Upload folder using huggingface_hub
Browse files- .ipynb_checkpoints/config-checkpoint.json +28 -0
- README.md +61 -0
- config.json +29 -0
- mergekit_config.yml +30 -0
- model-00001-of-00024.safetensors +3 -0
- model-00002-of-00024.safetensors +3 -0
- model-00003-of-00024.safetensors +3 -0
- model-00004-of-00024.safetensors +3 -0
- model-00005-of-00024.safetensors +3 -0
- model-00006-of-00024.safetensors +3 -0
- model-00007-of-00024.safetensors +3 -0
- model-00008-of-00024.safetensors +3 -0
- model-00009-of-00024.safetensors +3 -0
- model-00010-of-00024.safetensors +3 -0
- model-00011-of-00024.safetensors +3 -0
- model-00012-of-00024.safetensors +3 -0
- model-00013-of-00024.safetensors +3 -0
- model-00014-of-00024.safetensors +3 -0
- model-00015-of-00024.safetensors +3 -0
- model-00016-of-00024.safetensors +3 -0
- model-00017-of-00024.safetensors +3 -0
- model-00018-of-00024.safetensors +3 -0
- model-00019-of-00024.safetensors +3 -0
- model-00020-of-00024.safetensors +3 -0
- model-00021-of-00024.safetensors +3 -0
- model-00022-of-00024.safetensors +3 -0
- model-00023-of-00024.safetensors +3 -0
- model-00024-of-00024.safetensors +3 -0
- model.safetensors.index.json +1 -0
- special_tokens_map.json +23 -0
- tokenizer.json +0 -0
- tokenizer.model +3 -0
- tokenizer_config.json +42 -0
.ipynb_checkpoints/config-checkpoint.json
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "lizpreciatior/lzlv_70b_fp16_hf",
|
3 |
+
"architectures": [
|
4 |
+
"LlamaForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": false,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"bos_token_id": 1,
|
9 |
+
"eos_token_id": 2,
|
10 |
+
"hidden_act": "silu",
|
11 |
+
"hidden_size": 8192,
|
12 |
+
"initializer_range": 0.02,
|
13 |
+
"intermediate_size": 28672,
|
14 |
+
"max_position_embeddings": 4096,
|
15 |
+
"model_type": "llama",
|
16 |
+
"num_attention_heads": 64,
|
17 |
+
"num_hidden_layers": 137,
|
18 |
+
"num_key_value_heads": 8,
|
19 |
+
"pretraining_tp": 1,
|
20 |
+
"rms_norm_eps": 1e-05,
|
21 |
+
"rope_scaling": null,
|
22 |
+
"rope_theta": 10000.0,
|
23 |
+
"tie_word_embeddings": false,
|
24 |
+
"torch_dtype": "bfloat16",
|
25 |
+
"transformers_version": "4.40.0.dev0",
|
26 |
+
"use_cache": true,
|
27 |
+
"vocab_size": 32000
|
28 |
+
}
|
README.md
ADDED
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
base_model:
|
3 |
+
- lizpreciatior/lzlv_70b_fp16_hf
|
4 |
+
- 152334H/miqu-1-70b-sf
|
5 |
+
library_name: transformers
|
6 |
+
tags:
|
7 |
+
- mergekit
|
8 |
+
- merge
|
9 |
+
|
10 |
+
---
|
11 |
+
# Miquliz-120b-137l
|
12 |
+
|
13 |
+
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
14 |
+
|
15 |
+
## Merge Details
|
16 |
+
### Merge Method
|
17 |
+
|
18 |
+
This model was merged using the passthrough merge method.
|
19 |
+
|
20 |
+
### Models Merged
|
21 |
+
|
22 |
+
The following models were included in the merge:
|
23 |
+
* [lizpreciatior/lzlv_70b_fp16_hf](https://huggingface.co/lizpreciatior/lzlv_70b_fp16_hf)
|
24 |
+
* [152334H/miqu-1-70b-sf](https://huggingface.co/152334H/miqu-1-70b-sf)
|
25 |
+
|
26 |
+
### Configuration
|
27 |
+
|
28 |
+
The following YAML configuration was used to produce this model:
|
29 |
+
|
30 |
+
```yaml
|
31 |
+
slices:
|
32 |
+
- sources:
|
33 |
+
- model: 152334H/miqu-1-70b-sf
|
34 |
+
layer_range: [0, 16]
|
35 |
+
- sources:
|
36 |
+
- model: lizpreciatior/lzlv_70b_fp16_hf
|
37 |
+
layer_range: [8, 24]
|
38 |
+
- sources:
|
39 |
+
- model: 152334H/miqu-1-70b-sf
|
40 |
+
layer_range: [17, 32]
|
41 |
+
- sources:
|
42 |
+
- model: lizpreciatior/lzlv_70b_fp16_hf
|
43 |
+
layer_range: [25, 40]
|
44 |
+
- sources:
|
45 |
+
- model: 152334H/miqu-1-70b-sf
|
46 |
+
layer_range: [33, 48]
|
47 |
+
- sources:
|
48 |
+
- model: lizpreciatior/lzlv_70b_fp16_hf
|
49 |
+
layer_range: [41, 56]
|
50 |
+
- sources:
|
51 |
+
- model: 152334H/miqu-1-70b-sf
|
52 |
+
layer_range: [49, 64]
|
53 |
+
- sources:
|
54 |
+
- model: lizpreciatior/lzlv_70b_fp16_hf
|
55 |
+
layer_range: [57, 72]
|
56 |
+
- sources:
|
57 |
+
- model: 152334H/miqu-1-70b-sf
|
58 |
+
layer_range: [65, 80]
|
59 |
+
merge_method: passthrough
|
60 |
+
dtype: bfloat16
|
61 |
+
```
|
config.json
ADDED
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "lizpreciatior/lzlv_70b_fp16_hf",
|
3 |
+
"architectures": [
|
4 |
+
"LlamaForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": false,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"bos_token_id": 1,
|
9 |
+
"eos_token_id": 2,
|
10 |
+
"hidden_act": "silu",
|
11 |
+
"hidden_size": 8192,
|
12 |
+
"initializer_range": 0.02,
|
13 |
+
"intermediate_size": 28672,
|
14 |
+
"max_position_embeddings": 32764,
|
15 |
+
"model_type": "llama",
|
16 |
+
"num_attention_heads": 64,
|
17 |
+
"num_hidden_layers": 137,
|
18 |
+
"num_key_value_heads": 8,
|
19 |
+
"pad_token_id": 0,
|
20 |
+
"pretraining_tp": 1,
|
21 |
+
"rms_norm_eps": 1e-05,
|
22 |
+
"rope_scaling": null,
|
23 |
+
"rope_theta": 1000000,
|
24 |
+
"tie_word_embeddings": false,
|
25 |
+
"torch_dtype": "bfloat16",
|
26 |
+
"transformers_version": "4.40.0.dev0",
|
27 |
+
"use_cache": true,
|
28 |
+
"vocab_size": 32000
|
29 |
+
}
|
mergekit_config.yml
ADDED
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
slices:
|
2 |
+
- sources:
|
3 |
+
- model: 152334H/miqu-1-70b-sf
|
4 |
+
layer_range: [0, 16]
|
5 |
+
- sources:
|
6 |
+
- model: lizpreciatior/lzlv_70b_fp16_hf
|
7 |
+
layer_range: [8, 24]
|
8 |
+
- sources:
|
9 |
+
- model: 152334H/miqu-1-70b-sf
|
10 |
+
layer_range: [17, 32]
|
11 |
+
- sources:
|
12 |
+
- model: lizpreciatior/lzlv_70b_fp16_hf
|
13 |
+
layer_range: [25, 40]
|
14 |
+
- sources:
|
15 |
+
- model: 152334H/miqu-1-70b-sf
|
16 |
+
layer_range: [33, 48]
|
17 |
+
- sources:
|
18 |
+
- model: lizpreciatior/lzlv_70b_fp16_hf
|
19 |
+
layer_range: [41, 56]
|
20 |
+
- sources:
|
21 |
+
- model: 152334H/miqu-1-70b-sf
|
22 |
+
layer_range: [49, 64]
|
23 |
+
- sources:
|
24 |
+
- model: lizpreciatior/lzlv_70b_fp16_hf
|
25 |
+
layer_range: [57, 72]
|
26 |
+
- sources:
|
27 |
+
- model: 152334H/miqu-1-70b-sf
|
28 |
+
layer_range: [65, 80]
|
29 |
+
merge_method: passthrough
|
30 |
+
dtype: bfloat16
|
model-00001-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b6fd57bd981a857e510124c14a2cf3d991fecac8b3265804011b1b837a2ac6c5
|
3 |
+
size 9986791240
|
model-00002-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ae98f278a86bb644449f3d908e2f2389355b1f4471f0bb8c99bdcfb0f1d1c61a
|
3 |
+
size 9965885936
|
model-00003-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:89ab00727f43c82f49993f536432ce8ef59782de7b401b4a9d900b11ce5d82ce
|
3 |
+
size 9831650896
|
model-00004-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b0586a7d8f1b9edb2eb42416eb252361e3b7b01bbcd87ba892617d35e85673fe
|
3 |
+
size 9798063960
|
model-00005-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:79e42ff79e7e9083e567dabbe4cbb50071691f47fe093a6e0f59ac9de60d9fd6
|
3 |
+
size 9630291432
|
model-00006-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:86623c3c03fc7f35a043f516ed8976e971a74d997c2ae6bcbe38cb14ee950d1f
|
3 |
+
size 9965852984
|
model-00007-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:23d9894bd561b92057bd0a3422adc2f1fb21274238341f8e506c06250952ad57
|
3 |
+
size 9630340944
|
model-00008-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7b06936c00eabbf3583fbb9af714d93f91596497a322a84cb8d52a347b000552
|
3 |
+
size 9965868768
|
model-00009-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8c0770c8f5de73782d04f0d485ad507f32f6758da26e899f38be73e35dab78bd
|
3 |
+
size 9798063960
|
model-00010-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bbb7c7dabcb8eb17147129ba2552c4af7fb4ecbfc4607153f531b39b8e7ced1a
|
3 |
+
size 9965869480
|
model-00011-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fdbe6ce431c9d91bca833be2d9c7b6722349e634e432ee6b6aa3b065de5470e2
|
3 |
+
size 9965868776
|
model-00012-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9f62696112ef4c38d94385c263e41c9c8b8f20d14957f5587584613bd80e3c18
|
3 |
+
size 9932314816
|
model-00013-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ff9212abc81c5088dbd3b1f4491e34663b94a3ae828b6fc87c08cd552af86cd2
|
3 |
+
size 9965852288
|
model-00014-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:dbc954b9ae7f3ba462fd554f71f5d1579b200c20a59714054c211f02a709d2d5
|
3 |
+
size 9798080496
|
model-00015-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1d3f57d52e0ded42377c6e3714d9ad5818e99b8f1352f9c6f14571ea45680e07
|
3 |
+
size 9798097008
|
model-00016-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6e5e3a78e31155037bbde9bfeac4ffe99ca5499eaba3291af2175032ca43c528
|
3 |
+
size 9798097000
|
model-00017-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1abebe8c1152d3c139e429152ca6098a89d28a1de5f5a029bfdce28e4aae2140
|
3 |
+
size 9831650920
|
model-00018-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7e851163fd187f7d16dfe7d7a595daf30fdf7bb2e9f273f30e78b4376a087ee8
|
3 |
+
size 9932314808
|
model-00019-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:89f26d80a869d491092995395074a2f0251cc95ea1485062cb36d235ec0af75e
|
3 |
+
size 9663879112
|
model-00020-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2457174ae66b5b798d6e299d8638cef216f824a4c0e5d8c74eb529be6da5fe8e
|
3 |
+
size 9630258464
|
model-00021-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:45d1c56f6f2a68ea590cae1013b23b14b5f51fbad2a703819e94a0dbf9328c18
|
3 |
+
size 9798130000
|
model-00022-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c507dea1649cfaad0df72023fa0286ba65b491e80d5bc44012f47727cfe1af25
|
3 |
+
size 9965835816
|
model-00023-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7d6b97333661f415131338b19eb126cdea0650729d8780e716388132ede06fe4
|
3 |
+
size 9630325208
|
model-00024-of-00024.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e49676d7609421300d7c020bf221b74a343d8c39ad504193f60423147b78c288
|
3 |
+
size 9248658696
|
model.safetensors.index.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"metadata": {"mergekit_version": "0.0.4.2", "total_size": 235497897984}, "weight_map": {"model.layers.2.mlp.gate_proj.weight": "model-00001-of-00024.safetensors", "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00024.safetensors", "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00024.safetensors", "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00024.safetensors", "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00024.safetensors", "model.layers.1.mlp.down_proj.weight": "model-00001-of-00024.safetensors", "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00024.safetensors", "model.layers.1.mlp.up_proj.weight": "model-00001-of-00024.safetensors", "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00024.safetensors", "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00024.safetensors", "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00024.safetensors", "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00024.safetensors", "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00024.safetensors", "model.layers.1.input_layernorm.weight": "model-00001-of-00024.safetensors", "model.layers.0.mlp.down_proj.weight": "model-00001-of-00024.safetensors", "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00024.safetensors", "model.layers.0.mlp.up_proj.weight": "model-00001-of-00024.safetensors", "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00024.safetensors", "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00024.safetensors", "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00024.safetensors", "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00024.safetensors", "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00024.safetensors", "model.layers.0.input_layernorm.weight": "model-00001-of-00024.safetensors", "model.embed_tokens.weight": "model-00001-of-00024.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00024.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00024.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00024.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00024.safetensors", "model.layers.18.mlp.down_proj.weight": "model-00001-of-00024.safetensors", "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00024.safetensors", "model.layers.18.mlp.up_proj.weight": "model-00001-of-00024.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00024.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00024.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00024.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00024.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00024.safetensors", "model.layers.18.input_layernorm.weight": "model-00001-of-00024.safetensors", "model.layers.17.mlp.down_proj.weight": "model-00001-of-00024.safetensors", "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00024.safetensors", "model.layers.17.mlp.up_proj.weight": "model-00001-of-00024.