Hugging Face's logo

moxin-org
/
moxin-chat-7b

PyTorch
GGUF
mistral
Inference Endpoints
conversational
Model card Files Community
3
piuzha commited on
Commit
a14b4fb
•
0 Parent(s):

upload

Browse files
Files changed (26) hide show
  1. .gitattributes +36 -0
  2. README.md +103 -0
  3. config.json +25 -0
  4. generation_config.json +6 -0
  5. moxin-chat-7b.gguf +3 -0
  6. pytorch_model-00001.bin +3 -0
  7. pytorch_model-00002.bin +3 -0
  8. pytorch_model-00003.bin +3 -0
  9. pytorch_model-00004.bin +3 -0
  10. pytorch_model-00005.bin +3 -0
  11. pytorch_model-00006.bin +3 -0
  12. pytorch_model-00007.bin +3 -0
  13. pytorch_model-00008.bin +3 -0
  14. pytorch_model-00009.bin +3 -0
  15. pytorch_model-00010.bin +3 -0
  16. pytorch_model-00011.bin +3 -0
  17. pytorch_model-00012.bin +3 -0
  18. pytorch_model-00013.bin +3 -0
  19. pytorch_model-00014.bin +3 -0
  20. pytorch_model-00015.bin +3 -0
  21. pytorch_model-00016.bin +3 -0
  22. pytorch_model.bin.index.json +334 -0
  23. special_tokens_map.json +23 -0
  24. tokenizer.json +0 -0
  25. tokenizer.model +3 -0
  26. tokenizer_config.json +44 -0
.gitattributes ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ *.7z filter=lfs diff=lfs merge=lfs -text
2
+ *.arrow filter=lfs diff=lfs merge=lfs -text
3
+ *.bin filter=lfs diff=lfs merge=lfs -text
4
+ *.bz2 filter=lfs diff=lfs merge=lfs -text
5
+ *.ckpt filter=lfs diff=lfs merge=lfs -text
6
+ *.ftz filter=lfs diff=lfs merge=lfs -text
7
+ *.gz filter=lfs diff=lfs merge=lfs -text
8
+ *.h5 filter=lfs diff=lfs merge=lfs -text
9
+ *.joblib filter=lfs diff=lfs merge=lfs -text
10
+ *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
+ *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
+ *.model filter=lfs diff=lfs merge=lfs -text
13
+ *.msgpack filter=lfs diff=lfs merge=lfs -text
14
+ *.npy filter=lfs diff=lfs merge=lfs -text
15
+ *.npz filter=lfs diff=lfs merge=lfs -text
16
+ *.onnx filter=lfs diff=lfs merge=lfs -text
17
+ *.ot filter=lfs diff=lfs merge=lfs -text
18
+ *.parquet filter=lfs diff=lfs merge=lfs -text
19
+ *.pb filter=lfs diff=lfs merge=lfs -text
20
+ *.pickle filter=lfs diff=lfs merge=lfs -text
21
+ *.pkl filter=lfs diff=lfs merge=lfs -text
22
+ *.pt filter=lfs diff=lfs merge=lfs -text
23
+ *.pth filter=lfs diff=lfs merge=lfs -text
24
+ *.rar filter=lfs diff=lfs merge=lfs -text
25
+ *.safetensors filter=lfs diff=lfs merge=lfs -text
26
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
+ *.tar.* filter=lfs diff=lfs merge=lfs -text
28
+ *.tar filter=lfs diff=lfs merge=lfs -text
29
+ *.tflite filter=lfs diff=lfs merge=lfs -text
30
+ *.tgz filter=lfs diff=lfs merge=lfs -text
31
+ *.wasm filter=lfs diff=lfs merge=lfs -text
32
+ *.xz filter=lfs diff=lfs merge=lfs -text
33
+ *.zip filter=lfs diff=lfs merge=lfs -text
34
+ *.zst filter=lfs diff=lfs merge=lfs -text
35
+ *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ *.gguf filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,103 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ ---
4
+
5
+
6
+ <h1 align="center"> Moxin Chat 7B </h1>
7
+
8
+ <p align="center"> <a href="https://github.com/moxin-org/Moxin-LLM">Home Page</a> &nbsp&nbsp | &nbsp&nbsp <a href="https://github.com/moxin-org/Moxin-LLM/blob/main/report/Moxin_7B.pdf">Technical Report</a> &nbsp&nbsp | &nbsp&nbsp <a href="https://huggingface.co/moxin-org/moxin-llm-7b">Base Model</a> &nbsp&nbsp | &nbsp&nbsp <a href="https://huggingface.co/moxin-org/moxin-chat-7b">Chat Model</a> </p>
9
+
10
+
11
+
12
+
13
+ ## Model
14
+ You can download our base 7B model from this [link](https://huggingface.co/moxin-org/moxin-llm-7b) and our chat 7B model from this [link](https://huggingface.co/moxin-org/moxin-chat-7b).
15
+
16
+
17
+
18
+ ## Inference
19
+
20
+ You can use the following code to run inference with the model. The model is saved under './model/' directory. Change the model directory accordingly or use the Huggingface link.
21
+
22
+ ```
23
+ import torch
24
+ from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
25
+
26
+ torch.backends.cuda.enable_mem_efficient_sdp(False)
27
+ torch.backends.cuda.enable_flash_sdp(False)
28
+
29
+ model_name = 'moxin-org/moxin-chat-7b'
30
+ tokenizer = AutoTokenizer.from_pretrained(model_name)
31
+ model = AutoModelForCausalLM.from_pretrained(
32
+ model_name,
33
+ torch_dtype=torch.bfloat16,
34
+ device_map="auto",
35
+ trust_remote_code=True,
36
+ )
37
+
38
+ pipe = pipeline(
39
+ "text-generation",
40
+ model=model,
41
+ tokenizer = tokenizer,
42
+ torch_dtype=torch.bfloat16,
43
+ device_map="auto"
44
+ )
45
+
46
+ prompt = "Can you explain the concept of regularization in machine learning?"
