:
41
+
42
+ <details><summary>mergekit_config.yml</summary>
43
+
44
+ ```yaml
45
+ merge_method: linear
46
+ parameters:
47
+ weight: 1.0
48
+ slices:
49
+ - sources:
50
+ - model: 152334H/miqu-1-70b-sf
51
+ layer_range: [0, 1]
52
+ - model: grimulkan/aurelian-v0.5-70b-rope8-32K-fp16
53
+ layer_range: [0, 1]
54
+ parameters:
55
+ weight: 0
56
+ - sources:
57
+ - model: 152334H/miqu-1-70b-sf
58
+ layer_range: [1, 20]
59
+ - sources:
60
+ - model: grimulkan/aurelian-v0.5-70b-rope8-32K-fp16
61
+ layer_range: [10, 30]
62
+ - sources:
63
+ - model: 152334H/miqu-1-70b-sf
64
+ layer_range: [20, 40]
65
+ - sources:
66
+ - model: grimulkan/aurelian-v0.5-70b-rope8-32K-fp16
67
+ layer_range: [30, 50]
68
+ - sources:
69
+ - model: 152334H/miqu-1-70b-sf
70
+ layer_range: [40, 60]
71
+ - sources:
72
+ - model: grimulkan/aurelian-v0.5-70b-rope8-32K-fp16
73
+ layer_range: [50, 70]
74
+ - sources:
75
+ - model: 152334H/miqu-1-70b-sf
76
+ layer_range: [60, 79]
77
+ - sources:
78
+ - model: 152334H/miqu-1-70b-sf
79
+ layer_range: [79, 80]
80
+ - model: grimulkan/aurelian-v0.5-70b-rope8-32K-fp16
81
+ layer_range: [79, 80]
82
+ parameters:
83
+ weight: 0
84
+ dtype: float16
85
+ tokenizer_source: model:152334H/miqu-1-70b-sf
86
+ ```
87
+
88
+ </details>
config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "LlamaForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 1,
8
+ "eos_token_id": 2,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 8192,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 28672,
13
+ "max_position_embeddings": 32768,
14
+ "model_type": "llama",
15
+ "num_attention_heads": 64,
16
+ "num_hidden_layers": 140,
17
+ "num_key_value_heads": 8,
18
+ "pad_token_id": 0,
19
+ "pretraining_tp": 1,
20
+ "rms_norm_eps": 1e-05,
21
+ "rope_scaling": null,
22
+ "rope_theta": 1000000,
23
+ "tie_word_embeddings": false,
24
+ "torch_dtype": "float16",
25
+ "transformers_version": "4.37.2",
26
+ "use_cache": true,
27
+ "vocab_size": 32000
28
+ }
mergekit_config.yml ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ merge_method: linear
2
+ parameters:
3
+ weight: 1.0
4
+ slices:
5
+ - sources:
6
+ - model: 152334H/miqu-1-70b-sf
7
+ layer_range: [0, 1]
8
+ - model: lizpreciatior/lzlv_70b_fp16_hf
9
+ layer_range: [0, 1]
10
+ parameters:
11
+ weight: 0
12
+ - sources:
13
+ - model: 152334H/miqu-1-70b-sf
14
+ layer_range: [1, 20]
15
+ - sources:
16
+ - model: lizpreciatior/lzlv_70b_fp16_hf
17
+ layer_range: [10, 30]
18
+ - sources:
19
+ - model: 152334H/miqu-1-70b-sf
20
+ layer_range: [20, 40]
21
+ - sources:
22
+ - model: lizpreciatior/lzlv_70b_fp16_hf
23
+ layer_range: [30, 50]
24
+ - sources:
25
+ - model: 152334H/miqu-1-70b-sf
26
+ layer_range: [40, 60]
27
+ - sources:
28
+ - model: lizpreciatior/lzlv_70b_fp16_hf
29
+ layer_range: [50, 70]
30
+ - sources:
31
+ - model: 152334H/miqu-1-70b-sf
32
+ layer_range: [60, 79]
33
+ - sources:
34
+ - model: 152334H/miqu-1-70b-sf
35
+ layer_range: [79, 80]
36
+ - model: lizpreciatior/lzlv_70b_fp16_hf
37
+ layer_range: [79, 80]
38
+ parameters:
39
+ weight: 0
40
+ dtype: float16
41
+ tokenizer_source: model:152334H/miqu-1-70b-sf
model-00001-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5267a871ea7d0e9b39a415233527c8b94ce91c5085780ebd411e446c5fc80dc7
3
+ size 9684785016
model-00002-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23d6c24bdf927f8429fc17cd18c2f9711f9c84baa07aa7fb7d2de4a55809c56e
3
+ size 9965885208
model-00003-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3468ea7ad4222b07a25ff6ff979ae68b644f4e731a8323f21d64d6880afdc22d
3
+ size 9663862536
model-00004-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc4115fd7eb8c6a2b2e3bdfe34765ee87e729dd480f3a6ce68fe26708d7b9fc7
3
+ size 9965852216
model-00005-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:524b20e7ed3d7725be6e11a41b473e0be2c8d566af525de020db82966dce2687
3
+ size 9932298248
model-00006-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7eacb8b107e39915821c279ce2d984cc8aaeca86931cbb2028aa507c5ae80a8f
3
+ size 9965852928
model-00007-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cd6d93637682e069bff1f1097f0c81a63213e2588c926d9c30f9cc03fada479b
3
+ size 9630307904
model-00008-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f2e79779ae9b6c74a6433c57c516657e6d4b6dd3e229159f8009617700fe5223
3
+ size 9798129904
model-00009-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:54a9e9f2928636c73304e413322daa9b90008fb7b63eca553a4b6c740a089cbf
3
+ size 9965835712
model-00010-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f79b570ec0d928c0ff2ecf409ed182ed2dec78a50ee5f5242029e08c8b547046
3
+ size 9965869432
model-00011-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ae638234734e4faa97ad5c12d12dcde2f6310775a96d300d9a0c074cd71d430
3
+ size 9965852232
model-00012-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e269f62164b802b38c7ba87df980d30b221c7be9678c6bfbbc3475ac648bdf2
3
+ size 9932331280
model-00013-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa5b944d17441943b1cc88615afa0b01ab4f63cc9f20d9585151dc5c5eb925e1
3
+ size 9965852248
model-00014-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4a73980ca4da20f80f1dcc4114853d962cdead878c4b10a8160e1722cab9b4e
3
+ size 9798080464
model-00015-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:53731c85ccb29a82b42228cb00deef88b2b58b73c4d5c0964097c631786af676
3
+ size 9798096936
model-00016-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4909113d9ba9cdb0154d34aef1f5fc457629eaf2f67ed8fa9f21fbfa291035b4
3
+ size 9798096936
model-00017-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b0b0e1dd2a23122838f66efda12b8445da99146d23347c090254a655e923624
3
+ size 9831650880
model-00018-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0456ec80cb0ce551e0d5faf3e030b5467085dfc1241809cbefcd29944983929c
3
+ size 9932281752
model-00019-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d75887f06ae4462f58a8c07421fb518e53d3073b5d019d20c63866d6a0cb03b4
3
+ size 9965902416
model-00020-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf464369a1c18111863ff99712b3e39d6b164d8aa510419e624ad1a57c416b59
3
+ size 9798063208
model-00021-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1be625c98aa5de6f5e4ca7d84e4cb9630a438818dac5390ea2c53da77903f5d4
3
+ size 9630325120
model-00022-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c1f7256a50b5049d003294e8f6b16ac4133562c591194306051631b41dbf4c01
3
+ size 9831650920
model-00023-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07ad2d931edaafa4b77d4a1846baff44cfb04ffacbf53046951d3a0228eef670
3
+ size 9932281808
model-00024-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a627ca60440bb5d6ce9ffd6e0f56ca5ec5b326f50e391ee629857c6c7eac630
3
+ size 9663879112
model-00025-of-00025.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:874e744333015dc58203f1e3cd3ec0313cb560acbcdadabcc80b49e397e3f3ea
3
+ size 4248946944
model.safetensors.index.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"metadata": {"mergekit_version": "0.0.4.1"}, "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.21.self_attn.o_proj.weight": "model-00001-of-00025.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00001-of-00025.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00001-of-00025.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00001-of-00025.safetensors", "model.layers.20.mlp.down_proj.weight": "model-00001-of-00025.safetensors", "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00025.safetensors", "model.layers.20.mlp.up_proj.weight": "model-00001-of-00025.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00001-of-00025.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00025.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00025.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00025.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00025.safetensors", "model.layers.20.input_layernorm.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-00002-of-00025.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00002-of-00025.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00002-of-00025.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00002-of-00025.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00002-of-00025.safetensors", "model.layers.3.input_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.2.mlp.down_proj.weight": "model-00002-of-00025.safetensors", "model.layers.2.mlp.up_proj.weight": "model-00002-of-00025.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.2.input_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.26.mlp.down_proj.weight": "model-00002-of-00025.safetensors", "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00025.safetensors", "model.layers.26.mlp.up_proj.weight": "model-00002-of-00025.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00025.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00025.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00025.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00025.safetensors", "model.layers.26.input_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.25.mlp.down_proj.weight": "model-00002-of-00025.safetensors", "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00025.safetensors", "model.layers.25.mlp.up_proj.weight": "model-00002-of-00025.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00025.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00025.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00025.