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00024.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00024.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00024.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00024.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00024.safetensors", "model.layers.17.input_layernorm.weight": "model-00001-of-00024.safetensors", "model.layers.16.mlp.down_proj.weight": "model-00001-of-00024.safetensors", "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00024.safetensors", "model.layers.16.mlp.up_proj.weight": "model-00001-of-00024.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00024.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00024.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00024.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00024.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00024.safetensors", "model.layers.16.input_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.5.self_attn.o_proj.weight": "model-00002-of-00024.safetensors", "model.layers.5.self_attn.v_proj.weight": "model-00002-of-00024.safetensors", "model.layers.5.self_attn.k_proj.weight": "model-00002-of-00024.safetensors", "model.layers.5.self_attn.q_proj.weight": "model-00002-of-00024.safetensors", "model.layers.4.mlp.down_proj.weight": "model-00002-of-00024.safetensors", "model.layers.4.mlp.gate_proj.weight": "model-00002-of-00024.safetensors", "model.layers.4.mlp.up_proj.weight": "model-00002-of-00024.safetensors", "model.layers.4.post_attention_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.4.self_attn.o_proj.weight": "model-00002-of-00024.safetensors", "model.layers.4.self_attn.v_proj.weight": "model-00002-of-00024.safetensors", "model.layers.4.self_attn.k_proj.weight": "model-00002-of-00024.safetensors", "model.layers.4.self_attn.q_proj.weight": "model-00002-of-00024.safetensors", "model.layers.4.input_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.3.mlp.down_proj.weight": "model-00002-of-00024.safetensors", "model.layers.3.mlp.gate_proj.weight": "model-00002-of-00024.safetensors", "model.layers.3.mlp.up_proj.weight": "model-00002-of-00024.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00002-of-00024.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00002-of-00024.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00002-of-00024.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00002-of-00024.safetensors", "model.layers.3.input_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.2.mlp.down_proj.weight": "model-00002-of-00024.safetensors", "model.layers.2.mlp.up_proj.weight": "model-00002-of-00024.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.2.input_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.24.mlp.down_proj.weight": "model-00002-of-00024.safetensors", "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00024.safetensors", "model.layers.24.mlp.up_proj.weight": "model-00002-of-00024.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00024.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00024.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00024.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00024.safetensors", "model.layers.24.input_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.23.mlp.down_proj.weight": "model-00002-of-00024.safetensors", "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00024.safetensors", "model.layers.23.mlp.up_proj.weight": "model-00002-of-00024.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00024.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00024.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00024.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00024.safetensors", "model.layers.23.input_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.22.mlp.down_proj.weight": "model-00002-of-00024.safetensors", "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00024.safetensors", "model.layers.22.mlp.up_proj.weight": "model-00002-of-00024.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00024.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00024.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00024.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00024.safetensors", "model.layers.22.input_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.21.mlp.down_proj.weight": "model-00003-of-00024.safetensors", "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00024.safetensors", "model.layers.21.mlp.up_proj.weight": "model-00003-of-00024.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00024.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00024.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00024.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00024.safetensors", "model.layers.21.input_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.20.mlp.down_proj.weight": "model-00003-of-00024.safetensors", "model.layers.20.mlp.gate_proj.weight": "model-00003-of-00024.safetensors", "model.layers.20.mlp.up_proj.weight": "model-00003-of-00024.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00003-of-00024.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00003-of-00024.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00003-of-00024.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00003-of-00024.safetensors", "model.layers.20.input_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.19.mlp.down_proj.weight": "model-00003-of-00024.safetensors", "model.layers.19.mlp.gate_proj.weight": "model-00003-of-00024.safetensors", "model.layers.19.mlp.up_proj.weight": "model-00003-of-00024.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.19.input_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.8.self_attn.v_proj.weight": "model-00003-of-00024.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00003-of-00024.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00003-of-00024.safetensors", "model.layers.7.mlp.down_proj.weight": "model-00003-of-00024.safetensors", "model.layers.7.mlp.gate_proj.weight": "model-00003-of-00024.safetensors", "model.layers.7.mlp.up_proj.weight": "model-00003-of-00024.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00003-of-00024.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00003-of-00024.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00003-of-00024.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00003-of-00024.safetensors", "model.layers.7.input_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.6.mlp.down_proj.weight": "model-00003-of-00024.safetensors", "model.layers.6.mlp.gate_proj.weight": "model-00003-of-00024.safetensors", "model.layers.6.mlp.up_proj.weight": "model-00003-of-00024.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00003-of-00024.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00003-of-00024.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00003-of-00024.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00003-of-00024.safetensors", "model.layers.6.input_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.5.mlp.down_proj.weight": "model-00003-of-00024.safetensors", "model.layers.5.mlp.gate_proj.weight": "model-00003-of-00024.safetensors", "model.layers.5.mlp.up_proj.weight": "model-00003-of-00024.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.5.input_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.30.mlp.gate_proj.weight": "model-00004-of-00024.safetensors", "model.layers.30.mlp.up_proj.weight": "model-00004-of-00024.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00004-of-00024.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00004-of-00024.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00004-of-00024.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00004-of-00024.safetensors", "model.layers.29.mlp.down_proj.weight": "model-00004-of-00024.safetensors", "model.layers.29.mlp.gate_proj.weight": "model-00004-of-00024.safetensors", "model.layers.29.mlp.up_proj.weight": "model-00004-of-00024.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00004-of-00024.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00004-of-00024.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00004-of-00024.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00004-of-00024.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00004-of-00024.safetensors", "model.layers.29.input_layernorm.weight": "model-00004-of-00024.safetensors", "model.layers.28.mlp.down_proj.weight": "model-00004-of-00024.safetensors", "model.layers.28.mlp.gate_proj.weight": "model-00004-of-00024.safetensors", "model.layers.28.mlp.up_proj.weight": "model-00004-of-00024.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00004-of-00024.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00004-of-00024.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00004-of-00024.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00004-of-00024.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00004-of-00024.safetensors", "model.layers.28.input_layernorm.weight": "model-00004-of-00024.safetensors", "model.layers.27.mlp.down_proj.weight": "model-00004-of-00024.safetensors", "model.layers.27.mlp.gate_proj.weight": "model-00004-of-00024.safetensors", "model.layers.27.mlp.up_proj.weight": "model-00004-of-00024.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00004-of-00024.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00004-of-00024.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00004-of-00024.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00004-of-00024.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00004-of-00024.safetensors", "model.layers.27.input_layernorm.weight": "model-00004-of-00024.safetensors", "model.layers.26.mlp.down_proj.weight": "model-00004-of-00024.safetensors", "model.layers.26.mlp.gate_proj.weight": "model-00004-of-00024.safetensors", "model.layers.26.mlp.up_proj.weight": "model-00004-of-00024.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00004-of-00024.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00004-of-00024.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00004-of-00024.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00004-of-00024.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00004-of-00024.safetensors", "model.layers.26.input_layernorm.weight": "model-00004-of-00024.safetensors", "model.layers.25.mlp.down_proj.weight": "model-00004-of-00024.safetensors", "model.layers.25.mlp.gate_proj.weight": "model-00004-of-00024.safetensors", "model.layers.25.mlp.up_proj.weight": "model-00004-of-00024.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00004-of-00024.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00004-of-00024.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00004-of-00024.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00004-of-00024.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00004-of-00024.safetensors", "model.layers.25.input_layernorm.weight": "model-00004-of-00024.safetensors", "model.layers.10.mlp.down_proj.weight": "model-00005-of-00024.safetensors", "model.layers.10.mlp.gate_proj.weight": "model-00005-of-00024.safetensors", "model.layers.10.mlp.up_proj.weight": "model-00005-of-00024.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00005-of-00024.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00005-of-00024.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00005-of-00024.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00005-of-00024.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00005-of-00024.safetensors", "model.layers.10.input_layernorm.weight": "model-00005-of-00024.safetensors", "model.layers.9.mlp.down_proj.weight": "model-00005-of-00024.safetensors", "model.layers.9.mlp.gate_proj.weight": "model-00005-of-00024.safetensors", "model.layers.9.mlp.up_proj.weight": "model-00005-of-00024.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00005-of-00024.safetensors", "model.layers.9.self_attn.o_proj.weight": "model-00005-of-00024.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00005-of-00024.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00005-of-00024.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00005-of-00024.safetensors", "model.layers.9.input_layernorm.weight": "model-00005-of-00024.safetensors", "model.layers.8.mlp.down_proj.weight": "model-00005-of-00024.safetensors", "model.layers.8.mlp.gate_proj.weight": "model-00005-of-00024.safetensors", "model.layers.8.mlp.up_proj.weight": "model-00005-of-00024.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00005-of-00024.