47
+
48
+ sequences = pipe(
49
+ prompt,
50
+ do_sample=True,
51
+ max_new_tokens=1000,
52
+ temperature=0.7,
53
+ top_k=50,
54
+ top_p=0.95,
55
+ num_return_sequences=1,
56
+ )
57
+ print(sequences[0]['generated_text'])
58
+ ```
59
+
60
+ ## Chat template
61
+
62
+
63
+ ## Evaluation
64
+
65
+ We test the performance of our model with [lm-evaluation-harness](https://github.com/EleutherAI/lm-evaluation-harness). The evaluation results on common datasets are shown below. We test on AI2 Reasoning Challenge (25-shot), HellaSwag (10-shot), MMLU (5-shot), and Winogrande (5-shot).
66
+
67
+ | Models | ARC-C | Hellaswag | MMLU | WinoGrade | Ave |
68
+ |:----------------------:|:-----:|:---------:|:-----:|:---------:|:-----:|
69
+ | Mistral-7B | 57.59 | 83.25 | 62.42 | 78.77 | 70.51 |
70
+ | LLaMA 3.1-8B | 54.61 | 81.95 | 65.16 | 77.35 | 69.77 |
71
+ | LLaMA 3-8B | 55.46 | 82.09 | 65.29 | 77.82 | 70.17 |
72
+ | LLaMA 2-7B | 49.74 | 78.94 | 45.89 | 74.27 | 62.21 |
73
+ | Qwen 2-7B | 57.68 | 80.76 | 70.42 | 77.43 | 71.57 |
74
+ | gemma-7b | 56.48 | 82.31 | 63.02 | 78.3 | 70.03 |
75
+ | internlm2.5-7b | 54.78 | 79.7 | 68.17 | 80.9 | 70.89 |
76
+ | Baichuan2-7B | 47.87 | 73.89 | 54.13 | 70.8 | 61.67 |
77
+ | Yi-1.5-9B | 58.36 | 80.36 | 69.54 | 77.53 | 71.48 |
78
+ | Moxin-7B-original | 53.75 | 75.46 | 59.43 | 70.32 | 64.74 |
79
+ | Moxin-7B-finetuned | 59.47 | 83.08 | 60.97 | 78.69 | 70.55 |
80
+
81
+
82
+ We also test the zero shot performance on AI2 Reasoning Challenge (0-shot), AI2 Reasoning Easy (0-shot), HellaSwag (0-shot), PIQA (0-shot) and Winogrande (0-shot). The results are shown below.
83
+
84
+ | Models | HellaSwag | WinoGrade | PIQA | ARC-E | ARC-C | Ave |
85
+ |:-----------------:|:---------:|:---------:|:-----:|:-----:|:-----:|:-----:|
86
+ | Mistral-7B | 80.39 | 73.4 | 82.15 | 78.28 | 52.22 | 73.29 |
87
+ | LLaMA 2-7B | 75.99 | 69.06 | 79.11 | 74.54 | 46.42 | 69.02 |
88
+ | LLaMA 2-13B | 79.37 | 72.22 | 80.52 | 77.4 | 49.06 | 71.71 |
89
+ | LLaMA 3.1-8B | 78.92 | 74.19 | 81.12 | 81.06 | 53.67 | 73.79 |
90
+ | gemma-7b | 80.45 | 73.72 | 80.9 | 79.97 | 54.1 | 73.83 |
91
+ | Qwen v2-7B | 78.9 | 72.38 | 79.98 | 74.71 | 50.09 | 71.21 |
92
+ | internlm2.5-7b | 79.14 | 77.9 | 80.52 | 76.16 | 51.37 | 73.02 |
93
+ | Baichuan2-7B | 72.25 | 67.17 | 77.26 | 72.98 | 42.15 | 66.36 |
94
+ | Yi-1.5-9B | 77.86 | 73.01 | 80.74 | 79.04 | 55.03 | 73.14 |
95
+ | deepseek-7b | 76.13 | 69.77 | 79.76 | 71.04 | 44.8 | 68.3 |
96
+ | Moxin-7B-original | 72.06 | 66.31 | 78.07 | 71.47 | 48.15 | 67.21 |
97
+ | Moxin-7B-finetune | 80.03 | 75.17 | 82.24 | 81.12 | 58.64 | 75.44 |
98
+
99
+
100
+
101
+
102
+
103
+
config.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Moxin-chat",
3
+ "architectures": [
4
+ "MistralForCausalLM"
5
+ ],
6
+ "bos_token_id": 1,
7
+ "eos_token_id": 2,
8
+ "hidden_act": "silu",
9
+ "hidden_size": 4096,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 14336,
12
+ "max_position_embeddings": 32768,
13
+ "model_type": "mistral",
14
+ "num_attention_heads": 32,
15
+ "num_hidden_layers": 36,
16
+ "num_key_value_heads": 8,
17
+ "rms_norm_eps": 1e-05,
18
+ "rope_theta": 10000.0,
19
+ "sliding_window": 4096,
20
+ "tie_word_embeddings": false,
21
+ "torch_dtype": "float16",
22
+ "transformers_version": "4.34.0",
23
+ "use_cache": true,
24
+ "vocab_size": 32000
25
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "transformers_version": "4.34.0"
6
+ }
moxin-chat-7b.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8129919414b519683cd4ebc140ccd19997afb80144960b9304d3c6a15716d163
3
+ size 16229696704
pytorch_model-00001.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f1f5702c04992ac171e4e7cd0576099f359888163cd6db2d3f38ceea7ece6897
3
+ size 1017140493
pytorch_model-00002.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e3af6540c7b3281da70181a8d090eae1e274791b08be0f9d39cf93878073f86
3
+ size 1040244373
pytorch_model-00003.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e6db8a91458c8da3d960e98bd5e43175ec4a4422e68cf75c1bc3f97d126bfb9
3
+ size 1023449521
pytorch_model-00004.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e85b29c00c0a564a7019ea74eed40cdef20d4573b5edea2b8996c68c9a3f5f91
3
+ size 989894824
pytorch_model-00005.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c1b2ef96bdc075e68f770271b0c2096d80bef120e37a306ba9c07fc2e508a73
3
+ size 1040244373
pytorch_model-00006.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2b16f2d358f2b518996625dcbef50447f5b4cb8d8e5cdbc839aae12eb4f2b7e1
3
+ size 1023449585
pytorch_model-00007.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:340b6ba2d2be2daf6f9b9d64803445b1dc69825a1afe2d7ff8bc360ac01fbc7f