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00025.safetensors", "model.layers.25.input_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.24.mlp.down_proj.weight": "model-00002-of-00025.safetensors", "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00025.safetensors", "model.layers.24.mlp.up_proj.weight": "model-00002-of-00025.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00025.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00025.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00025.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00025.safetensors", "model.layers.24.input_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.23.mlp.down_proj.weight": "model-00002-of-00025.safetensors", "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00025.safetensors", "model.layers.23.mlp.up_proj.weight": "model-00002-of-00025.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00025.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00025.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00025.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00025.safetensors", "model.layers.23.input_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.22.mlp.down_proj.weight": "model-00002-of-00025.safetensors", "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00025.safetensors", "model.layers.22.mlp.up_proj.weight": "model-00002-of-00025.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00025.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00025.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00025.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00025.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00025.safetensors", "model.layers.22.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.21.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.21.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.21.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.8.self_attn.v_proj.weight": "model-00003-of-00025.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00003-of-00025.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00003-of-00025.safetensors", "model.layers.7.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.7.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.7.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00003-of-00025.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00003-of-00025.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00003-of-00025.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00003-of-00025.safetensors", "model.layers.7.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.6.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.6.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.6.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00003-of-00025.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00003-of-00025.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00003-of-00025.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00003-of-00025.safetensors", "model.layers.6.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.5.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.5.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.5.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.5.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.32.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.32.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.32.self_attn.o_proj.weight": "model-00003-of-00025.safetensors", "model.layers.32.self_attn.v_proj.weight": "model-00003-of-00025.safetensors", "model.layers.32.self_attn.k_proj.weight": "model-00003-of-00025.safetensors", "model.layers.32.self_attn.q_proj.weight": "model-00003-of-00025.safetensors", "model.layers.31.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.31.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00025.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00025.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00025.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00025.safetensors", "model.layers.31.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.30.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.30.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.30.mlp.up_proj.weight": "model-00004-of-00025.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.30.input_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.29.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.29.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.29.mlp.up_proj.weight": "model-00004-of-00025.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.29.input_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.28.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.28.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.28.mlp.up_proj.weight": "model-00004-of-00025.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.28.input_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.27.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.27.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.27.mlp.up_proj.weight": "model-00004-of-00025.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.27.input_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.10.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.10.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.10.mlp.up_proj.weight": "model-00004-of-00025.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.10.input_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.9.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.9.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.9.mlp.up_proj.weight": "model-00004-of-00025.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.9.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.9.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.8.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.8.mlp.gate_proj.weight": "model-00005-of-00025.safetensors", "model.layers.8.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.8.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.38.mlp.gate_proj.weight": "model-00005-of-00025.safetensors", "model.layers.38.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.38.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.38.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.38.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.37.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.37.mlp.gate_proj.weight": "model-00005-of-00025.safetensors", "model.layers.37.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.37.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.37.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.37.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.37.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.37.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.37.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.36.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.36.mlp.gate_proj.weight": "model-00005-of-00025.safetensors", "model.layers.36.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.36.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.36.self_attn.o_proj.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-00005-of-00025.safetensors", "model.layers.36.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.35.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.35.mlp.gate_proj.weight": "model-00005-of-00025.safetensors", "model.layers.35.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.35.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.35.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.35.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.35.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.35.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.35.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.34.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.34.mlp.gate_proj.weight": "model-00005-of-00025.safetensors", "model.layers.34.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.34.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.34.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.34.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.34.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.34.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.34.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.33.mlp.down_proj.weight": "model-00005-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.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.33.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.33.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.33.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.33.input_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.32.mlp.down_proj.weight": "model-00006-of-00025.safetensors", "model.layers.32.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.32.input_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.13.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.13.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.12.mlp.down_proj.weight": "model-00006-of-00025.safetensors", "model.layers.12.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.12.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.12.input_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.11.mlp.down_proj.weight": "model-00006-of-00025.safetensors", "model.layers.11.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.11.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.11.input_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.64.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.64.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.64.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.64.