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00005-of-00024.safetensors", "model.layers.8.input_layernorm.weight": "model-00005-of-00024.safetensors", "model.layers.50.mlp.gate_proj.weight": "model-00005-of-00024.safetensors", "model.layers.50.self_attn.o_proj.weight": "model-00005-of-00024.safetensors", "model.layers.50.self_attn.v_proj.weight": "model-00005-of-00024.safetensors", "model.layers.50.self_attn.k_proj.weight": "model-00005-of-00024.safetensors", "model.layers.50.self_attn.q_proj.weight": "model-00005-of-00024.safetensors", "model.layers.49.mlp.down_proj.weight": "model-00005-of-00024.safetensors", "model.layers.49.mlp.gate_proj.weight": "model-00005-of-00024.safetensors", "model.layers.49.mlp.up_proj.weight": "model-00005-of-00024.safetensors", "model.layers.49.post_attention_layernorm.weight": "model-00005-of-00024.safetensors", "model.layers.49.self_attn.o_proj.weight": "model-00005-of-00024.safetensors", "model.layers.49.self_attn.v_proj.weight": "model-00005-of-00024.safetensors", "model.layers.49.self_attn.k_proj.weight": "model-00005-of-00024.safetensors", "model.layers.49.self_attn.q_proj.weight": "model-00005-of-00024.safetensors", "model.layers.49.input_layernorm.weight": "model-00005-of-00024.safetensors", "model.layers.48.mlp.down_proj.weight": "model-00005-of-00024.safetensors", "model.layers.48.mlp.gate_proj.weight": "model-00005-of-00024.safetensors", "model.layers.48.mlp.up_proj.weight": "model-00005-of-00024.safetensors", "model.layers.48.post_attention_layernorm.weight": "model-00005-of-00024.safetensors", "model.layers.48.self_attn.o_proj.weight": "model-00005-of-00024.safetensors", "model.layers.48.self_attn.v_proj.weight": "model-00005-of-00024.safetensors", "model.layers.48.self_attn.k_proj.weight": "model-00005-of-00024.safetensors", "model.layers.48.self_attn.q_proj.weight": "model-00005-of-00024.safetensors", "model.layers.48.input_layernorm.weight": "model-00005-of-00024.safetensors", "model.layers.47.mlp.down_proj.weight": "model-00005-of-00024.safetensors", "model.layers.47.mlp.gate_proj.weight": "model-00006-of-00024.safetensors", "model.layers.47.mlp.up_proj.weight": "model-00006-of-00024.safetensors", "model.layers.47.post_attention_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.47.self_attn.o_proj.weight": "model-00006-of-00024.safetensors", "model.layers.47.self_attn.v_proj.weight": "model-00006-of-00024.safetensors", "model.layers.47.self_attn.k_proj.weight": "model-00006-of-00024.safetensors", "model.layers.47.self_attn.q_proj.weight": "model-00006-of-00024.safetensors", "model.layers.47.input_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.31.mlp.down_proj.weight": "model-00006-of-00024.safetensors", "model.layers.31.mlp.gate_proj.weight": "model-00006-of-00024.safetensors", "model.layers.31.mlp.up_proj.weight": "model-00006-of-00024.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00006-of-00024.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00006-of-00024.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00006-of-00024.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00006-of-00024.safetensors", "model.layers.31.input_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.30.mlp.down_proj.weight": "model-00006-of-00024.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.30.input_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.13.mlp.gate_proj.weight": "model-00006-of-00024.safetensors", "model.layers.13.mlp.up_proj.weight": "model-00006-of-00024.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00006-of-00024.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00006-of-00024.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00006-of-00024.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00006-of-00024.safetensors", "model.layers.12.mlp.down_proj.weight": "model-00006-of-00024.safetensors", "model.layers.12.mlp.gate_proj.weight": "model-00006-of-00024.safetensors", "model.layers.12.mlp.up_proj.weight": "model-00006-of-00024.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00006-of-00024.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00006-of-00024.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00006-of-00024.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00006-of-00024.safetensors", "model.layers.12.input_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.11.mlp.down_proj.weight": "model-00006-of-00024.safetensors", "model.layers.11.mlp.gate_proj.weight": "model-00006-of-00024.safetensors", "model.layers.11.mlp.up_proj.weight": "model-00006-of-00024.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00006-of-00024.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00006-of-00024.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00006-of-00024.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00006-of-00024.safetensors", "model.layers.11.input_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.56.self_attn.o_proj.weight": "model-00006-of-00024.safetensors", "model.layers.56.self_attn.v_proj.weight": "model-00006-of-00024.safetensors", "model.layers.56.self_attn.k_proj.weight": "model-00006-of-00024.safetensors", "model.layers.56.self_attn.q_proj.weight": "model-00006-of-00024.safetensors", "model.layers.55.mlp.down_proj.weight": "model-00006-of-00024.safetensors", "model.layers.55.mlp.gate_proj.weight": "model-00006-of-00024.safetensors", "model.layers.55.mlp.up_proj.weight": "model-00006-of-00024.safetensors", "model.layers.55.post_attention_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.55.self_attn.o_proj.weight": "model-00006-of-00024.safetensors", "model.layers.55.self_attn.v_proj.weight": "model-00006-of-00024.safetensors", "model.layers.55.self_attn.k_proj.weight": "model-00006-of-00024.safetensors", "model.layers.55.self_attn.q_proj.weight": "model-00007-of-00024.safetensors", "model.layers.55.input_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.54.mlp.down_proj.weight": "model-00007-of-00024.safetensors", "model.layers.54.mlp.gate_proj.weight": "model-00007-of-00024.safetensors", "model.layers.54.mlp.up_proj.weight": "model-00007-of-00024.safetensors", "model.layers.54.post_attention_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.54.self_attn.o_proj.weight": "model-00007-of-00024.safetensors", "model.layers.54.self_attn.v_proj.weight": "model-00007-of-00024.safetensors", "model.layers.54.self_attn.k_proj.weight": "model-00007-of-00024.safetensors", "model.layers.54.self_attn.q_proj.weight": "model-00007-of-00024.safetensors", "model.layers.54.input_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.53.mlp.down_proj.weight": "model-00007-of-00024.safetensors", "model.layers.53.mlp.gate_proj.weight": "model-00007-of-00024.safetensors", "model.layers.53.mlp.up_proj.weight": "model-00007-of-00024.safetensors", "model.layers.53.post_attention_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.53.self_attn.o_proj.weight": "model-00007-of-00024.safetensors", "model.layers.53.self_attn.v_proj.weight": "model-00007-of-00024.safetensors", "model.layers.53.self_attn.k_proj.weight": "model-00007-of-00024.safetensors", "model.layers.53.self_attn.q_proj.weight": "model-00007-of-00024.safetensors", "model.layers.53.input_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.52.mlp.down_proj.weight": "model-00007-of-00024.safetensors", "model.layers.52.mlp.gate_proj.weight": "model-00007-of-00024.safetensors", "model.layers.52.mlp.up_proj.weight": "model-00007-of-00024.safetensors", "model.layers.52.post_attention_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.52.self_attn.o_proj.weight": "model-00007-of-00024.safetensors", "model.layers.52.self_attn.v_proj.weight": "model-00007-of-00024.safetensors", "model.layers.52.self_attn.k_proj.weight": "model-00007-of-00024.safetensors", "model.layers.52.self_attn.q_proj.weight": "model-00007-of-00024.safetensors", "model.layers.52.input_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.51.mlp.down_proj.weight": "model-00007-of-00024.safetensors", "model.layers.51.mlp.gate_proj.weight": "model-00007-of-00024.safetensors", "model.layers.51.mlp.up_proj.weight": "model-00007-of-00024.safetensors", "model.layers.51.post_attention_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.51.self_attn.o_proj.weight": "model-00007-of-00024.safetensors", "model.layers.51.self_attn.v_proj.weight": "model-00007-of-00024.safetensors", "model.layers.51.self_attn.k_proj.weight": "model-00007-of-00024.safetensors", "model.layers.51.self_attn.q_proj.weight": "model-00007-of-00024.safetensors", "model.layers.51.input_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.50.mlp.down_proj.weight": "model-00007-of-00024.safetensors", "model.layers.50.mlp.up_proj.weight": "model-00007-of-00024.safetensors", "model.layers.50.post_attention_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.50.input_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.15.mlp.down_proj.weight": "model-00007-of-00024.safetensors", "model.layers.15.mlp.gate_proj.weight": "model-00007-of-00024.safetensors", "model.layers.15.mlp.up_proj.weight": "model-00007-of-00024.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00007-of-00024.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00007-of-00024.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00007-of-00024.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00007-of-00024.safetensors", "model.layers.15.input_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.14.mlp.down_proj.weight": "model-00008-of-00024.safetensors", "model.layers.14.mlp.gate_proj.weight": "model-00008-of-00024.safetensors", "model.layers.14.mlp.up_proj.weight": "model-00008-of-00024.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00008-of-00024.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00008-of-00024.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00008-of-00024.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00008-of-00024.safetensors", "model.layers.14.input_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.13.mlp.down_proj.weight": "model-00008-of-00024.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.13.input_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.61.mlp.down_proj.weight": "model-00008-of-00024.safetensors", "model.layers.61.mlp.gate_proj.weight": "model-00008-of-00024.safetensors", "model.layers.61.mlp.up_proj.weight": "model-00008-of-00024.safetensors", "model.layers.61.post_attention_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.61.self_attn.o_proj.weight": "model-00008-of-00024.safetensors", "model.layers.61.self_attn.v_proj.weight": "model-00008-of-00024.safetensors", "model.layers.61.self_attn.k_proj.weight": "model-00008-of-00024.safetensors", "model.layers.61.self_attn.q_proj.weight": "model-00008-of-00024.safetensors", "model.layers.61.input_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.60.mlp.down_proj.weight": "model-00008-of-00024.safetensors", "model.layers.60.mlp.gate_proj.weight": "model-00008-of-00024.safetensors", "model.layers.60.mlp.up_proj.weight": "model-00008-of-00024.safetensors", "model.layers.60.post_attention_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.60.self_attn.o_proj.weight": "model-00008-of-00024.safetensors", "model.layers.60.self_attn.v_proj.weight": "model-00008-of-00024.safetensors", "model.layers.60.self_attn.k_proj.weight": "model-00008-of-00024.safetensors", "model.layers.60.self_attn.q_proj.weight": "model-00008-of-00024.safetensors", "model.layers.60.input_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.59.mlp.down_proj.weight": "model-00008-of-00024.safetensors", "model.layers.59.mlp.gate_proj.weight": "model-00008-of-00024.safetensors", "model.layers.59.mlp.up_proj.weight": "model-00008-of-00024.safetensors", "model.layers.59.post_attention_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.59.self_attn.o_proj.weight": "model-00008-of-00024.safetensors", "model.layers.59.self_attn.v_proj.weight": "model-00008-of-00024.safetensors", "model.layers.59.self_attn.k_proj.weight": "model-00008-of-00024.safetensors", "model.layers.59.self_attn.q_proj.weight": "model-00008-of-00024.safetensors", "model.layers.59.input_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.58.mlp.down_proj.weight": "model-00008-of-00024.safetensors", "model.layers.58.mlp.gate_proj.weight": "model-00008-of-00024.safetensors", "model.layers.58.mlp.up_proj.weight": "model-00008-of-00024.safetensors", "model.layers.58.post_attention_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.58.self_attn.o_proj.weight": "model-00008-of-00024.safetensors", "model.layers.58.self_attn.v_proj.weight": "model-00008-of-00024.safetensors", "model.layers.58.self_attn.k_proj.weight": "model-00008-of-00024.safetensors", "model.layers.58.self_attn.q_proj.weight": "model-00008-of-00024.safetensors", "model.layers.58.input_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.57.mlp.down_proj.weight": "model-00008-of-00024.safetensors", "model.layers.57.mlp.gate_proj.weight": "model-00008-of-00024.safetensors", "model.layers.57.mlp.up_proj.weight": "model-00009-of-00024.safetensors", "model.layers.57