3
+ size 989894824
pytorch_model-00008.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d87fdccfc33d1ab01cc3ae40afbc224b786acea40d63ca86c09a0b3b2012a50c
3
+ size 1040244373
pytorch_model-00009.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5207b89fb981e402f3093d57b9a8b2e9ae42b97c0824066e29c82bf53db4af8d
3
+ size 1023449585
pytorch_model-00010.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a5400c092714924f1e7ec332189da56e8853f3be0f4faf9cdd418ca1e114f952
3
+ size 989894824
pytorch_model-00011.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b738633863df5d75ea6cc9a4443c8fe456360e84900377168565aff427003dbc
3
+ size 1040244373
pytorch_model-00012.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ee1ec9e0aba30597659255172fa7fa8425913314448c91246bb073cac900f9a
3
+ size 1023449585
pytorch_model-00013.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:52e37238ca11b88f9a031dc63449e1e6dafa46264a2c4cffff52dbb4a79fc388
3
+ size 989894824
pytorch_model-00014.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d60738c69a8bcf9362f794ba982bbfaaaee02e06fd444beb82fe89dff14c2aa3
3
+ size 1040244373
pytorch_model-00015.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3f8c20b21075b3d6e11424a0cb014793aa662a30f93ad1d3d12f80dda7ba670
3
+ size 1023449585
pytorch_model-00016.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4aad5422a6cb5efe3a60790275d16c0041781eabaa73e4392084f827cf1737cc
3
+ size 933278660
pytorch_model.bin.index.json ADDED
@@ -0,0 +1,334 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 15476.5703125
4
+ },
5
+ "weight_map": {
6
+ "model.embed_tokens.weight": "pytorch_model-00001.bin",
7
+ "model.layers.0.self_attn.q_proj.weight": "pytorch_model-00001.bin",
8
+ "model.layers.0.self_attn.k_proj.weight": "pytorch_model-00001.bin",
9
+ "model.layers.0.self_attn.v_proj.weight": "pytorch_model-00001.bin",
10
+ "model.layers.0.self_attn.o_proj.weight": "pytorch_model-00001.bin",
11
+ "model.layers.0.mlp.gate_proj.weight": "pytorch_model-00001.bin",
12
+ "model.layers.0.mlp.up_proj.weight": "pytorch_model-00001.bin",
13
+ "model.layers.0.mlp.down_proj.weight": "pytorch_model-00001.bin",
14
+ "model.layers.0.input_layernorm.weight": "pytorch_model-00001.bin",
15
+ "model.layers.0.post_attention_layernorm.weight": "pytorch_model-00001.bin",
16
+ "model.layers.1.self_attn.q_proj.weight": "pytorch_model-00001.bin",
17
+ "model.layers.1.self_attn.k_proj.weight": "pytorch_model-00001.bin",
18
+ "model.layers.1.self_attn.v_proj.weight": "pytorch_model-00001.bin",
19
+ "model.layers.1.self_attn.o_proj.weight": "pytorch_model-00001.bin",
20
+ "model.layers.1.mlp.gate_proj.weight": "pytorch_model-00001.bin",
21
+ "model.layers.1.mlp.up_proj.weight": "pytorch_model-00001.bin",
22
+ "model.layers.1.mlp.down_proj.weight": "pytorch_model-00002.bin",
23
+ "model.layers.1.input_layernorm.weight": "pytorch_model-00002.bin",
24
+ "model.layers.1.post_attention_layernorm.weight": "pytorch_model-00002.bin",
25
+ "model.layers.2.self_attn.q_proj.weight": "pytorch_model-00002.bin",
26
+ "model.layers.2.self_attn.k_proj.weight": "pytorch_model-00002.bin",
27
+ "model.layers.2.self_attn.v_proj.weight": "pytorch_model-00002.bin",
28
+ "model.layers.2.self_attn.o_proj.weight": "pytorch_model-00002.bin",
29
+ "model.layers.2.mlp.gate_proj.weight": "pytorch_model-00002.bin",
30
+ "model.layers.2.mlp.up_proj.weight": "pytorch_model-00002.bin",
31
+ "model.layers.2.mlp.down_proj.weight": "pytorch_model-00002.bin",
32
+ "model.layers.2.input_layernorm.weight": "pytorch_model-00002.bin",
33
+ "model.layers.2.post_attention_layernorm.weight": "pytorch_model-00002.bin",
34
+ "model.layers.3.self_attn.q_proj.weight": "pytorch_model-00002.bin",
35
+ "model.layers.3.self_attn.k_proj.weight": "pytorch_model-00002.bin",
36
+ "model.layers.3.self_attn.v_proj.weight": "pytorch_model-00002.bin",
37
+ "model.layers.3.self_attn.o_proj.weight": "pytorch_model-00002.bin",
38
+ "model.layers.3.mlp.gate_proj.weight": "pytorch_model-00002.bin",
39
+ "model.layers.3.mlp.up_proj.weight": "pytorch_model-00002.bin",
40
+ "model.layers.3.mlp.down_proj.weight": "pytorch_model-00002.bin",
41
+ "model.layers.3.input_layernorm.weight": "pytorch_model-00002.bin",
42
+ "model.layers.3.post_attention_layernorm.weight": "pytorch_model-00002.bin",
43
+ "model.layers.4.self_attn.q_proj.weight": "pytorch_model-00002.bin",
44
+ "model.layers.4.self_attn.k_proj.weight": "pytorch_model-00002.bin",
45
+ "model.layers.4.self_attn.v_proj.weight": "pytorch_model-00002.bin",
46
+ "model.layers.4.self_attn.o_proj.weight": "pytorch_model-00003.bin",
47