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.63.mlp.down_proj.weight": "model-00006-of-00025.safetensors", "model.layers.63.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.63.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.63.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.63.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.63.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.63.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.63.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.63.input_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.62.mlp.down_proj.weight": "model-00006-of-00025.safetensors", "model.layers.62.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.62.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.62.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.62.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.62.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.62.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.62.self_attn.q_proj.weight": "model-00007-of-00025.safetensors", "model.layers.62.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.61.mlp.down_proj.weight": "model-00007-of-00025.safetensors", "model.layers.61.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.61.mlp.up_proj.weight": "model-00007-of-00025.safetensors", "model.layers.61.post_attention_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.61.self_attn.o_proj.weight": "model-00007-of-00025.safetensors", "model.layers.61.self_attn.v_proj.weight": "model-00007-of-00025.safetensors", "model.layers.61.self_attn.k_proj.weight": "model-00007-of-00025.safetensors", "model.layers.61.self_attn.q_proj.weight": "model-00007-of-00025.safetensors", "model.layers.61.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.60.mlp.down_proj.weight": "model-00007-of-00025.safetensors", "model.layers.60.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.60.mlp.up_proj.weight": "model-00007-of-00025.safetensors", "model.layers.60.post_attention_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.60.self_attn.o_proj.weight": "model-00007-of-00025.safetensors", "model.layers.60.self_attn.v_proj.weight": "model-00007-of-00025.safetensors", "model.layers.60.self_attn.k_proj.weight": "model-00007-of-00025.safetensors", "model.layers.60.self_attn.q_proj.weight": "model-00007-of-00025.safetensors", "model.layers.60.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.39.mlp.down_proj.weight": "model-00007-of-00025.safetensors", "model.layers.39.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.39.mlp.up_proj.weight": "model-00007-of-00025.safetensors", "model.layers.39.post_attention_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.39.self_attn.o_proj.weight": "model-00007-of-00025.safetensors", "model.layers.39.self_attn.v_proj.weight": "model-00007-of-00025.safetensors", "model.layers.39.self_attn.k_proj.weight": "model-00007-of-00025.safetensors", "model.layers.39.self_attn.q_proj.weight": "model-00007-of-00025.safetensors", "model.layers.39.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.38.mlp.down_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.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.16.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00007-of-00025.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00007-of-00025.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00007-of-00025.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00007-of-00025.safetensors", "model.layers.15.mlp.down_proj.weight": "model-00007-of-00025.safetensors", "model.layers.15.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.15.mlp.up_proj.weight": "model-00007-of-00025.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00007-of-00025.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00007-of-00025.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00007-of-00025.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00007-of-00025.safetensors", "model.layers.15.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.14.mlp.down_proj.weight": "model-00007-of-00025.safetensors", "model.layers.14.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.14.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.14.input_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.13.mlp.down_proj.weight": "model-00008-of-00025.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.13.input_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.69.mlp.down_proj.weight": "model-00008-of-00025.safetensors", "model.layers.69.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.69.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.69.post_attention_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.69.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.69.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.69.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.69.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.69.input_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.68.mlp.down_proj.weight": "model-00008-of-00025.safetensors", "model.layers.68.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.68.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.68.post_attention_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.68.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.68.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.68.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.68.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.68.input_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.67.mlp.down_proj.weight": "model-00008-of-00025.safetensors", "model.layers.67.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.67.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.67.post_attention_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.67.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.67.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.67.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.67.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.67.input_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.66.mlp.down_proj.weight": "model-00008-of-00025.safetensors", "model.layers.66.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.66.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.66.post_attention_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.66.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.66.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.66.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.66.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.66.input_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.65.mlp.down_proj.weight": "model-00008-of-00025.safetensors", "model.layers.65.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.65.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.65.post_attention_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.65.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.65.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.65.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.65.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.65.input_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.64.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.64.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.64.mlp.up_proj.weight": "model-00009-of-00025.safetensors", "model.layers.64.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.64.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00009-of-00025.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00009-of-00025.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.18.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.18.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.18.mlp.up_proj.weight": "model-00009-of-00025.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00009-of-00025.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00009-of-00025.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.18.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.17.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.17.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.17.mlp.up_proj.weight": "model-00009-of-00025.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00009-of-00025.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00009-of-00025.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.17.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.16.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.16.mlp.up_proj.weight": "model-00009-of-00025.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.16.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.75.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.75.mlp.up_proj.weight": "model-00009-of-00025.safetensors", "model.layers.75.self_attn.o_proj.weight": "model-00009-of-00025.safetensors", "model.layers.75.self_attn.v_proj.weight": "model-00009-of-00025.safetensors", "model.layers.75.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.75.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.74.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.74.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.74.mlp.up_proj.weight": "model-00009-of-00025.safetensors", "model.layers.74.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.74.self_attn.o_proj.weight": "model-00009-of-00025.safetensors", "model.layers.74.self_attn.v_proj.weight": "model-00009-of-00025.safetensors", "model.layers.74.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.74.