.post_attention_layernorm.weight": "model-00009-of-00024.safetensors", "model.layers.57.self_attn.o_proj.weight": "model-00009-of-00024.safetensors", "model.layers.57.self_attn.v_proj.weight": "model-00009-of-00024.safetensors", "model.layers.57.self_attn.k_proj.weight": "model-00009-of-00024.safetensors", "model.layers.57.self_attn.q_proj.weight": "model-00009-of-00024.safetensors", "model.layers.57.input_layernorm.weight": "model-00009-of-00024.safetensors", "model.layers.56.mlp.down_proj.weight": "model-00009-of-00024.safetensors", "model.layers.56.mlp.gate_proj.weight": "model-00009-of-00024.safetensors", "model.layers.56.mlp.up_proj.weight": "model-00009-of-00024.safetensors", "model.layers.56.post_attention_layernorm.weight": "model-00009-of-00024.safetensors", "model.layers.56.input_layernorm.weight": "model-00009-of-00024.safetensors", "model.layers.34.self_attn.o_proj.weight": "model-00009-of-00024.safetensors", "model.layers.34.self_attn.v_proj.weight": "model-00009-of-00024.safetensors", "model.layers.34.self_attn.k_proj.weight": "model-00009-of-00024.safetensors", "model.layers.34.self_attn.q_proj.weight": "model-00009-of-00024.safetensors", "model.layers.33.mlp.down_proj.weight": "model-00009-of-00024.safetensors", "model.layers.33.mlp.gate_proj.weight": "model-00009-of-00024.safetensors", "model.layers.33.mlp.up_proj.weight": "model-00009-of-00024.safetensors", "model.layers.33.post_attention_layernorm.weight": "model-00009-of-00024.safetensors", "model.layers.33.self_attn.o_proj.weight": "model-00009-of-00024.safetensors", "model.layers.33.self_attn.v_proj.weight": "model-00009-of-00024.safetensors", "model.layers.33.self_attn.k_proj.weight": "model-00009-of-00024.safetensors", "model.layers.33.self_attn.q_proj.weight": "model-00009-of-00024.safetensors", "model.layers.33.input_layernorm.weight": "model-00009-of-00024.safetensors", "model.layers.32.mlp.down_proj.weight": "model-00009-of-00024.safetensors", "model.layers.32.mlp.gate_proj.weight": "model-00009-of-00024.safetensors", "model.layers.32.mlp.up_proj.weight": "model-00009-of-00024.safetensors", "model.layers.32.post_attention_layernorm.weight": "model-00009-of-00024.safetensors", "model.layers.32.self_attn.o_proj.weight": "model-00009-of-00024.safetensors", "model.layers.32.self_attn.v_proj.weight": "model-00009-of-00024.safetensors", "model.layers.32.self_attn.k_proj.weight": "model-00009-of-00024.safetensors", "model.layers.32.self_attn.q_proj.weight": "model-00009-of-00024.safetensors", "model.layers.32.input_layernorm.weight": "model-00009-of-00024.safetensors", "model.layers.81.mlp.gate_proj.weight": "model-00009-of-00024.safetensors", "model.layers.81.mlp.up_proj.weight": "model-00009-of-00024.safetensors", "model.layers.81.self_attn.o_proj.weight": "model-00009-of-00024.safetensors", "model.layers.81.self_attn.v_proj.weight": "model-00009-of-00024.safetensors", "model.layers.81.self_attn.k_proj.weight": "model-00009-of-00024.safetensors", "model.layers.81.self_attn.q_proj.weight": "model-00009-of-00024.safetensors", "model.layers.80.mlp.down_proj.weight": "model-00009-of-00024.safetensors", "model.layers.80.mlp.gate_proj.weight": "model-00009-of-00024.safetensors", "model.layers.80.mlp.up_proj.weight": "model-00009-of-00024.safetensors", "model.layers.80.post_attention_layernorm.weight": "model-00009-of-00024.safetensors", "model.layers.80.self_attn.o_proj.weight": "model-00009-of-00024.safetensors", "model.layers.80.self_attn.v_proj.weight": "model-00009-of-00024.safetensors", "model.layers.80.self_attn.k_proj.weight": "model-00009-of-00024.safetensors", "model.layers.80.self_attn.q_proj.weight": "model-00009-of-00024.safetensors", "model.layers.80.input_layernorm.weight": "model-00009-of-00024.safetensors", "model.layers.79.mlp.down_proj.weight": "model-00009-of-00024.safetensors", "model.layers.79.mlp.gate_proj.weight": "model-00009-of-00024.safetensors", "model.layers.79.mlp.up_proj.weight": "model-00010-of-00024.safetensors", "model.layers.79.post_attention_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.79.self_attn.o_proj.weight": "model-00010-of-00024.safetensors", "model.layers.79.self_attn.v_proj.weight": "model-00010-of-00024.safetensors", "model.layers.79.self_attn.k_proj.weight": "model-00010-of-00024.safetensors", "model.layers.79.self_attn.q_proj.weight": "model-00010-of-00024.safetensors", "model.layers.79.input_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.78.mlp.down_proj.weight": "model-00010-of-00024.safetensors", "model.layers.78.mlp.gate_proj.weight": "model-00010-of-00024.safetensors", "model.layers.78.mlp.up_proj.weight": "model-00010-of-00024.safetensors", "model.layers.78.post_attention_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.78.self_attn.o_proj.weight": "model-00010-of-00024.safetensors", "model.layers.78.self_attn.v_proj.weight": "model-00010-of-00024.safetensors", "model.layers.78.self_attn.k_proj.weight": "model-00010-of-00024.safetensors", "model.layers.78.self_attn.q_proj.weight": "model-00010-of-00024.safetensors", "model.layers.78.input_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.77.mlp.down_proj.weight": "model-00010-of-00024.safetensors", "model.layers.77.mlp.gate_proj.weight": "model-00010-of-00024.safetensors", "model.layers.77.mlp.up_proj.weight": "model-00010-of-00024.safetensors", "model.layers.77.post_attention_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.77.self_attn.o_proj.weight": "model-00010-of-00024.safetensors", "model.layers.77.self_attn.v_proj.weight": "model-00010-of-00024.safetensors", "model.layers.77.self_attn.k_proj.weight": "model-00010-of-00024.safetensors", "model.layers.77.self_attn.q_proj.weight": "model-00010-of-00024.safetensors", "model.layers.77.input_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.37.self_attn.v_proj.weight": "model-00010-of-00024.safetensors", "model.layers.37.self_attn.k_proj.weight": "model-00010-of-00024.safetensors", "model.layers.37.self_attn.q_proj.weight": "model-00010-of-00024.safetensors", "model.layers.36.mlp.down_proj.weight": "model-00010-of-00024.safetensors", "model.layers.36.mlp.gate_proj.weight": "model-00010-of-00024.safetensors", "model.layers.36.mlp.up_proj.weight": "model-00010-of-00024.safetensors", "model.layers.36.post_attention_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.36.self_attn.o_proj.weight": "model-00010-of-00024.safetensors", "model.layers.36.self_attn.v_proj.weight": "model-00010-of-00024.safetensors", "model.layers.36.self_attn.k_proj.weight": "model-00010-of-00024.safetensors", "model.layers.36.self_attn.q_proj.weight": "model-00010-of-00024.safetensors", "model.layers.36.input_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.35.mlp.down_proj.weight": "model-00010-of-00024.safetensors", "model.layers.35.mlp.gate_proj.weight": "model-00010-of-00024.safetensors", "model.layers.35.mlp.up_proj.weight": "model-00010-of-00024.safetensors", "model.layers.35.post_attention_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.35.self_attn.o_proj.weight": "model-00010-of-00024.safetensors", "model.layers.35.self_attn.v_proj.weight": "model-00010-of-00024.safetensors", "model.layers.35.self_attn.k_proj.weight": "model-00010-of-00024.safetensors", "model.layers.35.self_attn.q_proj.weight": "model-00010-of-00024.safetensors", "model.layers.35.input_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.34.mlp.down_proj.weight": "model-00010-of-00024.safetensors", "model.layers.34.mlp.gate_proj.weight": "model-00010-of-00024.safetensors", "model.layers.34.mlp.up_proj.weight": "model-00010-of-00024.safetensors", "model.layers.34.post_attention_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.34.input_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.87.mlp.gate_proj.weight": "model-00010-of-00024.safetensors", "model.layers.87.self_attn.o_proj.weight": "model-00010-of-00024.safetensors", "model.layers.87.self_attn.v_proj.weight": "model-00010-of-00024.safetensors", "model.layers.87.self_attn.k_proj.weight": "model-00010-of-00024.safetensors", "model.layers.87.self_attn.q_proj.weight": "model-00010-of-00024.safetensors", "model.layers.86.mlp.down_proj.weight": "model-00011-of-00024.safetensors", "model.layers.86.mlp.gate_proj.weight": "model-00011-of-00024.safetensors", "model.layers.86.mlp.up_proj.weight": "model-00011-of-00024.safetensors", "model.layers.86.post_attention_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.86.self_attn.o_proj.weight": "model-00011-of-00024.safetensors", "model.layers.86.self_attn.v_proj.weight": "model-00011-of-00024.safetensors", "model.layers.86.self_attn.k_proj.weight": "model-00011-of-00024.safetensors", "model.layers.86.self_attn.q_proj.weight": "model-00011-of-00024.safetensors", "model.layers.86.input_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.85.mlp.down_proj.weight": "model-00011-of-00024.safetensors", "model.layers.85.mlp.gate_proj.weight": "model-00011-of-00024.safetensors", "model.layers.85.mlp.up_proj.weight": "model-00011-of-00024.safetensors", "model.layers.85.post_attention_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.85.self_attn.o_proj.weight": "model-00011-of-00024.safetensors", "model.layers.85.self_attn.v_proj.weight": "model-00011-of-00024.safetensors", "model.layers.85.self_attn.k_proj.weight": "model-00011-of-00024.safetensors", "model.layers.85.self_attn.q_proj.weight": "model-00011-of-00024.safetensors", "model.layers.85.input_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.84.mlp.down_proj.weight": "model-00011-of-00024.safetensors", "model.layers.84.mlp.gate_proj.weight": "model-00011-of-00024.safetensors", "model.layers.84.mlp.up_proj.weight": "model-00011-of-00024.safetensors", "model.layers.84.post_attention_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.84.self_attn.o_proj.weight": "model-00011-of-00024.safetensors", "model.layers.84.self_attn.v_proj.weight": "model-00011-of-00024.safetensors", "model.layers.84.self_attn.k_proj.weight": "model-00011-of-00024.safetensors", "model.layers.84.self_attn.q_proj.weight": "model-00011-of-00024.safetensors", "model.layers.84.input_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.83.mlp.down_proj.weight": "model-00011-of-00024.safetensors", "model.layers.83.mlp.gate_proj.weight": "model-00011-of-00024.safetensors", "model.layers.83.mlp.up_proj.weight": "model-00011-of-00024.safetensors", "model.layers.83.post_attention_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.83.self_attn.o_proj.weight": "model-00011-of-00024.safetensors", "model.layers.83.self_attn.v_proj.weight": "model-00011-of-00024.safetensors", "model.layers.83.self_attn.k_proj.weight": "model-00011-of-00024.safetensors", "model.layers.83.self_attn.q_proj.weight": "model-00011-of-00024.safetensors", "model.layers.83.input_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.82.mlp.down_proj.weight": "model-00011-of-00024.safetensors", "model.layers.82.mlp.gate_proj.weight": "model-00011-of-00024.safetensors", "model.layers.82.mlp.up_proj.weight": "model-00011-of-00024.safetensors", "model.layers.82.post_attention_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.82.self_attn.o_proj.weight": "model-00011-of-00024.safetensors", "model.layers.82.self_attn.v_proj.weight": "model-00011-of-00024.safetensors", "model.layers.82.self_attn.k_proj.weight": "model-00011-of-00024.safetensors", "model.layers.82.self_attn.q_proj.weight": "model-00011-of-00024.safetensors", "model.layers.82.input_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.81.mlp.down_proj.weight": "model-00011-of-00024.safetensors", "model.layers.81.post_attention_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.81.input_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.39.mlp.down_proj.weight": "model-00011-of-00024.safetensors", "model.layers.39.mlp.gate_proj.weight": "model-00011-of-00024.safetensors", "model.layers.39.mlp.up_proj.weight": "model-00012-of-00024.safetensors", "model.layers.39.post_attention_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.39.self_attn.o_proj.weight": "model-00012-of-00024.safetensors", "model.layers.39.self_attn.v_proj.weight": "model-00012-of-00024.safetensors", "model.layers.39.self_attn.k_proj.weight": "model-00012-of-00024.safetensors", "model.layers.39.self_attn.q_proj.weight": "model-00012-of-00024.safetensors", "model.layers.39.input_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.38.mlp.down_proj.weight": "model-00012-of-00024.safetensors", "model.layers.38.mlp.gate_proj.weight": "model-00012-of-00024.safetensors", "model.layers.38.mlp.up_proj.weight": "model-00012-of-00024.safetensors", "model.layers.38.post_attention_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.38.self_attn.o_proj.weight": "model-00012-of-00024.safetensors", "model.layers.38.self_attn.v_proj.weight": "model-00012-of-00024