+ "model.layers.4.mlp.gate_proj.weight": "pytorch_model-00003.bin",
48
+ "model.layers.4.mlp.up_proj.weight": "pytorch_model-00003.bin",
49
+ "model.layers.4.mlp.down_proj.weight": "pytorch_model-00003.bin",
50
+ "model.layers.4.input_layernorm.weight": "pytorch_model-00003.bin",
51
+ "model.layers.4.post_attention_layernorm.weight": "pytorch_model-00003.bin",
52
+ "model.layers.5.self_attn.q_proj.weight": "pytorch_model-00003.bin",
53
+ "model.layers.5.self_attn.k_proj.weight": "pytorch_model-00003.bin",
54
+ "model.layers.5.self_attn.v_proj.weight": "pytorch_model-00003.bin",
55
+ "model.layers.5.self_attn.o_proj.weight": "pytorch_model-00003.bin",
56
+ "model.layers.5.mlp.gate_proj.weight": "pytorch_model-00003.bin",
57
+ "model.layers.5.mlp.up_proj.weight": "pytorch_model-00003.bin",
58
+ "model.layers.5.mlp.down_proj.weight": "pytorch_model-00003.bin",
59
+ "model.layers.5.input_layernorm.weight": "pytorch_model-00003.bin",
60
+ "model.layers.5.post_attention_layernorm.weight": "pytorch_model-00003.bin",
61
+ "model.layers.6.self_attn.q_proj.weight": "pytorch_model-00003.bin",
62
+ "model.layers.6.self_attn.k_proj.weight": "pytorch_model-00003.bin",
63
+ "model.layers.6.self_attn.v_proj.weight": "pytorch_model-00003.bin",
64
+ "model.layers.6.self_attn.o_proj.weight": "pytorch_model-00003.bin",
65
+ "model.layers.6.mlp.gate_proj.weight": "pytorch_model-00003.bin",
66
+ "model.layers.6.mlp.up_proj.weight": "pytorch_model-00004.bin",
67
+ "model.layers.6.mlp.down_proj.weight": "pytorch_model-00004.bin",
68
+ "model.layers.6.input_layernorm.weight": "pytorch_model-00004.bin",
69
+ "model.layers.6.post_attention_layernorm.weight": "pytorch_model-00004.bin",
70
+ "model.layers.7.self_attn.q_proj.weight": "pytorch_model-00004.bin",
71
+ "model.layers.7.self_attn.k_proj.weight": "pytorch_model-00004.bin",
72
+ "model.layers.7.self_attn.v_proj.weight": "pytorch_model-00004.bin",
73
+ "model.layers.7.self_attn.o_proj.weight": "pytorch_model-00004.bin",
74
+ "model.layers.7.mlp.gate_proj.weight": "pytorch_model-00004.bin",
75
+ "model.layers.7.mlp.up_proj.weight": "pytorch_model-00004.bin",
76
+ "model.layers.7.mlp.down_proj.weight": "pytorch_model-00004.bin",
77
+ "model.layers.7.input_layernorm.weight": "pytorch_model-00004.bin",
78
+ "model.layers.7.post_attention_layernorm.weight": "pytorch_model-00004.bin",
79
+ "model.layers.8.self_attn.q_proj.weight": "pytorch_model-00004.bin",
80
+ "model.layers.8.self_attn.k_proj.weight": "pytorch_model-00004.bin",
81
+ "model.layers.8.self_attn.v_proj.weight": "pytorch_model-00004.bin",
82
+ "model.layers.8.self_attn.o_proj.weight": "pytorch_model-00004.bin",
83
+ "model.layers.8.mlp.gate_proj.weight": "pytorch_model-00004.bin",
84
+ "model.layers.8.mlp.up_proj.weight": "pytorch_model-00004.bin",
85
+ "model.layers.8.mlp.down_proj.weight": "pytorch_model-00005.bin",
86
+ "model.layers.8.input_layernorm.weight": "pytorch_model-00005.bin",
87
+ "model.layers.8.post_attention_layernorm.weight": "pytorch_model-00005.bin",
88
+ "model.layers.9.self_attn.q_proj.weight": "pytorch_model-00005.bin",
89
+ "model.layers.9.self_attn.k_proj.weight": "pytorch_model-00005.bin",
90
+ "model.layers.9.self_attn.v_proj.weight": "pytorch_model-00005.bin",
91
+ "model.layers.9.self_attn.o_proj.weight": "pytorch_model-00005.bin",
92
+ "model.layers.9.mlp.gate_proj.weight": "pytorch_model-00005.bin",
93
+ "model.layers.9.mlp.up_proj.weight": "pytorch_model-00005.bin",
94
+ "model.layers.9.mlp.down_proj.weight": "pytorch_model-00005.bin",
95
+ "model.layers.9.input_layernorm.weight": "pytorch_model-00005.bin",
96
+ "model.layers.9.post_attention_layernorm.weight": "pytorch_model-00005.bin",
97
+ "model.layers.10.self_attn.q_proj.weight": "pytorch_model-00005.bin",
98
+ "model.layers.10.self_attn.k_proj.weight": "pytorch_model-00005.bin",
99
+ "model.layers.10.self_attn.v_proj.weight": "pytorch_model-00005.bin",
100
+ "model.layers.10.self_attn.o_proj.weight": "pytorch_model-00005.bin",
101
+ "model.layers.10.mlp.gate_proj.weight": "pytorch_model-00005.bin",
102
+ "model.layers.10.mlp.up_proj.weight": "pytorch_model-00005.bin",
103
+ "model.layers.10.mlp.down_proj.weight": "pytorch_model-00005.bin",
104
+ "model.layers.10.input_layernorm.weight": "pytorch_model-00005.bin",
105
+ "model.layers.10.post_attention_layernorm.weight": "pytorch_model-00005.bin",
106
+ "model.layers.11.self_attn.q_proj.weight": "pytorch_model-00005.bin",
107
+ "model.layers.11.self_attn.k_proj.weight": "pytorch_model-00005.bin",
108
+ "model.layers.11.self_attn.v_proj.weight": "pytorch_model-00005.bin",
109
+ "model.layers.11.self_attn.o_proj.weight": "pytorch_model-00006.bin",