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.74.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.73.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.73.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.73.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.73.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.73.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.73.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.73.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.73.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.73.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.72.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.72.mlp.gate_proj.weight": "model-00010-of-00025.safetensors", "model.layers.72.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.72.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.72.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.72.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.72.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.72.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.72.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.71.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.71.mlp.gate_proj.weight": "model-00010-of-00025.safetensors", "model.layers.71.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.71.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.71.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.71.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.71.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.71.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.71.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.70.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.70.mlp.gate_proj.weight": "model-00010-of-00025.safetensors", "model.layers.70.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.70.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.70.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.70.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.70.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.70.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.70.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.42.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.42.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.42.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.41.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.41.mlp.gate_proj.weight": "model-00010-of-00025.safetensors", "model.layers.41.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.41.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.41.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.41.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.41.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.41.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.41.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.40.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.40.mlp.gate_proj.weight": "model-00010-of-00025.safetensors", "model.layers.40.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.40.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.40.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.40.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.40.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.40.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.40.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.19.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.19.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.19.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.19.input_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.101.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.101.self_attn.o_proj.weight": "model-00011-of-00025.safetensors", "model.layers.101.self_attn.v_proj.weight": "model-00011-of-00025.safetensors", "model.layers.101.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.101.self_attn.q_proj.weight": "model-00011-of-00025.safetensors", "model.layers.100.mlp.down_proj.weight": "model-00011-of-00025.safetensors", "model.layers.100.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.100.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.100.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.100.self_attn.o_proj.weight": "model-00011-of-00025.safetensors", "model.layers.100.self_attn.v_proj.weight": "model-00011-of-00025.safetensors", "model.layers.100.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.100.self_attn.q_proj.weight": "model-00011-of-00025.safetensors", "model.layers.100.input_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.79.mlp.down_proj.weight": "model-00011-of-00025.safetensors", "model.layers.79.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.79.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.79.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.79.self_attn.o_proj.weight": "model-00011-of-00025.safetensors", "model.layers.79.self_attn.v_proj.weight": "model-00011-of-00025.safetensors", "model.layers.79.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.79.self_attn.q_proj.weight": "model-00011-of-00025.safetensors", "model.layers.79.input_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.78.mlp.down_proj.weight": "model-00011-of-00025.safetensors", "model.layers.78.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.78.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.78.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.78.self_attn.o_proj.weight": "model-00011-of-00025.safetensors", "model.layers.78.self_attn.v_proj.weight": "model-00011-of-00025.safetensors", "model.layers.78.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.78.self_attn.q_proj.weight": "model-00011-of-00025.safetensors", "model.layers.78.input_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.77.mlp.down_proj.weight": "model-00011-of-00025.safetensors", "model.layers.77.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.77.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.77.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.77.self_attn.o_proj.weight": "model-00011-of-00025.safetensors", "model.layers.77.self_attn.v_proj.weight": "model-00011-of-00025.safetensors", "model.layers.77.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.77.self_attn.q_proj.weight": "model-00011-of-00025.safetensors", "model.layers.77.input_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.76.mlp.down_proj.weight": "model-00011-of-00025.safetensors", "model.layers.76.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.76.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.76.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.76.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.76.self_attn.v_proj.weight": "model-00012-of-00025.safetensors", "model.layers.76.self_attn.k_proj.weight": "model-00012-of-00025.safetensors", "model.layers.76.self_attn.q_proj.weight": "model-00012-of-00025.safetensors", "model.layers.76.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.75.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.75.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.75.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.44.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.44.mlp.gate_proj.weight": "model-00012-of-00025.safetensors", "model.layers.44.mlp.up_proj.weight": "model-00012-of-00025.safetensors", "model.layers.44.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.44.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.44.self_attn.v_proj.weight": "model-00012-of-00025.safetensors", "model.layers.44.self_attn.k_proj.weight": "model-00012-of-00025.safetensors", "model.layers.44.self_attn.q_proj.weight": "model-00012-of-00025.safetensors", "model.layers.44.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.43.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.43.mlp.gate_proj.weight": "model-00012-of-00025.safetensors", "model.layers.43.mlp.up_proj.weight": "model-00012-of-00025.safetensors", "model.layers.43.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.43.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.43.self_attn.v_proj.weight": "model-00012-of-00025.safetensors", "model.layers.43.self_attn.k_proj.weight": "model-00012-of-00025.safetensors", "model.layers.43.self_attn.q_proj.weight": "model-00012-of-00025.safetensors", "model.layers.43.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.42.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.42.mlp.gate_proj.weight": "model-00012-of-00025.safetensors", "model.layers.42.mlp.up_proj.weight": "model-00012-of-00025.safetensors", "model.layers.42.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.42.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.42.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.107.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.107.self_attn.v_proj.weight": "model-00012-of-00025.safetensors", "model.layers.107.self_attn.k_proj.weight": "model-00012-of-00025.safetensors", "model.layers.107.self_attn.q_proj.weight": "model-00012-of-00025.safetensors", "model.layers.106.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.106.mlp.gate_proj.weight": "model-00012-of-00025.safetensors", "model.layers.106.mlp.up_proj.weight": "model-00012-of-00025.safetensors", "model.layers.106.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.106.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.106.self_attn.v_proj.weight": "model-00012-of-00025.safetensors", "model.layers.106.self_attn.k_proj.weight": "model-00012-of-00025.safetensors", "model.layers.106.self_attn.q_proj.weight": "model-00012-of-00025.safetensors", "model.layers.106.