.safetensors", "model.layers.38.self_attn.k_proj.weight": "model-00012-of-00024.safetensors", "model.layers.38.self_attn.q_proj.weight": "model-00012-of-00024.safetensors", "model.layers.38.input_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.37.mlp.down_proj.weight": "model-00012-of-00024.safetensors", "model.layers.37.mlp.gate_proj.weight": "model-00012-of-00024.safetensors", "model.layers.37.mlp.up_proj.weight": "model-00012-of-00024.safetensors", "model.layers.37.post_attention_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.37.self_attn.o_proj.weight": "model-00012-of-00024.safetensors", "model.layers.37.input_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.107.self_attn.o_proj.weight": "model-00012-of-00024.safetensors", "model.layers.107.self_attn.v_proj.weight": "model-00012-of-00024.safetensors", "model.layers.107.self_attn.k_proj.weight": "model-00012-of-00024.safetensors", "model.layers.107.self_attn.q_proj.weight": "model-00012-of-00024.safetensors", "model.layers.91.mlp.down_proj.weight": "model-00012-of-00024.safetensors", "model.layers.91.mlp.gate_proj.weight": "model-00012-of-00024.safetensors", "model.layers.91.mlp.up_proj.weight": "model-00012-of-00024.safetensors", "model.layers.91.post_attention_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.91.self_attn.o_proj.weight": "model-00012-of-00024.safetensors", "model.layers.91.self_attn.v_proj.weight": "model-00012-of-00024.safetensors", "model.layers.91.self_attn.k_proj.weight": "model-00012-of-00024.safetensors", "model.layers.91.self_attn.q_proj.weight": "model-00012-of-00024.safetensors", "model.layers.91.input_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.90.mlp.down_proj.weight": "model-00012-of-00024.safetensors", "model.layers.90.mlp.gate_proj.weight": "model-00012-of-00024.safetensors", "model.layers.90.mlp.up_proj.weight": "model-00012-of-00024.safetensors", "model.layers.90.post_attention_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.90.self_attn.o_proj.weight": "model-00012-of-00024.safetensors", "model.layers.90.self_attn.v_proj.weight": "model-00012-of-00024.safetensors", "model.layers.90.self_attn.k_proj.weight": "model-00012-of-00024.safetensors", "model.layers.90.self_attn.q_proj.weight": "model-00012-of-00024.safetensors", "model.layers.90.input_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.89.mlp.down_proj.weight": "model-00012-of-00024.safetensors", "model.layers.89.mlp.gate_proj.weight": "model-00012-of-00024.safetensors", "model.layers.89.mlp.up_proj.weight": "model-00012-of-00024.safetensors", "model.layers.89.post_attention_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.89.self_attn.o_proj.weight": "model-00012-of-00024.safetensors", "model.layers.89.self_attn.v_proj.weight": "model-00012-of-00024.safetensors", "model.layers.89.self_attn.k_proj.weight": "model-00012-of-00024.safetensors", "model.layers.89.self_attn.q_proj.weight": "model-00012-of-00024.safetensors", "model.layers.89.input_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.88.mlp.down_proj.weight": "model-00012-of-00024.safetensors", "model.layers.88.mlp.gate_proj.weight": "model-00013-of-00024.safetensors", "model.layers.88.mlp.up_proj.weight": "model-00013-of-00024.safetensors", "model.layers.88.post_attention_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.88.self_attn.o_proj.weight": "model-00013-of-00024.safetensors", "model.layers.88.self_attn.v_proj.weight": "model-00013-of-00024.safetensors", "model.layers.88.self_attn.k_proj.weight": "model-00013-of-00024.safetensors", "model.layers.88.self_attn.q_proj.weight": "model-00013-of-00024.safetensors", "model.layers.88.input_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.87.mlp.down_proj.weight": "model-00013-of-00024.safetensors", "model.layers.87.mlp.up_proj.weight": "model-00013-of-00024.safetensors", "model.layers.87.post_attention_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.87.input_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.42.mlp.gate_proj.weight": "model-00013-of-00024.safetensors", "model.layers.42.mlp.up_proj.weight": "model-00013-of-00024.safetensors", "model.layers.42.self_attn.o_proj.weight": "model-00013-of-00024.safetensors", "model.layers.42.self_attn.v_proj.weight": "model-00013-of-00024.safetensors", "model.layers.42.self_attn.k_proj.weight": "model-00013-of-00024.safetensors", "model.layers.42.self_attn.q_proj.weight": "model-00013-of-00024.safetensors", "model.layers.41.mlp.down_proj.weight": "model-00013-of-00024.safetensors", "model.layers.41.mlp.gate_proj.weight": "model-00013-of-00024.safetensors", "model.layers.41.mlp.up_proj.weight": "model-00013-of-00024.safetensors", "model.layers.41.post_attention_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.41.self_attn.o_proj.weight": "model-00013-of-00024.safetensors", "model.layers.41.self_attn.v_proj.weight": "model-00013-of-00024.safetensors", "model.layers.41.self_attn.k_proj.weight": "model-00013-of-00024.safetensors", "model.layers.41.self_attn.q_proj.weight": "model-00013-of-00024.safetensors", "model.layers.41.input_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.40.mlp.down_proj.weight": "model-00013-of-00024.safetensors", "model.layers.40.mlp.gate_proj.weight": "model-00013-of-00024.safetensors", "model.layers.40.mlp.up_proj.weight": "model-00013-of-00024.safetensors", "model.layers.40.post_attention_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.40.self_attn.o_proj.weight": "model-00013-of-00024.safetensors", "model.layers.40.self_attn.v_proj.weight": "model-00013-of-00024.safetensors", "model.layers.40.self_attn.k_proj.weight": "model-00013-of-00024.safetensors", "model.layers.40.self_attn.q_proj.weight": "model-00013-of-00024.safetensors", "model.layers.40.input_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.112.mlp.down_proj.weight": "model-00013-of-00024.safetensors", "model.layers.112.mlp.gate_proj.weight": "model-00013-of-00024.safetensors", "model.layers.112.mlp.up_proj.weight": "model-00013-of-00024.safetensors", "model.layers.112.post_attention_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.112.self_attn.o_proj.weight": "model-00013-of-00024.safetensors", "model.layers.112.self_attn.v_proj.weight": "model-00013-of-00024.safetensors", "model.layers.112.self_attn.k_proj.weight": "model-00013-of-00024.safetensors", "model.layers.112.self_attn.q_proj.weight": "model-00013-of-00024.safetensors", "model.layers.112.input_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.111.mlp.down_proj.weight": "model-00013-of-00024.safetensors", "model.layers.111.mlp.gate_proj.weight": "model-00013-of-00024.safetensors", "model.layers.111.mlp.up_proj.weight": "model-00013-of-00024.safetensors", "model.layers.111.post_attention_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.111.self_attn.o_proj.weight": "model-00014-of-00024.safetensors", "model.layers.111.self_attn.v_proj.weight": "model-00014-of-00024.safetensors", "model.layers.111.self_attn.k_proj.weight": "model-00014-of-00024.safetensors", "model.layers.111.self_attn.q_proj.weight": "model-00014-of-00024.safetensors", "model.layers.111.input_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.110.mlp.down_proj.weight": "model-00014-of-00024.safetensors", "model.layers.110.mlp.gate_proj.weight": "model-00014-of-00024.safetensors", "model.layers.110.mlp.up_proj.weight": "model-00014-of-00024.safetensors", "model.layers.110.post_attention_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.110.self_attn.o_proj.weight": "model-00014-of-00024.safetensors", "model.layers.110.self_attn.v_proj.weight": "model-00014-of-00024.safetensors", "model.layers.110.self_attn.k_proj.weight": "model-00014-of-00024.safetensors", "model.layers.110.self_attn.q_proj.weight": "model-00014-of-00024.safetensors", "model.layers.110.input_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.109.mlp.down_proj.weight": "model-00014-of-00024.safetensors", "model.layers.109.mlp.gate_proj.weight": "model-00014-of-00024.safetensors", "model.layers.109.mlp.up_proj.weight": "model-00014-of-00024.safetensors", "model.layers.109.post_attention_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.109.self_attn.o_proj.weight": "model-00014-of-00024.safetensors", "model.layers.109.self_attn.v_proj.weight": "model-00014-of-00024.safetensors", "model.layers.109.self_attn.k_proj.weight": "model-00014-of-00024.safetensors", "model.layers.109.self_attn.q_proj.weight": "model-00014-of-00024.safetensors", "model.layers.109.input_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.108.mlp.down_proj.weight": "model-00014-of-00024.safetensors", "model.layers.108.mlp.gate_proj.weight": "model-00014-of-00024.safetensors", "model.layers.108.mlp.up_proj.weight": "model-00014-of-00024.safetensors", "model.layers.108.post_attention_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.108.self_attn.o_proj.weight": "model-00014-of-00024.safetensors", "model.layers.108.self_attn.v_proj.weight": "model-00014-of-00024.safetensors", "model.layers.108.self_attn.k_proj.weight": "model-00014-of-00024.safetensors", "model.layers.108.self_attn.q_proj.weight": "model-00014-of-00024.safetensors", "model.layers.108.input_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.107.mlp.down_proj.weight": "model-00014-of-00024.safetensors", "model.layers.107.mlp.gate_proj.weight": "model-00014-of-00024.safetensors", "model.layers.107.mlp.up_proj.weight": "model-00014-of-00024.safetensors", "model.layers.107.post_attention_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.107.input_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.45.mlp.gate_proj.weight": "model-00014-of-00024.safetensors", "model.layers.45.self_attn.o_proj.weight": "model-00014-of-00024.safetensors", "model.layers.45.self_attn.v_proj.weight": "model-00014-of-00024.safetensors", "model.layers.45.self_attn.k_proj.weight": "model-00014-of-00024.safetensors", "model.layers.45.self_attn.q_proj.weight": "model-00014-of-00024.safetensors", "model.layers.44.mlp.down_proj.weight": "model-00014-of-00024.safetensors", "model.layers.44.mlp.gate_proj.weight": "model-00014-of-00024.safetensors", "model.layers.44.mlp.up_proj.weight": "model-00014-of-00024.safetensors", "model.layers.44.post_attention_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.44.self_attn.o_proj.weight": "model-00014-of-00024.safetensors", "model.layers.44.self_attn.v_proj.weight": "model-00014-of-00024.safetensors", "model.layers.44.self_attn.k_proj.weight": "model-00014-of-00024.safetensors", "model.layers.44.self_attn.q_proj.weight": "model-00014-of-00024.safetensors", "model.layers.44.input_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.43.mlp.down_proj.weight": "model-00014-of-00024.safetensors", "model.layers.43.mlp.gate_proj.weight": "model-00015-of-00024.safetensors", "model.layers.43.mlp.up_proj.weight": "model-00015-of-00024.safetensors", "model.layers.43.post_attention_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.43.self_attn.o_proj.weight": "model-00015-of-00024.safetensors", "model.layers.43.self_attn.v_proj.weight": "model-00015-of-00024.safetensors", "model.layers.43.self_attn.k_proj.weight": "model-00015-of-00024.safetensors", "model.layers.43.self_attn.q_proj.weight": "model-00015-of-00024.safetensors", "model.layers.43.input_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.42.mlp.down_proj.weight": "model-00015-of-00024.safetensors", "model.layers.42.post_attention_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.42.input_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.118.mlp.gate_proj.weight": "model-00015-of-00024.safetensors", "model.layers.118.mlp.up_proj.weight": "model-00015-of-00024.safetensors", "model.layers.118.self_attn.o_proj.weight": "model-00015-of-00024.safetensors", "model.layers.118.self_attn.v_proj.weight": "model-00015-of-00024.safetensors", "model.layers.118.self_attn.k_proj.weight": "model-00015-of-00024.safetensors", "model.layers.118.self_attn.q_proj.weight": "model-00015-of-00024.safetensors", "model.layers.117.mlp.down_proj.weight": "model-00015-of-00024.safetensors", "model.layers.117.mlp.gate_proj.weight": "model-00015-of-00024.safetensors", "model.layers.117.mlp.up_proj.weight": "model-00015-of-00024.safetensors", "model.layers.117.post_attention_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.117.self_attn.o_proj.weight": "model-00015-of-00024.safetensors", "model.layers.117.self_attn.v_proj.weight": "model-00015-of-00024.safetensors", "model.layers.117.self_attn.k_proj.weight": "model-00015-of-00024.safetensors", "model.layers.117.self_attn.q_proj.weight": "model-00015-of-00024.safetensors", "model.layers.117.input_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.116.mlp.down_proj.weight": "model-00015-of-00024.safetensors", "model.layers.116.mlp.gate_proj.weight": "model