110
+ "model.layers.11.mlp.gate_proj.weight": "pytorch_model-00006.bin",
111
+ "model.layers.11.mlp.up_proj.weight": "pytorch_model-00006.bin",
112
+ "model.layers.11.mlp.down_proj.weight": "pytorch_model-00006.bin",
113
+ "model.layers.11.input_layernorm.weight": "pytorch_model-00006.bin",
114
+ "model.layers.11.post_attention_layernorm.weight": "pytorch_model-00006.bin",
115
+ "model.layers.12.self_attn.q_proj.weight": "pytorch_model-00006.bin",
116
+ "model.layers.12.self_attn.k_proj.weight": "pytorch_model-00006.bin",
117
+ "model.layers.12.self_attn.v_proj.weight": "pytorch_model-00006.bin",
118
+ "model.layers.12.self_attn.o_proj.weight": "pytorch_model-00006.bin",
119
+ "model.layers.12.mlp.gate_proj.weight": "pytorch_model-00006.bin",
120
+ "model.layers.12.mlp.up_proj.weight": "pytorch_model-00006.bin",
121
+ "model.layers.12.mlp.down_proj.weight": "pytorch_model-00006.bin",
122
+ "model.layers.12.input_layernorm.weight": "pytorch_model-00006.bin",
123
+ "model.layers.12.post_attention_layernorm.weight": "pytorch_model-00006.bin",
124
+ "model.layers.13.self_attn.q_proj.weight": "pytorch_model-00006.bin",
125
+ "model.layers.13.self_attn.k_proj.weight": "pytorch_model-00006.bin",
126
+ "model.layers.13.self_attn.v_proj.weight": "pytorch_model-00006.bin",
127
+ "model.layers.13.self_attn.o_proj.weight": "pytorch_model-00006.bin",
128
+ "model.layers.13.mlp.gate_proj.weight": "pytorch_model-00006.bin",
129
+ "model.layers.13.mlp.up_proj.weight": "pytorch_model-00007.bin",
130
+ "model.layers.13.mlp.down_proj.weight": "pytorch_model-00007.bin",
131
+ "model.layers.13.input_layernorm.weight": "pytorch_model-00007.bin",
132
+ "model.layers.13.post_attention_layernorm.weight": "pytorch_model-00007.bin",
133
+ "model.layers.14.self_attn.q_proj.weight": "pytorch_model-00007.bin",
134
+ "model.layers.14.self_attn.k_proj.weight": "pytorch_model-00007.bin",
135
+ "model.layers.14.self_attn.v_proj.weight": "pytorch_model-00007.bin",
136
+ "model.layers.14.self_attn.o_proj.weight": "pytorch_model-00007.bin",
137
+ "model.layers.14.mlp.gate_proj.weight": "pytorch_model-00007.bin",
138
+ "model.layers.14.mlp.up_proj.weight": "pytorch_model-00007.bin",
139
+ "model.layers.14.mlp.down_proj.weight": "pytorch_model-00007.bin",
140
+ "model.layers.14.input_layernorm.weight": "pytorch_model-00007.bin",
141
+ "model.layers.14.post_attention_layernorm.weight": "pytorch_model-00007.bin",
142
+ "model.layers.15.self_attn.q_proj.weight": "pytorch_model-00007.bin",
143
+ "model.layers.15.self_attn.k_proj.weight": "pytorch_model-00007.bin",
144
+ "model.layers.15.self_attn.v_proj.weight": "pytorch_model-00007.bin",
145
+ "model.layers.15.self_attn.o_proj.weight": "pytorch_model-00007.bin",
146
+ "model.layers.15.mlp.gate_proj.weight": "pytorch_model-00007.bin",
147
+ "model.layers.15.mlp.up_proj.weight": "pytorch_model-00007.bin",
148
+ "model.layers.15.mlp.down_proj.weight": "pytorch_model-00008.bin",
149
+ "model.layers.15.input_layernorm.weight": "pytorch_model-00008.bin",
150
+ "model.layers.15.post_attention_layernorm.weight": "pytorch_model-00008.bin",
151
+ "model.layers.16.self_attn.q_proj.weight": "pytorch_model-00008.bin",
152
+ "model.layers.16.self_attn.k_proj.weight": "pytorch_model-00008.bin",
153
+ "model.layers.16.self_attn.v_proj.weight": "pytorch_model-00008.bin",
154
+ "model.layers.16.self_attn.o_proj.weight": "pytorch_model-00008.bin",
155
+ "model.layers.16.mlp.gate_proj.weight": "pytorch_model-00008.bin",
156
+ "model.layers.16.mlp.up_proj.weight": "pytorch_model-00008.bin",
157
+ "model.layers.16.mlp.down_proj.weight": "pytorch_model-00008.bin",
158
+ "model.layers.16.input_layernorm.weight": "pytorch_model-00008.bin",
159
+ "model.layers.16.post_attention_layernorm.weight": "pytorch_model-00008.bin",
160
+ "model.layers.17.self_attn.q_proj.weight": "pytorch_model-00008.bin",
161
+ "model.layers.17.self_attn.k_proj.weight": "pytorch_model-00008.bin",
162
+ "model.layers.17.self_attn.v_proj.weight": "pytorch_model-00008.bin",
163
+ "model.layers.17.self_attn.o_proj.weight": "pytorch_model-00008.bin",
164
+ "model.layers.17.mlp.gate_proj.weight": "pytorch_model-00008.bin",
165
+ "model.layers.17.mlp.up_proj.weight": "pytorch_model-00008.bin",
166
+ "model.layers.17.mlp.down_proj.weight": "pytorch_model-00008.bin",
167
+ "model.layers.17.input_layernorm.weight": "pytorch_model-00008.bin",
168
+ "model.layers.17.post_attention_layernorm.weight": "pytorch_model-00008.bin",
169
+ "model.layers.18.self_attn.q_proj.weight": "pytorch_model-00008.bin",
170
+ "model.layers.18.self_attn.k_proj.weight": "pytorch_model-00008.bin",