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.105.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.105.mlp.gate_proj.weight": "model-00012-of-00025.safetensors", "model.layers.105.mlp.up_proj.weight": "model-00012-of-00025.safetensors", "model.layers.105.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.105.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.105.self_attn.v_proj.weight": "model-00012-of-00025.safetensors", "model.layers.105.self_attn.k_proj.weight": "model-00012-of-00025.safetensors", "model.layers.105.self_attn.q_proj.weight": "model-00012-of-00025.safetensors", "model.layers.105.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.104.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.104.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.104.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.104.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.104.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.104.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.104.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.104.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.104.input_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.103.mlp.down_proj.weight": "model-00013-of-00025.safetensors", "model.layers.103.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.103.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.103.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.103.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.103.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.103.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.103.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.103.input_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.102.mlp.down_proj.weight": "model-00013-of-00025.safetensors", "model.layers.102.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.102.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.102.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.102.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.102.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.102.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.102.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.102.input_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.101.mlp.down_proj.weight": "model-00013-of-00025.safetensors", "model.layers.101.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.101.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.101.input_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.47.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.47.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.47.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.47.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.47.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.47.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.46.mlp.down_proj.weight": "model-00013-of-00025.safetensors", "model.layers.46.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.46.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.46.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.46.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.46.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.46.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.46.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.46.input_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.45.mlp.down_proj.weight": "model-00013-of-00025.safetensors", "model.layers.45.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.45.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.45.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.45.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.45.self_attn.v_proj.weight": "model-00014-of-00025.safetensors", "model.layers.45.self_attn.k_proj.weight": "model-00014-of-00025.safetensors", "model.layers.45.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.45.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.112.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.112.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.112.mlp.up_proj.weight": "model-00014-of-00025.safetensors", "model.layers.112.post_attention_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.112.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.112.self_attn.v_proj.weight": "model-00014-of-00025.safetensors", "model.layers.112.self_attn.k_proj.weight": "model-00014-of-00025.safetensors", "model.layers.112.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.112.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.111.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.111.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.111.mlp.up_proj.weight": "model-00014-of-00025.safetensors", "model.layers.111.post_attention_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.111.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.111.self_attn.v_proj.weight": "model-00014-of-00025.safetensors", "model.layers.111.self_attn.k_proj.weight": "model-00014-of-00025.safetensors", "model.layers.111.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.111.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.110.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.110.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.110.mlp.up_proj.weight": "model-00014-of-00025.safetensors", "model.layers.110.post_attention_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.110.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.110.self_attn.v_proj.weight": "model-00014-of-00025.safetensors", "model.layers.110.self_attn.k_proj.weight": "model-00014-of-00025.safetensors", "model.layers.110.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.110.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.109.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.109.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.109.mlp.up_proj.weight": "model-00014-of-00025.safetensors", "model.layers.109.post_attention_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.109.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.109.self_attn.v_proj.weight": "model-00014-of-00025.safetensors", "model.layers.109.self_attn.k_proj.weight": "model-00014-of-00025.safetensors", "model.layers.109.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.109.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.108.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.108.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.108.mlp.up_proj.weight": "model-00014-of-00025.safetensors", "model.layers.108.post_attention_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.108.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.108.self_attn.v_proj.weight": "model-00014-of-00025.safetensors", "model.layers.108.self_attn.k_proj.weight": "model-00014-of-00025.safetensors", "model.layers.108.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.108.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.107.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.107.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.107.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.107.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.107.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.50.mlp.gate_proj.weight": "model-00015-of-00025.safetensors", "model.layers.50.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.50.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.50.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.50.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.49.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.49.mlp.gate_proj.weight": "model-00015-of-00025.safetensors", "model.layers.49.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.49.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.49.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.49.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.49.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.49.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.49.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.48.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.48.mlp.gate_proj.weight": "model-00015-of-00025.safetensors", "model.layers.48.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.48.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.48.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.48.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.48.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.48.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.48.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.47.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.47.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.47.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.118.mlp.gate_proj.weight": "model-00015-of-00025.safetensors", "model.layers.118.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.118.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.118.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.118.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.118.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.117.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.117.mlp.gate_proj.weight": "model-00015-of-00025.safetensors", "model.layers.117.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.117.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.117.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.117.