-00015-of-00024.safetensors", "model.layers.116.mlp.up_proj.weight": "model-00015-of-00024.safetensors", "model.layers.116.post_attention_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.116.self_attn.o_proj.weight": "model-00015-of-00024.safetensors", "model.layers.116.self_attn.v_proj.weight": "model-00015-of-00024.safetensors", "model.layers.116.self_attn.k_proj.weight": "model-00015-of-00024.safetensors", "model.layers.116.self_attn.q_proj.weight": "model-00015-of-00024.safetensors", "model.layers.116.input_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.115.mlp.down_proj.weight": "model-00015-of-00024.safetensors", "model.layers.115.mlp.gate_proj.weight": "model-00015-of-00024.safetensors", "model.layers.115.mlp.up_proj.weight": "model-00015-of-00024.safetensors", "model.layers.115.post_attention_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.115.self_attn.o_proj.weight": "model-00015-of-00024.safetensors", "model.layers.115.self_attn.v_proj.weight": "model-00015-of-00024.safetensors", "model.layers.115.self_attn.k_proj.weight": "model-00015-of-00024.safetensors", "model.layers.115.self_attn.q_proj.weight": "model-00015-of-00024.safetensors", "model.layers.115.input_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.114.mlp.down_proj.weight": "model-00015-of-00024.safetensors", "model.layers.114.mlp.gate_proj.weight": "model-00015-of-00024.safetensors", "model.layers.114.mlp.up_proj.weight": "model-00015-of-00024.safetensors", "model.layers.114.post_attention_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.114.self_attn.o_proj.weight": "model-00015-of-00024.safetensors", "model.layers.114.self_attn.v_proj.weight": "model-00015-of-00024.safetensors", "model.layers.114.self_attn.k_proj.weight": "model-00015-of-00024.safetensors", "model.layers.114.self_attn.q_proj.weight": "model-00015-of-00024.safetensors", "model.layers.114.input_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.113.mlp.down_proj.weight": "model-00016-of-00024.safetensors", "model.layers.113.mlp.gate_proj.weight": "model-00016-of-00024.safetensors", "model.layers.113.mlp.up_proj.weight": "model-00016-of-00024.safetensors", "model.layers.113.post_attention_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.113.self_attn.o_proj.weight": "model-00016-of-00024.safetensors", "model.layers.113.self_attn.v_proj.weight": "model-00016-of-00024.safetensors", "model.layers.113.self_attn.k_proj.weight": "model-00016-of-00024.safetensors", "model.layers.113.self_attn.q_proj.weight": "model-00016-of-00024.safetensors", "model.layers.113.input_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.62.self_attn.o_proj.weight": "model-00016-of-00024.safetensors", "model.layers.62.self_attn.v_proj.weight": "model-00016-of-00024.safetensors", "model.layers.62.self_attn.k_proj.weight": "model-00016-of-00024.safetensors", "model.layers.62.self_attn.q_proj.weight": "model-00016-of-00024.safetensors", "model.layers.46.mlp.down_proj.weight": "model-00016-of-00024.safetensors", "model.layers.46.mlp.gate_proj.weight": "model-00016-of-00024.safetensors", "model.layers.46.mlp.up_proj.weight": "model-00016-of-00024.safetensors", "model.layers.46.post_attention_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.46.self_attn.o_proj.weight": "model-00016-of-00024.safetensors", "model.layers.46.self_attn.v_proj.weight": "model-00016-of-00024.safetensors", "model.layers.46.self_attn.k_proj.weight": "model-00016-of-00024.safetensors", "model.layers.46.self_attn.q_proj.weight": "model-00016-of-00024.safetensors", "model.layers.46.input_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.45.mlp.down_proj.weight": "model-00016-of-00024.safetensors", "model.layers.45.mlp.up_proj.weight": "model-00016-of-00024.safetensors", "model.layers.45.post_attention_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.45.input_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.121.mlp.down_proj.weight": "model-00016-of-00024.safetensors", "model.layers.121.mlp.gate_proj.weight": "model-00016-of-00024.safetensors", "model.layers.121.mlp.up_proj.weight": "model-00016-of-00024.safetensors", "model.layers.121.post_attention_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.121.self_attn.o_proj.weight": "model-00016-of-00024.safetensors", "model.layers.121.self_attn.v_proj.weight": "model-00016-of-00024.safetensors", "model.layers.121.self_attn.k_proj.weight": "model-00016-of-00024.safetensors", "model.layers.121.self_attn.q_proj.weight": "model-00016-of-00024.safetensors", "model.layers.121.input_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.120.mlp.down_proj.weight": "model-00016-of-00024.safetensors", "model.layers.120.mlp.gate_proj.weight": "model-00016-of-00024.safetensors", "model.layers.120.mlp.up_proj.weight": "model-00016-of-00024.safetensors", "model.layers.120.post_attention_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.120.self_attn.o_proj.weight": "model-00016-of-00024.safetensors", "model.layers.120.self_attn.v_proj.weight": "model-00016-of-00024.safetensors", "model.layers.120.self_attn.k_proj.weight": "model-00016-of-00024.safetensors", "model.layers.120.self_attn.q_proj.weight": "model-00016-of-00024.safetensors", "model.layers.120.input_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.119.mlp.down_proj.weight": "model-00016-of-00024.safetensors", "model.layers.119.mlp.gate_proj.weight": "model-00016-of-00024.safetensors", "model.layers.119.mlp.up_proj.weight": "model-00016-of-00024.safetensors", "model.layers.119.post_attention_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.119.self_attn.o_proj.weight": "model-00016-of-00024.safetensors", "model.layers.119.self_attn.v_proj.weight": "model-00016-of-00024.safetensors", "model.layers.119.self_attn.k_proj.weight": "model-00016-of-00024.safetensors", "model.layers.119.self_attn.q_proj.weight": "model-00016-of-00024.safetensors", "model.layers.119.input_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.118.mlp.down_proj.weight": "model-00017-of-00024.safetensors", "model.layers.118.post_attention_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.118.input_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.65.self_attn.v_proj.weight": "model-00017-of-00024.safetensors", "model.layers.65.self_attn.k_proj.weight": "model-00017-of-00024.safetensors", "model.layers.65.self_attn.q_proj.weight": "model-00017-of-00024.safetensors", "model.layers.64.mlp.down_proj.weight": "model-00017-of-00024.safetensors", "model.layers.64.mlp.gate_proj.weight": "model-00017-of-00024.safetensors", "model.layers.64.mlp.up_proj.weight": "model-00017-of-00024.safetensors", "model.layers.64.post_attention_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.64.self_attn.o_proj.weight": "model-00017-of-00024.safetensors", "model.layers.64.self_attn.v_proj.weight": "model-00017-of-00024.safetensors", "model.layers.64.self_attn.k_proj.weight": "model-00017-of-00024.safetensors", "model.layers.64.self_attn.q_proj.weight": "model-00017-of-00024.safetensors", "model.layers.64.input_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.63.mlp.down_proj.weight": "model-00017-of-00024.safetensors", "model.layers.63.mlp.gate_proj.weight": "model-00017-of-00024.safetensors", "model.layers.63.mlp.up_proj.weight": "model-00017-of-00024.safetensors", "model.layers.63.post_attention_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.63.self_attn.o_proj.weight": "model-00017-of-00024.safetensors", "model.layers.63.self_attn.v_proj.weight": "model-00017-of-00024.safetensors", "model.layers.63.self_attn.k_proj.weight": "model-00017-of-00024.safetensors", "model.layers.63.self_attn.q_proj.weight": "model-00017-of-00024.safetensors", "model.layers.63.input_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.62.mlp.down_proj.weight": "model-00017-of-00024.safetensors", "model.layers.62.mlp.gate_proj.weight": "model-00017-of-00024.safetensors", "model.layers.62.mlp.up_proj.weight": "model-00017-of-00024.safetensors", "model.layers.62.post_attention_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.62.input_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.67.mlp.down_proj.weight": "model-00017-of-00024.safetensors", "model.layers.67.mlp.gate_proj.weight": "model-00017-of-00024.safetensors", "model.layers.67.mlp.up_proj.weight": "model-00017-of-00024.safetensors", "model.layers.67.post_attention_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.67.self_attn.o_proj.weight": "model-00017-of-00024.safetensors", "model.layers.67.self_attn.v_proj.weight": "model-00017-of-00024.safetensors", "model.layers.67.self_attn.k_proj.weight": "model-00017-of-00024.safetensors", "model.layers.67.self_attn.q_proj.weight": "model-00017-of-00024.safetensors", "model.layers.67.input_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.66.mlp.down_proj.weight": "model-00017-of-00024.safetensors", "model.layers.66.mlp.gate_proj.weight": "model-00017-of-00024.safetensors", "model.layers.66.mlp.up_proj.weight": "model-00017-of-00024.safetensors", "model.layers.66.post_attention_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.66.self_attn.o_proj.weight": "model-00017-of-00024.safetensors", "model.layers.66.self_attn.v_proj.weight": "model-00017-of-00024.safetensors", "model.layers.66.self_attn.k_proj.weight": "model-00017-of-00024.safetensors", "model.layers.66.self_attn.q_proj.weight": "model-00017-of-00024.safetensors", "model.layers.66.input_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.65.mlp.down_proj.weight": "model-00017-of-00024.safetensors", "model.layers.65.mlp.gate_proj.weight": "model-00017-of-00024.safetensors", "model.layers.65.mlp.up_proj.weight": "model-00018-of-00024.safetensors", "model.layers.65.post_attention_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.65.self_attn.o_proj.weight": "model-00018-of-00024.safetensors", "model.layers.65.input_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.70.mlp.gate_proj.weight": "model-00018-of-00024.safetensors", "model.layers.70.mlp.up_proj.weight": "model-00018-of-00024.safetensors", "model.layers.70.self_attn.o_proj.weight": "model-00018-of-00024.safetensors", "model.layers.70.self_attn.v_proj.weight": "model-00018-of-00024.safetensors", "model.layers.70.self_attn.k_proj.weight": "model-00018-of-00024.safetensors", "model.layers.70.self_attn.q_proj.weight": "model-00018-of-00024.safetensors", "model.layers.69.mlp.down_proj.weight": "model-00018-of-00024.safetensors", "model.layers.69.mlp.gate_proj.weight": "model-00018-of-00024.safetensors", "model.layers.69.mlp.up_proj.weight": "model-00018-of-00024.safetensors", "model.layers.69.post_attention_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.69.self_attn.o_proj.weight": "model-00018-of-00024.safetensors", "model.layers.69.self_attn.v_proj.weight": "model-00018-of-00024.safetensors", "model.layers.69.self_attn.k_proj.weight": "model-00018-of-00024.safetensors", "model.layers.69.self_attn.q_proj.weight": "model-00018-of-00024.safetensors", "model.layers.69.input_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.68.mlp.down_proj.weight": "model-00018-of-00024.safetensors", "model.layers.68.mlp.gate_proj.weight": "model-00018-of-00024.safetensors", "model.layers.68.mlp.up_proj.weight": "model-00018-of-00024.safetensors", "model.layers.68.post_attention_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.68.self_attn.o_proj.weight": "model-00018-of-00024.safetensors", "model.layers.68.self_attn.v_proj.weight": "model-00018-of-00024.safetensors", "model.layers.68.self_attn.k_proj.weight": "model-00018-of-00024.safetensors", "model.layers.68.self_attn.q_proj.weight": "model-00018-of-00024.safetensors", "model.layers.68.input_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.73.mlp.gate_proj.weight": "model-00018-of-00024.safetensors", "model.layers.73.self_attn.o_proj.weight": "model-00018-of-00024.safetensors", "model.layers.73.self_attn.v_proj.weight": "model-00018-of-00024.safetensors", "model.layers.73.self_attn.k_proj.weight": "model-00018-of-00024.safetensors", "model.layers.73.self_attn.q_proj.weight": "model-00018-of-00024.safetensors", "model.layers.72.mlp.down_proj.weight": "model-00018-of-00024.safetensors", "model.layers.72.mlp.gate_proj.weight": "model-00018-of-00024.safetensors", "model.layers.72.mlp.up_proj.weight": "model-00018-of-00024.safetensors", "model.layers.72.post_attention_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.72.self_attn.o_proj.weight": "model-00018-of-00024.safetensors", "model.layers.72.self_attn.v_proj.weight": "model-00018-of-00024.safetensors", "model.layers.72.self_attn.k_proj.weight": "model-00018-of-00024.safetensors", "model.layers.72.self_attn.q_proj.weight": "model-00018-of-00024.safetensors", "model.layers.72.input_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.71.mlp.down_proj.weight": "model