171
+ "model.layers.18.self_attn.v_proj.weight": "pytorch_model-00008.bin",
172
+ "model.layers.18.self_attn.o_proj.weight": "pytorch_model-00009.bin",
173
+ "model.layers.18.mlp.gate_proj.weight": "pytorch_model-00009.bin",
174
+ "model.layers.18.mlp.up_proj.weight": "pytorch_model-00009.bin",
175
+ "model.layers.18.mlp.down_proj.weight": "pytorch_model-00009.bin",
176
+ "model.layers.18.input_layernorm.weight": "pytorch_model-00009.bin",
177
+ "model.layers.18.post_attention_layernorm.weight": "pytorch_model-00009.bin",
178
+ "model.layers.19.self_attn.q_proj.weight": "pytorch_model-00009.bin",
179
+ "model.layers.19.self_attn.k_proj.weight": "pytorch_model-00009.bin",
180
+ "model.layers.19.self_attn.v_proj.weight": "pytorch_model-00009.bin",
181
+ "model.layers.19.self_attn.o_proj.weight": "pytorch_model-00009.bin",
182
+ "model.layers.19.mlp.gate_proj.weight": "pytorch_model-00009.bin",
183
+ "model.layers.19.mlp.up_proj.weight": "pytorch_model-00009.bin",
184
+ "model.layers.19.mlp.down_proj.weight": "pytorch_model-00009.bin",
185
+ "model.layers.19.input_layernorm.weight": "pytorch_model-00009.bin",
186
+ "model.layers.19.post_attention_layernorm.weight": "pytorch_model-00009.bin",
187
+ "model.layers.20.self_attn.q_proj.weight": "pytorch_model-00009.bin",
188
+ "model.layers.20.self_attn.k_proj.weight": "pytorch_model-00009.bin",
189
+ "model.layers.20.self_attn.v_proj.weight": "pytorch_model-00009.bin",
190
+ "model.layers.20.self_attn.o_proj.weight": "pytorch_model-00009.bin",
191
+ "model.layers.20.mlp.gate_proj.weight": "pytorch_model-00009.bin",
192
+ "model.layers.20.mlp.up_proj.weight": "pytorch_model-00010.bin",
193
+ "model.layers.20.mlp.down_proj.weight": "pytorch_model-00010.bin",
194
+ "model.layers.20.input_layernorm.weight": "pytorch_model-00010.bin",
195
+ "model.layers.20.post_attention_layernorm.weight": "pytorch_model-00010.bin",
196
+ "model.layers.21.self_attn.q_proj.weight": "pytorch_model-00010.bin",
197
+ "model.layers.21.self_attn.k_proj.weight": "pytorch_model-00010.bin",
198
+ "model.layers.21.self_attn.v_proj.weight": "pytorch_model-00010.bin",
199
+ "model.layers.21.self_attn.o_proj.weight": "pytorch_model-00010.bin",
200
+ "model.layers.21.mlp.gate_proj.weight": "pytorch_model-00010.bin",
201
+ "model.layers.21.mlp.up_proj.weight": "pytorch_model-00010.bin",
202
+ "model.layers.21.mlp.down_proj.weight": "pytorch_model-00010.bin",
203
+ "model.layers.21.input_layernorm.weight": "pytorch_model-00010.bin",
204
+ "model.layers.21.post_attention_layernorm.weight": "pytorch_model-00010.bin",
205
+ "model.layers.22.self_attn.q_proj.weight": "pytorch_model-00010.bin",
206
+ "model.layers.22.self_attn.k_proj.weight": "pytorch_model-00010.bin",
207
+ "model.layers.22.self_attn.v_proj.weight": "pytorch_model-00010.bin",
208
+ "model.layers.22.self_attn.o_proj.weight": "pytorch_model-00010.bin",
209
+ "model.layers.22.mlp.gate_proj.weight": "pytorch_model-00010.bin",
210
+ "model.layers.22.mlp.up_proj.weight": "pytorch_model-00010.bin",
211
+ "model.layers.22.mlp.down_proj.weight": "pytorch_model-00011.bin",
212
+ "model.layers.22.input_layernorm.weight": "pytorch_model-00011.bin",
213
+ "model.layers.22.post_attention_layernorm.weight": "pytorch_model-00011.bin",
214
+ "model.layers.23.self_attn.q_proj.weight": "pytorch_model-00011.bin",
215
+ "model.layers.23.self_attn.k_proj.weight": "pytorch_model-00011.bin",
216
+ "model.layers.23.self_attn.v_proj.weight": "pytorch_model-00011.bin",
217
+ "model.layers.23.self_attn.o_proj.weight": "pytorch_model-00011.bin",
218
+ "model.layers.23.mlp.gate_proj.weight": "pytorch_model-00011.bin",
219
+ "model.layers.23.mlp.up_proj.weight": "pytorch_model-00011.bin",
220
+ "model.layers.23.mlp.down_proj.weight": "pytorch_model-00011.bin",
221
+ "model.layers.23.input_layernorm.weight": "pytorch_model-00011.bin",
222
+ "model.layers.23.post_attention_layernorm.weight": "pytorch_model-00011.bin",
223
+ "model.layers.24.self_attn.q_proj.weight": "pytorch_model-00011.bin",
224
+ "model.layers.24.self_attn.k_proj.weight": "pytorch_model-00011.bin",
225
+ "model.layers.24.self_attn.v_proj.weight": "pytorch_model-00011.bin",
226
+ "model.layers.24.self_attn.o_proj.weight": "pytorch_model-00011.bin",
227
+ "model.layers.24.mlp.gate_proj.weight": "pytorch_model-00011.bin",
228
+ "model.layers.24.mlp.up_proj.weight": "pytorch_model-00011.bin",
229
+ "model.layers.24.mlp.down_proj.weight": "pytorch_model-00011.bin",
230
+ "model.layers.24.input_layernorm.weight": "pytorch_model-00011.bin",
231
+ "model.layers.24.post_attention_layernorm.weight": "pytorch_model-00011.bin",