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.117.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.117.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.117.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.116.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.116.mlp.gate_proj.weight": "model-00015-of-00025.safetensors", "model.layers.116.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.116.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.116.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.116.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.116.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.116.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.116.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.115.mlp.down_proj.weight": "model-00016-of-00025.safetensors", "model.layers.115.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.115.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.115.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.115.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.115.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.115.self_attn.k_proj.weight": "model-00016-of-00025.safetensors", "model.layers.115.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.115.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.114.mlp.down_proj.weight": "model-00016-of-00025.safetensors", "model.layers.114.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.114.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.114.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.114.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.114.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.114.self_attn.k_proj.weight": "model-00016-of-00025.safetensors", "model.layers.114.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.114.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.113.mlp.down_proj.weight": "model-00016-of-00025.safetensors", "model.layers.113.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.113.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.113.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.113.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.113.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.113.self_attn.k_proj.weight": "model-00016-of-00025.safetensors", "model.layers.113.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.113.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.53.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.53.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.53.self_attn.k_proj.weight": "model-00016-of-00025.safetensors", "model.layers.53.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.52.mlp.down_proj.weight": "model-00016-of-00025.safetensors", "model.layers.52.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.52.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.52.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.52.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.52.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.52.self_attn.k_proj.weight": "model-00016-of-00025.safetensors", "model.layers.52.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.52.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.51.mlp.down_proj.weight": "model-00016-of-00025.safetensors", "model.layers.51.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.51.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.51.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.51.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.51.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.51.self_attn.k_proj.weight": "model-00016-of-00025.safetensors", "model.layers.51.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.51.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.50.mlp.down_proj.weight": "model-00016-of-00025.safetensors", "model.layers.50.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.50.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.50.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.119.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.119.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.119.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.119.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.119.self_attn.o_proj.weight": "model-00017-of-00025.safetensors", "model.layers.119.self_attn.v_proj.weight": "model-00017-of-00025.safetensors", "model.layers.119.self_attn.k_proj.weight": "model-00017-of-00025.safetensors", "model.layers.119.self_attn.q_proj.weight": "model-00017-of-00025.safetensors", "model.layers.119.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.118.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.118.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.118.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.56.self_attn.v_proj.weight": "model-00017-of-00025.safetensors", "model.layers.56.self_attn.k_proj.weight": "model-00017-of-00025.safetensors", "model.layers.56.self_attn.q_proj.weight": "model-00017-of-00025.safetensors", "model.layers.55.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.55.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.55.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.55.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.55.self_attn.o_proj.weight": "model-00017-of-00025.safetensors", "model.layers.55.self_attn.v_proj.weight": "model-00017-of-00025.safetensors", "model.layers.55.self_attn.k_proj.weight": "model-00017-of-00025.safetensors", "model.layers.55.self_attn.q_proj.weight": "model-00017-of-00025.safetensors", "model.layers.55.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.54.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.54.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.54.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.54.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.54.self_attn.o_proj.weight": "model-00017-of-00025.safetensors", "model.layers.54.self_attn.v_proj.weight": "model-00017-of-00025.safetensors", "model.layers.54.self_attn.k_proj.weight": "model-00017-of-00025.safetensors", "model.layers.54.self_attn.q_proj.weight": "model-00017-of-00025.safetensors", "model.layers.54.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.53.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.53.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.53.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.53.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.53.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.58.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.58.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.58.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.58.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.58.self_attn.o_proj.weight": "model-00017-of-00025.safetensors", "model.layers.58.self_attn.v_proj.weight": "model-00017-of-00025.safetensors", "model.layers.58.self_attn.k_proj.weight": "model-00017-of-00025.safetensors", "model.layers.58.self_attn.q_proj.weight": "model-00017-of-00025.safetensors", "model.layers.58.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.57.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.57.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.57.mlp.up_proj.weight": "model-00018-of-00025.safetensors", "model.layers.57.post_attention_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.57.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.57.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.57.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.57.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.57.input_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.56.mlp.down_proj.weight": "model-00018-of-00025.safetensors", "model.layers.56.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.56.mlp.up_proj.weight": "model-00018-of-00025.safetensors", "model.layers.56.post_attention_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.56.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.56.input_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.81.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.81.mlp.up_proj.weight": "model-00018-of-00025.safetensors", "model.layers.81.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.81.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.81.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.81.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.80.mlp.down_proj.weight": "model-00018-of-00025.safetensors", "model.layers.80.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.80.mlp.up_proj.weight": "model-00018-of-00025.safetensors", "model.layers.80.post_attention_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.80.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.80.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.80.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.80.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.80.input_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.59.mlp.down_proj.weight": "model-00018-of-00025.safetensors", "model.layers.59.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.59.mlp.up_proj.weight": "model-00018-of-00025.safetensors", "model.layers.59.post_attention_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.59.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.59.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.59.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.59.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.59.input_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.84.