-00018-of-00024.safetensors", "model.layers.71.mlp.gate_proj.weight": "model-00018-of-00024.safetensors", "model.layers.71.mlp.up_proj.weight": "model-00018-of-00024.safetensors", "model.layers.71.post_attention_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.71.self_attn.o_proj.weight": "model-00018-of-00024.safetensors", "model.layers.71.self_attn.v_proj.weight": "model-00018-of-00024.safetensors", "model.layers.71.self_attn.k_proj.weight": "model-00018-of-00024.safetensors", "model.layers.71.self_attn.q_proj.weight": "model-00018-of-00024.safetensors", "model.layers.71.input_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.70.mlp.down_proj.weight": "model-00018-of-00024.safetensors", "model.layers.70.post_attention_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.70.input_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.76.self_attn.o_proj.weight": "model-00019-of-00024.safetensors", "model.layers.76.self_attn.v_proj.weight": "model-00019-of-00024.safetensors", "model.layers.76.self_attn.k_proj.weight": "model-00019-of-00024.safetensors", "model.layers.76.self_attn.q_proj.weight": "model-00019-of-00024.safetensors", "model.layers.75.mlp.down_proj.weight": "model-00019-of-00024.safetensors", "model.layers.75.mlp.gate_proj.weight": "model-00019-of-00024.safetensors", "model.layers.75.mlp.up_proj.weight": "model-00019-of-00024.safetensors", "model.layers.75.post_attention_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.75.self_attn.o_proj.weight": "model-00019-of-00024.safetensors", "model.layers.75.self_attn.v_proj.weight": "model-00019-of-00024.safetensors", "model.layers.75.self_attn.k_proj.weight": "model-00019-of-00024.safetensors", "model.layers.75.self_attn.q_proj.weight": "model-00019-of-00024.safetensors", "model.layers.75.input_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.74.mlp.down_proj.weight": "model-00019-of-00024.safetensors", "model.layers.74.mlp.gate_proj.weight": "model-00019-of-00024.safetensors", "model.layers.74.mlp.up_proj.weight": "model-00019-of-00024.safetensors", "model.layers.74.post_attention_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.74.self_attn.o_proj.weight": "model-00019-of-00024.safetensors", "model.layers.74.self_attn.v_proj.weight": "model-00019-of-00024.safetensors", "model.layers.74.self_attn.k_proj.weight": "model-00019-of-00024.safetensors", "model.layers.74.self_attn.q_proj.weight": "model-00019-of-00024.safetensors", "model.layers.74.input_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.73.mlp.down_proj.weight": "model-00019-of-00024.safetensors", "model.layers.73.mlp.up_proj.weight": "model-00019-of-00024.safetensors", "model.layers.73.post_attention_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.73.input_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.93.self_attn.v_proj.weight": "model-00019-of-00024.safetensors", "model.layers.93.self_attn.k_proj.weight": "model-00019-of-00024.safetensors", "model.layers.93.self_attn.q_proj.weight": "model-00019-of-00024.safetensors", "model.layers.92.mlp.down_proj.weight": "model-00019-of-00024.safetensors", "model.layers.92.mlp.gate_proj.weight": "model-00019-of-00024.safetensors", "model.layers.92.mlp.up_proj.weight": "model-00019-of-00024.safetensors", "model.layers.92.post_attention_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.92.self_attn.o_proj.weight": "model-00019-of-00024.safetensors", "model.layers.92.self_attn.v_proj.weight": "model-00019-of-00024.safetensors", "model.layers.92.self_attn.k_proj.weight": "model-00019-of-00024.safetensors", "model.layers.92.self_attn.q_proj.weight": "model-00019-of-00024.safetensors", "model.layers.92.input_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.76.mlp.down_proj.weight": "model-00019-of-00024.safetensors", "model.layers.76.mlp.gate_proj.weight": "model-00019-of-00024.safetensors", "model.layers.76.mlp.up_proj.weight": "model-00019-of-00024.safetensors", "model.layers.76.post_attention_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.76.input_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.95.mlp.down_proj.weight": "model-00019-of-00024.safetensors", "model.layers.95.mlp.gate_proj.weight": "model-00019-of-00024.safetensors", "model.layers.95.mlp.up_proj.weight": "model-00019-of-00024.safetensors", "model.layers.95.post_attention_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.95.self_attn.o_proj.weight": "model-00019-of-00024.safetensors", "model.layers.95.self_attn.v_proj.weight": "model-00019-of-00024.safetensors", "model.layers.95.self_attn.k_proj.weight": "model-00019-of-00024.safetensors", "model.layers.95.self_attn.q_proj.weight": "model-00019-of-00024.safetensors", "model.layers.95.input_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.94.mlp.down_proj.weight": "model-00020-of-00024.safetensors", "model.layers.94.mlp.gate_proj.weight": "model-00020-of-00024.safetensors", "model.layers.94.mlp.up_proj.weight": "model-00020-of-00024.safetensors", "model.layers.94.post_attention_layernorm.weight": "model-00020-of-00024.safetensors", "model.layers.94.self_attn.o_proj.weight": "model-00020-of-00024.safetensors", "model.layers.94.self_attn.v_proj.weight": "model-00020-of-00024.safetensors", "model.layers.94.self_attn.k_proj.weight": "model-00020-of-00024.safetensors", "model.layers.94.self_attn.q_proj.weight": "model-00020-of-00024.safetensors", "model.layers.94.input_layernorm.weight": "model-00020-of-00024.safetensors", "model.layers.93.mlp.down_proj.weight": "model-00020-of-00024.safetensors", "model.layers.93.mlp.gate_proj.weight": "model-00020-of-00024.safetensors", "model.layers.93.mlp.up_proj.weight": "model-00020-of-00024.safetensors", "model.layers.93.post_attention_layernorm.weight": "model-00020-of-00024.safetensors", "model.layers.93.self_attn.o_proj.weight": "model-00020-of-00024.safetensors", "model.layers.93.input_layernorm.weight": "model-00020-of-00024.safetensors", "model.layers.98.mlp.gate_proj.weight": "model-00020-of-00024.safetensors", "model.layers.98.mlp.up_proj.weight": "model-00020-of-00024.safetensors", "model.layers.98.self_attn.o_proj.weight": "model-00020-of-00024.safetensors", "model.layers.98.self_attn.v_proj.weight": "model-00020-of-00024.safetensors", "model.layers.98.self_attn.k_proj.weight": "model-00020-of-00024.safetensors", "model.layers.98.self_attn.q_proj.weight": "model-00020-of-00024.safetensors", "model.layers.97.mlp.down_proj.weight": "model-00020-of-00024.safetensors", "model.layers.97.mlp.gate_proj.weight": "model-00020-of-00024.safetensors", "model.layers.97.mlp.up_proj.weight": "model-00020-of-00024.safetensors", "model.layers.97.post_attention_layernorm.weight": "model-00020-of-00024.safetensors", "model.layers.97.self_attn.o_proj.weight": "model-00020-of-00024.safetensors", "model.layers.97.self_attn.v_proj.weight": "model-00020-of-00024.safetensors", "model.layers.97.self_attn.k_proj.weight": "model-00020-of-00024.safetensors", "model.layers.97.self_attn.q_proj.weight": "model-00020-of-00024.safetensors", "model.layers.97.input_layernorm.weight": "model-00020-of-00024.safetensors", "model.layers.96.mlp.down_proj.weight": "model-00020-of-00024.safetensors", "model.layers.96.mlp.gate_proj.weight": "model-00020-of-00024.safetensors", "model.layers.96.mlp.up_proj.weight": "model-00020-of-00024.safetensors", "model.layers.96.post_attention_layernorm.weight": "model-00020-of-00024.safetensors", "model.layers.96.self_attn.o_proj.weight": "model-00020-of-00024.safetensors", "model.layers.96.self_attn.v_proj.weight": "model-00020-of-00024.safetensors", "model.layers.96.self_attn.k_proj.weight": "model-00020-of-00024.safetensors", "model.layers.96.self_attn.q_proj.weight": "model-00020-of-00024.safetensors", "model.layers.96.input_layernorm.weight": "model-00020-of-00024.safetensors", "model.layers.101.mlp.gate_proj.weight": "model-00020-of-00024.safetensors", "model.layers.101.self_attn.o_proj.weight": "model-00020-of-00024.safetensors", "model.layers.101.self_attn.v_proj.weight": "model-00020-of-00024.safetensors", "model.layers.101.self_attn.k_proj.weight": "model-00020-of-00024.safetensors", "model.layers.101.self_attn.q_proj.weight": "model-00020-of-00024.safetensors", "model.layers.100.mlp.down_proj.weight": "model-00020-of-00024.safetensors", "model.layers.100.mlp.gate_proj.weight": "model-00020-of-00024.safetensors", "model.layers.100.mlp.up_proj.weight": "model-00021-of-00024.safetensors", "model.layers.100.post_attention_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.100.self_attn.o_proj.weight": "model-00021-of-00024.safetensors", "model.layers.100.self_attn.v_proj.weight": "model-00021-of-00024.safetensors", "model.layers.100.self_attn.k_proj.weight": "model-00021-of-00024.safetensors", "model.layers.100.self_attn.q_proj.weight": "model-00021-of-00024.safetensors", "model.layers.100.input_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.99.mlp.down_proj.weight": "model-00021-of-00024.safetensors", "model.layers.99.mlp.gate_proj.weight": "model-00021-of-00024.safetensors", "model.layers.99.mlp.up_proj.weight": "model-00021-of-00024.safetensors", "model.layers.99.post_attention_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.99.self_attn.o_proj.weight": "model-00021-of-00024.safetensors", "model.layers.99.self_attn.v_proj.weight": "model-00021-of-00024.safetensors", "model.layers.99.self_attn.k_proj.weight": "model-00021-of-00024.safetensors", "model.layers.99.self_attn.q_proj.weight": "model-00021-of-00024.safetensors", "model.layers.99.input_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.98.mlp.down_proj.weight": "model-00021-of-00024.safetensors", "model.layers.98.post_attention_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.98.input_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.104.self_attn.o_proj.weight": "model-00021-of-00024.safetensors", "model.layers.104.self_attn.v_proj.weight": "model-00021-of-00024.safetensors", "model.layers.104.self_attn.k_proj.weight": "model-00021-of-00024.safetensors", "model.layers.104.self_attn.q_proj.weight": "model-00021-of-00024.safetensors", "model.layers.103.mlp.down_proj.weight": "model-00021-of-00024.safetensors", "model.layers.103.mlp.gate_proj.weight": "model-00021-of-00024.safetensors", "model.layers.103.mlp.up_proj.weight": "model-00021-of-00024.safetensors", "model.layers.103.post_attention_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.103.self_attn.o_proj.weight": "model-00021-of-00024.safetensors", "model.layers.103.self_attn.v_proj.weight": "model-00021-of-00024.safetensors", "model.layers.103.self_attn.k_proj.weight": "model-00021-of-00024.safetensors", "model.layers.103.self_attn.q_proj.weight": "model-00021-of-00024.safetensors", "model.layers.103.input_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.102.mlp.down_proj.weight": "model-00021-of-00024.safetensors", "model.layers.102.mlp.gate_proj.weight": "model-00021-of-00024.safetensors", "model.layers.102.mlp.up_proj.weight": "model-00021-of-00024.safetensors", "model.layers.102.post_attention_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.102.self_attn.o_proj.weight": "model-00021-of-00024.safetensors", "model.layers.102.self_attn.v_proj.weight": "model-00021-of-00024.safetensors", "model.layers.102.self_attn.k_proj.weight": "model-00021-of-00024.safetensors", "model.layers.102.self_attn.q_proj.weight": "model-00021-of-00024.safetensors", "model.layers.102.input_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.101.mlp.down_proj.weight": "model-00021-of-00024.safetensors", "model.layers.101.mlp.up_proj.weight": "model-00021-of-00024.safetensors", "model.layers.101.post_attention_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.101.input_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.106.mlp.down_proj.weight": "model-00021-of-00024.safetensors", "model.layers.106.mlp.gate_proj.weight": "model-00021-of-00024.safetensors", "model.layers.106.mlp.up_proj.weight": "model-00021-of-00024.safetensors", "model.layers.106.post_attention_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.106.self_attn.o_proj.weight": "model-00021-of-00024.safetensors", "model.layers.106.self_attn.v_proj.weight": "model-00021-of-00024.safetensors", "model.layers.106.self_attn.k_proj.weight": "model-00021-of-00024.safetensors", "model.layers.106.self_attn.q_proj.weight": "model-00021-of-00024.safetensors", "model.layers.106.input_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.105.mlp.down_proj.weight": "model-00021-of-00024.safetensors", "model.layers.105.mlp.gate_proj.weight": "model-00022-of-00024.safetensors", "model.layers.105.mlp.up_proj.weight": "model-00022-of-00024.safetensors", "model.layers.105.post_attention_layernorm.weight": "model-00022-of-00024.safetensors", "model.layers.105.self_attn.o_proj.weight": "model-00022-of-00024.safetensors", "model