232
+ "model.layers.25.self_attn.q_proj.weight": "pytorch_model-00011.bin",
233
+ "model.layers.25.self_attn.k_proj.weight": "pytorch_model-00011.bin",
234
+ "model.layers.25.self_attn.v_proj.weight": "pytorch_model-00011.bin",
235
+ "model.layers.25.self_attn.o_proj.weight": "pytorch_model-00012.bin",
236
+ "model.layers.25.mlp.gate_proj.weight": "pytorch_model-00012.bin",
237
+ "model.layers.25.mlp.up_proj.weight": "pytorch_model-00012.bin",
238
+ "model.layers.25.mlp.down_proj.weight": "pytorch_model-00012.bin",
239
+ "model.layers.25.input_layernorm.weight": "pytorch_model-00012.bin",
240
+ "model.layers.25.post_attention_layernorm.weight": "pytorch_model-00012.bin",
241
+ "model.layers.26.self_attn.q_proj.weight": "pytorch_model-00012.bin",
242
+ "model.layers.26.self_attn.k_proj.weight": "pytorch_model-00012.bin",
243
+ "model.layers.26.self_attn.v_proj.weight": "pytorch_model-00012.bin",
244
+ "model.layers.26.self_attn.o_proj.weight": "pytorch_model-00012.bin",
245
+ "model.layers.26.mlp.gate_proj.weight": "pytorch_model-00012.bin",
246
+ "model.layers.26.mlp.up_proj.weight": "pytorch_model-00012.bin",
247
+ "model.layers.26.mlp.down_proj.weight": "pytorch_model-00012.bin",
248
+ "model.layers.26.input_layernorm.weight": "pytorch_model-00012.bin",
249
+ "model.layers.26.post_attention_layernorm.weight": "pytorch_model-00012.bin",
250
+ "model.layers.27.self_attn.q_proj.weight": "pytorch_model-00012.bin",
251
+ "model.layers.27.self_attn.k_proj.weight": "pytorch_model-00012.bin",
252
+ "model.layers.27.self_attn.v_proj.weight": "pytorch_model-00012.bin",
253
+ "model.layers.27.self_attn.o_proj.weight": "pytorch_model-00012.bin",
254
+ "model.layers.27.mlp.gate_proj.weight": "pytorch_model-00012.bin",
255
+ "model.layers.27.mlp.up_proj.weight": "pytorch_model-00013.bin",
256
+ "model.layers.27.mlp.down_proj.weight": "pytorch_model-00013.bin",
257
+ "model.layers.27.input_layernorm.weight": "pytorch_model-00013.bin",
258
+ "model.layers.27.post_attention_layernorm.weight": "pytorch_model-00013.bin",
259
+ "model.layers.28.self_attn.q_proj.weight": "pytorch_model-00013.bin",
260
+ "model.layers.28.self_attn.k_proj.weight": "pytorch_model-00013.bin",
261
+ "model.layers.28.self_attn.v_proj.weight": "pytorch_model-00013.bin",
262
+ "model.layers.28.self_attn.o_proj.weight": "pytorch_model-00013.bin",
263
+ "model.layers.28.mlp.gate_proj.weight": "pytorch_model-00013.bin",
264
+ "model.layers.28.mlp.up_proj.weight": "pytorch_model-00013.bin",
265
+ "model.layers.28.mlp.down_proj.weight": "pytorch_model-00013.bin",
266
+ "model.layers.28.input_layernorm.weight": "pytorch_model-00013.bin",
267
+ "model.layers.28.post_attention_layernorm.weight": "pytorch_model-00013.bin",
268
+ "model.layers.29.self_attn.q_proj.weight": "pytorch_model-00013.bin",
269
+ "model.layers.29.self_attn.k_proj.weight": "pytorch_model-00013.bin",
270
+ "model.layers.29.self_attn.v_proj.weight": "pytorch_model-00013.bin",
271
+ "model.layers.29.self_attn.o_proj.weight": "pytorch_model-00013.bin",
272
+ "model.layers.29.mlp.gate_proj.weight": "pytorch_model-00013.bin",
273
+ "model.layers.29.mlp.up_proj.weight": "pytorch_model-00013.bin",
274
+ "model.layers.29.mlp.down_proj.weight": "pytorch_model-00014.bin",
275
+ "model.layers.29.input_layernorm.weight": "pytorch_model-00014.bin",
276
+ "model.layers.29.post_attention_layernorm.weight": "pytorch_model-00014.bin",
277
+ "model.layers.30.self_attn.q_proj.weight": "pytorch_model-00014.bin",
278
+ "model.layers.30.self_attn.k_proj.weight": "pytorch_model-00014.bin",
279
+ "model.layers.30.self_attn.v_proj.weight": "pytorch_model-00014.bin",
280
+ "model.layers.30.self_attn.o_proj.weight": "pytorch_model-00014.bin",
281
+ "model.layers.30.mlp.gate_proj.weight": "pytorch_model-00014.bin",
282
+ "model.layers.30.mlp.up_proj.weight": "pytorch_model-00014.bin",
283
+ "model.layers.30.mlp.down_proj.weight": "pytorch_model-00014.bin",
284
+ "model.layers.30.input_layernorm.weight": "pytorch_model-00014.bin",
285
+ "model.layers.30.post_attention_layernorm.weight": "pytorch_model-00014.bin",
286
+ "model.layers.31.self_attn.q_proj.weight": "pytorch_model-00014.bin",
287
+ "model.layers.31.self_attn.k_proj.weight": "pytorch_model-00014.bin",
288
+ "model.layers.31.self_attn.v_proj.weight": "pytorch_model-00014.bin",
289
+ "model.layers.31.self_attn.o_proj.weight": "pytorch_model-00014.bin",
290
+ "model.layers.31.mlp.gate_proj.weight": "pytorch_model-00014.bin",
291
+ "model.layers.31.mlp.up_proj.weight": "pytorch_model-00014.bin",
292
+ "model.layers.31.mlp.down_proj.weight": "pytorch_model-00014.bin",
293