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.84.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.84.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.84.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.84.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.83.mlp.down_proj.weight": "model-00018-of-00025.safetensors", "model.layers.83.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.83.mlp.up_proj.weight": "model-00018-of-00025.safetensors", "model.layers.83.post_attention_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.83.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.83.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.83.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.83.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.83.input_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.82.mlp.down_proj.weight": "model-00018-of-00025.safetensors", "model.layers.82.mlp.gate_proj.weight": "model-00019-of-00025.safetensors", "model.layers.82.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.82.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.82.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.82.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.82.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.82.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.82.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.81.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.81.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.81.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.87.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.87.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.87.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.87.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.86.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.86.mlp.gate_proj.weight": "model-00019-of-00025.safetensors", "model.layers.86.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.86.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.86.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.86.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.86.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.86.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.86.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.85.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.85.mlp.gate_proj.weight": "model-00019-of-00025.safetensors", "model.layers.85.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.85.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.85.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.85.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.85.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.85.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.85.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.84.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.84.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.84.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.84.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.90.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.90.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.90.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.89.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.89.mlp.gate_proj.weight": "model-00019-of-00025.safetensors", "model.layers.89.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.89.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.89.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.89.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.89.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.89.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.89.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.88.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.88.mlp.gate_proj.weight": "model-00019-of-00025.safetensors", "model.layers.88.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.88.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.88.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.88.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.88.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.88.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.88.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.87.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.87.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.87.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.87.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.87.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.92.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.92.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.92.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.92.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.92.self_attn.o_proj.weight": "model-00020-of-00025.safetensors", "model.layers.92.self_attn.v_proj.weight": "model-00020-of-00025.safetensors", "model.layers.92.self_attn.k_proj.weight": "model-00020-of-00025.safetensors", "model.layers.92.self_attn.q_proj.weight": "model-00020-of-00025.safetensors", "model.layers.92.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.91.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.91.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.91.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.91.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.91.self_attn.o_proj.weight": "model-00020-of-00025.safetensors", "model.layers.91.self_attn.v_proj.weight": "model-00020-of-00025.safetensors", "model.layers.91.self_attn.k_proj.weight": "model-00020-of-00025.safetensors", "model.layers.91.self_attn.q_proj.weight": "model-00020-of-00025.safetensors", "model.layers.91.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.90.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.90.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.90.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.90.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.90.self_attn.o_proj.weight": "model-00020-of-00025.safetensors", "model.layers.90.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.95.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.95.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.95.self_attn.o_proj.weight": "model-00020-of-00025.safetensors", "model.layers.95.self_attn.v_proj.weight": "model-00020-of-00025.safetensors", "model.layers.95.self_attn.k_proj.weight": "model-00020-of-00025.safetensors", "model.layers.95.self_attn.q_proj.weight": "model-00020-of-00025.safetensors", "model.layers.94.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.94.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.94.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.94.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.94.self_attn.o_proj.weight": "model-00020-of-00025.safetensors", "model.layers.94.self_attn.v_proj.weight": "model-00020-of-00025.safetensors", "model.layers.94.self_attn.k_proj.weight": "model-00020-of-00025.safetensors", "model.layers.94.self_attn.q_proj.weight": "model-00020-of-00025.safetensors", "model.layers.94.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.93.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.93.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.93.mlp.up_proj.weight": "model-00021-of-00025.safetensors", "model.layers.93.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.93.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.93.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.93.self_attn.k_proj.weight": "model-00021-of-00025.safetensors", "model.layers.93.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.93.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.98.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.98.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.98.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.98.self_attn.k_proj.weight": "model-00021-of-00025.safetensors", "model.layers.98.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.97.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.97.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.97.mlp.up_proj.weight": "model-00021-of-00025.safetensors", "model.layers.97.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.97.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.97.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.97.self_attn.k_proj.weight": "model-00021-of-00025.safetensors", "model.layers.97.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.97.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.96.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.96.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.96.mlp.up_proj.weight": "model-00021-of-00025.safetensors", "model.layers.96.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.96.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.96.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.96.self_attn.k_proj.weight": "model-00021-of-00025.safetensors", "model.layers.96.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.96.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.95.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.95.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.95.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.121.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.121.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.121.self_attn.k_proj.weight": "model-00021-of-00025.safetensors", "model.layers.121.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.120.