.layers.105.self_attn.v_proj.weight": "model-00022-of-00024.safetensors", "model.layers.105.self_attn.k_proj.weight": "model-00022-of-00024.safetensors", "model.layers.105.self_attn.q_proj.weight": "model-00022-of-00024.safetensors", "model.layers.105.input_layernorm.weight": "model-00022-of-00024.safetensors", "model.layers.104.mlp.down_proj.weight": "model-00022-of-00024.safetensors", "model.layers.104.mlp.gate_proj.weight": "model-00022-of-00024.safetensors", "model.layers.104.mlp.up_proj.weight": "model-00022-of-00024.safetensors", "model.layers.104.post_attention_layernorm.weight": "model-00022-of-00024.safetensors", "model.layers.104.input_layernorm.weight": "model-00022-of-00024.safetensors", "model.layers.123.mlp.down_proj.weight": "model-00022-of-00024.safetensors", "model.layers.123.mlp.gate_proj.weight": "model-00022-of-00024.safetensors", "model.layers.123.mlp.up_proj.weight": "model-00022-of-00024.safetensors", "model.layers.123.post_attention_layernorm.weight": "model-00022-of-00024.safetensors", "model.layers.123.self_attn.o_proj.weight": "model-00022-of-00024.safetensors", "model.layers.123.self_attn.v_proj.weight": "model-00022-of-00024.safetensors", "model.layers.123.self_attn.k_proj.weight": "model-00022-of-00024.safetensors", "model.layers.123.self_attn.q_proj.weight": "model-00022-of-00024.safetensors", "model.layers.123.input_layernorm.weight": "model-00022-of-00024.safetensors", "model.layers.122.mlp.down_proj.weight": "model-00022-of-00024.safetensors", "model.layers.122.mlp.gate_proj.weight": "model-00022-of-00024.safetensors", "model.layers.122.mlp.up_proj.weight": "model-00022-of-00024.safetensors", "model.layers.122.post_attention_layernorm.weight": "model-00022-of-00024.safetensors", "model.layers.122.self_attn.o_proj.weight": "model-00022-of-00024.safetensors", "model.layers.122.self_attn.v_proj.weight": "model-00022-of-00024.safetensors", "model.layers.122.self_attn.k_proj.weight": "model-00022-of-00024.safetensors", "model.layers.122.self_attn.q_proj.weight": "model-00022-of-00024.safetensors", "model.layers.122.input_layernorm.weight": "model-00022-of-00024.safetensors", "model.layers.126.mlp.gate_proj.weight": "model-00022-of-00024.safetensors", "model.layers.126.mlp.up_proj.weight": "model-00022-of-00024.safetensors", "model.layers.126.self_attn.o_proj.weight": "model-00022-of-00024.safetensors", "model.layers.126.self_attn.v_proj.weight": "model-00022-of-00024.safetensors", "model.layers.126.self_attn.k_proj.weight": "model-00022-of-00024.safetensors", "model.layers.126.self_attn.q_proj.weight": "model-00022-of-00024.safetensors", "model.layers.125.mlp.down_proj.weight": "model-00022-of-00024.safetensors", "model.layers.125.mlp.gate_proj.weight": "model-00022-of-00024.safetensors", "model.layers.125.mlp.up_proj.weight": "model-00022-of-00024.safetensors", "model.layers.125.post_attention_layernorm.weight": "model-00022-of-00024.safetensors", "model.layers.125.self_attn.o_proj.weight": "model-00022-of-00024.safetensors", "model.layers.125.self_attn.v_proj.weight": "model-00022-of-00024.safetensors", "model.layers.125.self_attn.k_proj.weight": "model-00022-of-00024.safetensors", "model.layers.125.self_attn.q_proj.weight": "model-00022-of-00024.safetensors", "model.layers.125.input_layernorm.weight": "model-00022-of-00024.safetensors", "model.layers.124.mlp.down_proj.weight": "model-00022-of-00024.safetensors", "model.layers.124.mlp.gate_proj.weight": "model-00022-of-00024.safetensors", "model.layers.124.mlp.up_proj.weight": "model-00023-of-00024.safetensors", "model.layers.124.post_attention_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.124.self_attn.o_proj.weight": "model-00023-of-00024.safetensors", "model.layers.124.self_attn.v_proj.weight": "model-00023-of-00024.safetensors", "model.layers.124.self_attn.k_proj.weight": "model-00023-of-00024.safetensors", "model.layers.124.self_attn.q_proj.weight": "model-00023-of-00024.safetensors", "model.layers.124.input_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.129.mlp.gate_proj.weight": "model-00023-of-00024.safetensors", "model.layers.129.self_attn.o_proj.weight": "model-00023-of-00024.safetensors", "model.layers.129.self_attn.v_proj.weight": "model-00023-of-00024.safetensors", "model.layers.129.self_attn.k_proj.weight": "model-00023-of-00024.safetensors", "model.layers.129.self_attn.q_proj.weight": "model-00023-of-00024.safetensors", "model.layers.128.mlp.down_proj.weight": "model-00023-of-00024.safetensors", "model.layers.128.mlp.gate_proj.weight": "model-00023-of-00024.safetensors", "model.layers.128.mlp.up_proj.weight": "model-00023-of-00024.safetensors", "model.layers.128.post_attention_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.128.self_attn.o_proj.weight": "model-00023-of-00024.safetensors", "model.layers.128.self_attn.v_proj.weight": "model-00023-of-00024.safetensors", "model.layers.128.self_attn.k_proj.weight": "model-00023-of-00024.safetensors", "model.layers.128.self_attn.q_proj.weight": "model-00023-of-00024.safetensors", "model.layers.128.input_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.127.mlp.down_proj.weight": "model-00023-of-00024.safetensors", "model.layers.127.mlp.gate_proj.weight": "model-00023-of-00024.safetensors", "model.layers.127.mlp.up_proj.weight": "model-00023-of-00024.safetensors", "model.layers.127.post_attention_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.127.self_attn.o_proj.weight": "model-00023-of-00024.safetensors", "model.layers.127.self_attn.v_proj.weight": "model-00023-of-00024.safetensors", "model.layers.127.self_attn.k_proj.weight": "model-00023-of-00024.safetensors", "model.layers.127.self_attn.q_proj.weight": "model-00023-of-00024.safetensors", "model.layers.127.input_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.126.mlp.down_proj.weight": "model-00023-of-00024.safetensors", "model.layers.126.post_attention_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.126.input_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.132.self_attn.o_proj.weight": "model-00023-of-00024.safetensors", "model.layers.132.self_attn.v_proj.weight": "model-00023-of-00024.safetensors", "model.layers.132.self_attn.k_proj.weight": "model-00023-of-00024.safetensors", "model.layers.132.self_attn.q_proj.weight": "model-00023-of-00024.safetensors", "model.layers.131.mlp.down_proj.weight": "model-00023-of-00024.safetensors", "model.layers.131.mlp.gate_proj.weight": "model-00023-of-00024.safetensors", "model.layers.131.mlp.up_proj.weight": "model-00023-of-00024.safetensors", "model.layers.131.post_attention_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.131.self_attn.o_proj.weight": "model-00023-of-00024.safetensors", "model.layers.131.self_attn.v_proj.weight": "model-00023-of-00024.safetensors", "model.layers.131.self_attn.k_proj.weight": "model-00023-of-00024.safetensors", "model.layers.131.self_attn.q_proj.weight": "model-00023-of-00024.safetensors", "model.layers.131.input_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.130.mlp.down_proj.weight": "model-00023-of-00024.safetensors", "model.layers.130.mlp.gate_proj.weight": "model-00023-of-00024.safetensors", "model.layers.130.mlp.up_proj.weight": "model-00023-of-00024.safetensors", "model.layers.130.post_attention_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.130.self_attn.o_proj.weight": "model-00023-of-00024.safetensors", "model.layers.130.self_attn.v_proj.weight": "model-00023-of-00024.safetensors", "model.layers.130.self_attn.k_proj.weight": "model-00023-of-00024.safetensors", "model.layers.130.self_attn.q_proj.weight": "model-00023-of-00024.safetensors", "model.layers.130.input_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.129.mlp.down_proj.weight": "model-00023-of-00024.safetensors", "model.layers.129.mlp.up_proj.weight": "model-00024-of-00024.safetensors", "model.layers.129.post_attention_layernorm.weight": "model-00024-of-00024.safetensors", "model.layers.129.input_layernorm.weight": "model-00024-of-00024.safetensors", "model.layers.135.self_attn.v_proj.weight": "model-00024-of-00024.safetensors", "model.layers.135.self_attn.k_proj.weight": "model-00024-of-00024.safetensors", "model.layers.135.self_attn.q_proj.weight": "model-00024-of-00024.safetensors", "model.layers.134.mlp.down_proj.weight": "model-00024-of-00024.safetensors", "model.layers.134.mlp.gate_proj.weight": "model-00024-of-00024.safetensors", "model.layers.134.mlp.up_proj.weight": "model-00024-of-00024.safetensors", "model.layers.134.post_attention_layernorm.weight": "model-00024-of-00024.safetensors", "model.layers.134.self_attn.o_proj.weight": "model-00024-of-00024.safetensors", "model.layers.134.self_attn.v_proj.weight": "model-00024-of-00024.safetensors", "model.layers.134.self_attn.k_proj.weight": "model-00024-of-00024.safetensors", "model.layers.134.self_attn.q_proj.weight": "model-00024-of-00024.safetensors", "model.layers.134.input_layernorm.weight": "model-00024-of-00024.safetensors", "model.layers.133.mlp.down_proj.weight": "model-00024-of-00024.safetensors", "model.layers.133.mlp.gate_proj.weight": "model-00024-of-00024.safetensors", "model.layers.133.mlp.up_proj.weight": "model-00024-of-00024.safetensors", "model.layers.133.post_attention_layernorm.weight": "model-00024-of-00024.safetensors", "model.layers.133.self_attn.o_proj.weight": "model-00024-of-00024.safetensors", "model.layers.133.self_attn.v_proj.weight": "model-00024-of-00024.safetensors", "model.layers.133.self_attn.k_proj.weight": "model-00024-of-00024.safetensors", "model.layers.133.self_attn.q_proj.weight": "model-00024-of-00024.safetensors", "model.layers.133.input_layernorm.weight": "model-00024-of-00024.safetensors", "model.layers.132.mlp.down_proj.weight": "model-00024-of-00024.safetensors", "model.layers.132.mlp.gate_proj.weight": "model-00024-of-00024.safetensors", "model.layers.132.mlp.up_proj.weight": "model-00024-of-00024.safetensors", "model.layers.132.post_attention_layernorm.weight": "model-00024-of-00024.safetensors", "model.layers.132.input_layernorm.weight": "model-00024-of-00024.safetensors", "lm_head.weight": "model-00024-of-00024.safetensors", "model.norm.weight": "model-00024-of-00024.safetensors", "model.layers.136.mlp.down_proj.weight": "model-00024-of-00024.safetensors", "model.layers.136.mlp.gate_proj.weight": "model-00024-of-00024.safetensors", "model.layers.136.mlp.up_proj.weight": "model-00024-of-00024.safetensors", "model.layers.136.post_attention_layernorm.weight": "model-00024-of-00024.safetensors", "model.layers.136.self_attn.o_proj.weight": "model-00024-of-00024.safetensors", "model.layers.136.self_attn.v_proj.weight": "model-00024-of-00024.safetensors", "model.layers.136.self_attn.k_proj.weight": "model-00024-of-00024.safetensors", "model.layers.136.self_attn.q_proj.weight": "model-00024-of-00024.safetensors", "model.layers.136.input_layernorm.weight": "model-00024-of-00024.safetensors", "model.layers.135.mlp.down_proj.weight": "model-00024-of-00024.safetensors", "model.layers.135.mlp.gate_proj.weight": "model-00024-of-00024.safetensors", "model.layers.135.mlp.up_proj.weight": "model-00024-of-00024.safetensors", "model.layers.135.post_attention_layernorm.weight": "model-00024-of-00024.safetensors", "model.layers.135.self_attn.o_proj.weight": "model-00024-of-00024.safetensors", "model.layers.135.input_layernorm.weight": "model-00024-of-00024.safetensors"}}
|
special_tokens_map.json
ADDED
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": {
|
3 |
+
"content": "<s>",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": false,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"eos_token": {
|
10 |
+
"content": "</s>",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"unk_token": {
|
17 |
+
"content": "<unk>",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
}
|
23 |
+
}
|
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": false,
|
9 |
+
"rstrip": false,
|
10 |
+
"single_word": false,
|
11 |
+
"special": true
|
12 |
+
},
|
13 |
+
"1": {
|
14 |
+
"content": "<s>",
|
15 |
+
"lstrip": false,
|
16 |
+
"normalized": false,
|
17 |
+
"rstrip": false,
|
18 |
+
"single_word": false,
|
19 |
+
"special": true
|
20 |
+
},
|
21 |
+
"2": {
|
22 |
+
"content": "</s>",
|
23 |
+
"lstrip": false,
|
24 |
+
"normalized": false,
|
25 |
+
"rstrip": false,
|
26 |
+
"single_word": false,
|
27 |
+
"special": true
|
28 |
+
}
|
29 |
+
},
|
30 |
+
"bos_token": "<s>",
|
31 |
+
"clean_up_tokenization_spaces": false,
|
32 |
+
"eos_token": "</s>",
|
33 |
+
"legacy": false,
|
34 |
+
"model_max_length": 1000000000000000019884624838656,
|
35 |
+
"pad_token": null,
|
36 |
+
"padding_side": "right",
|
37 |
+
"sp_model_kwargs": {},
|
38 |
+
"spaces_between_special_tokens": false,
|
39 |
+
"tokenizer_class": "LlamaTokenizer",
|
40 |
+
"unk_token": "<unk>",
|
41 |
+
"use_default_system_prompt": true
|
42 |
+
}
|