+ "model.layers.31.input_layernorm.weight": "pytorch_model-00014.bin",
294
+ "model.layers.31.post_attention_layernorm.weight": "pytorch_model-00014.bin",
295
+ "model.layers.32.self_attn.q_proj.weight": "pytorch_model-00014.bin",
296
+ "model.layers.32.self_attn.k_proj.weight": "pytorch_model-00014.bin",
297
+ "model.layers.32.self_attn.v_proj.weight": "pytorch_model-00014.bin",
298
+ "model.layers.32.self_attn.o_proj.weight": "pytorch_model-00015.bin",
299
+ "model.layers.32.mlp.gate_proj.weight": "pytorch_model-00015.bin",
300
+ "model.layers.32.mlp.up_proj.weight": "pytorch_model-00015.bin",
301
+ "model.layers.32.mlp.down_proj.weight": "pytorch_model-00015.bin",
302
+ "model.layers.32.input_layernorm.weight": "pytorch_model-00015.bin",
303
+ "model.layers.32.post_attention_layernorm.weight": "pytorch_model-00015.bin",
304
+ "model.layers.33.self_attn.q_proj.weight": "pytorch_model-00015.bin",
305
+ "model.layers.33.self_attn.k_proj.weight": "pytorch_model-00015.bin",
306
+ "model.layers.33.self_attn.v_proj.weight": "pytorch_model-00015.bin",
307
+ "model.layers.33.self_attn.o_proj.weight": "pytorch_model-00015.bin",
308
+ "model.layers.33.mlp.gate_proj.weight": "pytorch_model-00015.bin",
309
+ "model.layers.33.mlp.up_proj.weight": "pytorch_model-00015.bin",
310
+ "model.layers.33.mlp.down_proj.weight": "pytorch_model-00015.bin",
311
+ "model.layers.33.input_layernorm.weight": "pytorch_model-00015.bin",
312
+ "model.layers.33.post_attention_layernorm.weight": "pytorch_model-00015.bin",
313
+ "model.layers.34.self_attn.q_proj.weight": "pytorch_model-00015.bin",
314
+ "model.layers.34.self_attn.k_proj.weight": "pytorch_model-00015.bin",
315
+ "model.layers.34.self_attn.v_proj.weight": "pytorch_model-00015.bin",
316
+ "model.layers.34.self_attn.o_proj.weight": "pytorch_model-00015.bin",
317
+ "model.layers.34.mlp.gate_proj.weight": "pytorch_model-00015.bin",
318
+ "model.layers.34.mlp.up_proj.weight": "pytorch_model-00016.bin",
319
+ "model.layers.34.mlp.down_proj.weight": "pytorch_model-00016.bin",
320
+ "model.layers.34.input_layernorm.weight": "pytorch_model-00016.bin",
321
+ "model.layers.34.post_attention_layernorm.weight": "pytorch_model-00016.bin",
322
+ "model.layers.35.self_attn.q_proj.weight": "pytorch_model-00016.bin",
323
+ "model.layers.35.self_attn.k_proj.weight": "pytorch_model-00016.bin",
324
+ "model.layers.35.self_attn.v_proj.weight": "pytorch_model-00016.bin",
325
+ "model.layers.35.self_attn.o_proj.weight": "pytorch_model-00016.bin",
326
+ "model.layers.35.mlp.gate_proj.weight": "pytorch_model-00016.bin",
327
+ "model.layers.35.mlp.up_proj.weight": "pytorch_model-00016.bin",
328
+ "model.layers.35.mlp.down_proj.weight": "pytorch_model-00016.bin",
329
+ "model.layers.35.input_layernorm.weight": "pytorch_model-00016.bin",
330
+ "model.layers.35.post_attention_layernorm.weight": "pytorch_model-00016.bin",
331
+ "model.norm.weight": "pytorch_model-00016.bin",
332
+ "lm_head.weight": "pytorch_model-00016.bin"
333
+ }
334
+ }
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:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
tokenizer_config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ }
30
+ },
31
+ "additional_special_tokens": [],
32
+ "bos_token": "<s>",
33
+ "chat_template": "{%- if messages[0]['role'] == 'system' %}\n {%- set system_message = messages[0]['content'] %}\n {%- set loop_messages = messages[1:] %}\n{%- else %}\n {%- set loop_messages = messages %}\n{%- endif %}\n\n{{- bos_token }}\n{%- for message in loop_messages %}\n {%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}\n {{- raise_exception('After the optional system message, conversation roles must alternate user/assistant/user/assistant/...') }}\n {%- endif %}\n {%- if message['role'] == 'user' %}\n {%- if loop.first and system_message is defined %}\n {{- ' [INST] ' + system_message + '\\n\\n' + message['content'] + ' [/INST]' }}\n {%- else %}\n {{- ' [INST] ' + message['content'] + ' [/INST]' }}\n {%- endif %}\n {%- elif message['role'] == 'assistant' %}\n {{- ' ' + message['content'] + eos_token}}\n {%- else %}\n {{- raise_exception('Only user and assistant roles are supported, with the exception of an initial optional system message!') }}\n {%- endif %}\n{%- endfor %}\n",
34
+ "clean_up_tokenization_spaces": false,
35
+ "eos_token": "</s>",
36
+ "legacy": false,
37
+ "model_max_length": 1000000000000000019884624838656,
38
+ "pad_token": null,
39
+ "sp_model_kwargs": {},
40
+ "spaces_between_special_tokens": false,
41
+ "tokenizer_class": "LlamaTokenizer",
42
+ "unk_token": "<unk>",
43
+ "use_default_system_prompt": false
44
+ }