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.120.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.120.mlp.up_proj.weight": "model-00021-of-00025.safetensors", "model.layers.120.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.120.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.120.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.120.self_attn.k_proj.weight": "model-00021-of-00025.safetensors", "model.layers.120.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.120.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.99.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.99.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.99.mlp.up_proj.weight": "model-00021-of-00025.safetensors", "model.layers.99.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.99.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.99.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.99.self_attn.k_proj.weight": "model-00021-of-00025.safetensors", "model.layers.99.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.99.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.98.mlp.down_proj.weight": "model-00022-of-00025.safetensors", "model.layers.98.mlp.up_proj.weight": "model-00022-of-00025.safetensors", "model.layers.98.post_attention_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.98.input_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.124.self_attn.v_proj.weight": "model-00022-of-00025.safetensors", "model.layers.124.self_attn.k_proj.weight": "model-00022-of-00025.safetensors", "model.layers.124.self_attn.q_proj.weight": "model-00022-of-00025.safetensors", "model.layers.123.mlp.down_proj.weight": "model-00022-of-00025.safetensors", "model.layers.123.mlp.gate_proj.weight": "model-00022-of-00025.safetensors", "model.layers.123.mlp.up_proj.weight": "model-00022-of-00025.safetensors", "model.layers.123.post_attention_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.123.self_attn.o_proj.weight": "model-00022-of-00025.safetensors", "model.layers.123.self_attn.v_proj.weight": "model-00022-of-00025.safetensors", "model.layers.123.self_attn.k_proj.weight": "model-00022-of-00025.safetensors", "model.layers.123.self_attn.q_proj.weight": "model-00022-of-00025.safetensors", "model.layers.123.input_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.122.mlp.down_proj.weight": "model-00022-of-00025.safetensors", "model.layers.122.mlp.gate_proj.weight": "model-00022-of-00025.safetensors", "model.layers.122.mlp.up_proj.weight": "model-00022-of-00025.safetensors", "model.layers.122.post_attention_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.122.self_attn.o_proj.weight": "model-00022-of-00025.safetensors", "model.layers.122.self_attn.v_proj.weight": "model-00022-of-00025.safetensors", "model.layers.122.self_attn.k_proj.weight": "model-00022-of-00025.safetensors", "model.layers.122.self_attn.q_proj.weight": "model-00022-of-00025.safetensors", "model.layers.122.input_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.121.mlp.down_proj.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.post_attention_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.121.input_layernorm.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.124.mlp.down_proj.weight": "model-00022-of-00025.safetensors", "model.layers.124.mlp.gate_proj.weight": "model-00023-of-00025.safetensors", "model.layers.124.mlp.up_proj.weight": "model-00023-of-00025.safetensors", "model.layers.124.post_attention_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.124.self_attn.o_proj.weight": "model-00023-of-00025.safetensors", "model.layers.124.input_layernorm.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.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.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.mlp.gate_proj.weight": "model-00023-of-00025.safetensors", "model.layers.127.mlp.up_proj.weight": "model-00023-of-00025.safetensors", "model.layers.127.post_attention_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.127.self_attn.o_proj.weight": "model-00023-of-00025.safetensors", "model.layers.127.self_attn.v_proj.weight": "model-00023-of-00025.safetensors", "model.layers.127.self_attn.k_proj.weight": "model-00023-of-00025.safetensors", "model.layers.127.self_attn.q_proj.weight": "model-00023-of-00025.safetensors", "model.layers.127.input_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.132.mlp.gate_proj.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.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-00023-of-00025.safetensors", "model.layers.131.post_attention_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.131.self_attn.o_proj.weight": "model-00023-of-00025.safetensors", "model.layers.131.self_attn.v_proj.weight": "model-00023-of-00025.safetensors", "model.layers.131.self_attn.k_proj.weight": "model-00023-of-00025.safetensors", "model.layers.131.self_attn.q_proj.weight": "model-00023-of-00025.safetensors", "model.layers.131.input_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.130.mlp.down_proj.weight": "model-00023-of-00025.safetensors", "model.layers.130.mlp.gate_proj.weight": "model-00023-of-00025.safetensors", "model.layers.130.mlp.up_proj.weight": "model-00023-of-00025.safetensors", "model.layers.130.post_attention_layernorm.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.130.input_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.129.mlp.down_proj.weight": "model-00024-of-00025.safetensors", "model.layers.129.post_attention_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.129.input_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.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.self_attn.o_proj.weight": "model-00024-of-00025.safetensors", "model.layers.133.self_attn.v_proj.weight": "model-00024-of-00025.safetensors", "model.layers.133.self_attn.k_proj.weight": "model-00024-of-00025.safetensors", "model.layers.133.self_attn.q_proj.weight": "model-00024-of-00025.safetensors", "model.layers.133.input_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.132.mlp.down_proj.weight": "model-00024-of-00025.safetensors", "model.layers.132.mlp.up_proj.weight": "model-00024-of-00025.safetensors", "model.layers.132.post_attention_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.132.input_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.138.self_attn.v_proj.weight": "model-00024-of-00025.safetensors", "model.layers.138.self_attn.k_proj.weight": "model-00024-of-00025.safetensors", "model.layers.138.self_attn.q_proj.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-00024-of-00025.safetensors", "model.layers.136.post_attention_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.136.self_attn.o_proj.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.136.input_layernorm.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-00025-of-00025.safetensors", "model.layers.135.post_attention_layernorm.weight": "model-00025-of-00025.safetensors", "model.layers.135.input_layernorm.weight": "model-00025-of-00025.safetensors", "lm_head.weight": "model-00025-of-00025.safetensors", "model.norm.weight": "model-00025-of-00025.safetensors", "model.layers.139.mlp.down_proj.weight": "model-00025-of-00025.safetensors", "model.layers.139.mlp.gate_proj.weight": "model-00025-of-00025.safetensors", "model.layers.139.mlp.up_proj.weight": "model-00025-of-00025.safetensors", "model.layers.139.post_attention_layernorm.weight": "model-00025-of-00025.safetensors", "model.layers.139.self_attn.o_proj.weight": "model-00025-of-00025.safetensors", "model.layers.139.self_attn.v_proj.weight": "model-00025-of-00025.safetensors", "model.layers.139.self_attn.k_proj.weight": "model-00025-of-00025.safetensors", "model.layers.139.self_attn.q_proj.weight": "model-00025-of-00025.safetensors", "model.layers.139.input_layernorm.weight": "model-00025-of-00025.safetensors", "model.layers.138.mlp.down_proj.weight": "model-00025-of-00025.safetensors", "model.layers.138.mlp.gate_proj.weight": "model-00025-of-00025.safetensors", "model.layers.138.mlp.up_proj.weight": "model-00025-of-00025.safetensors", "model.layers.138.post_attention_layernorm.weight": "model-00025-of-00025.safetensors", "model.layers.138.self_attn.o_proj.weight": "model-00025-of-00025.safetensors", "model.layers.138.input_layernorm.weight": "model-00025-of-00025.safetensors"}}
special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<unk>",
18
+ "lstrip": false,
19
+ "normalized": true,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<unk>",
25
+ "lstrip": false,
26
+ "normalized": true,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
tokenizer_config.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": true,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": true,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": true,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ }
29
+ },
30
+ "bos_token": "<s>",
31
+ "chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if message['role'] == 'user' %}{{ '[INST] ' + message['content'] + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ message['content'] + eos_token}}{% else %}{{ raise_exception('Only user and assistant roles are supported!') }}{% endif %}{% endfor %}",
32
+ "clean_up_tokenization_spaces": false,
33
+ "eos_token": "</s>",
34
+ "legacy": false,
35
+ "model_max_length": 1000000000000000019884624838656,
36
+ "pad_token": "<unk>",
37
+ "sp_model_kwargs": {},
38
+ "spaces_between_special_tokens": false,
39
+ "tokenizer_class": "LlamaTokenizer",
40
+ "unk_token": "<unk>",
41
+ "use_default_system_prompt": false
42
+ }