Hugging Face's logo

TheBloke
/
Mixtral_7Bx2_MoE-GPTQ

Text Generation
Transformers
Safetensors
mixtral
text-generation-inference
4-bit precision
gptq
Model card Files Community
TheBloke commited on
Commit
9dddeeb
•
1 Parent(s): 34f22cb

GPTQ model commit

Browse files
Files changed (6) hide show
  1. config.json +82 -0
  2. model.safetensors +3 -0
  3. quantize_config.json +11 -0
  4. special_tokens_map.json +29 -0
  5. tokenizer.json +0 -0
  6. tokenizer_config.json +48 -0
config.json ADDED
@@ -0,0 +1,82 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/workspace/process/cloudyu_mixtral_7bx2_moe/source",
3
+ "architectures": [
4
+ "MixtralForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 1,
8
+ "eos_token_id": 2,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 4096,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 14336,
13
+ "max_position_embeddings": 32768,
14
+ "model_type": "mixtral",
15
+ "num_attention_heads": 32,
16
+ "num_experts_per_tok": 2,
17
+ "num_hidden_layers": 32,
18
+ "num_key_value_heads": 8,
19
+ "num_local_experts": 2,
20
+ "output_router_logits": false,
21
+ "pad_token_id": 0,
22
+ "pretraining_tp": 1,
23
+ "rms_norm_eps": 1e-05,
24
+ "rope_theta": 10000.0,
25
+ "router_aux_loss_coef": 0.001,
26
+ "sliding_window": null,
27
+ "tie_word_embeddings": false,
28
+ "torch_dtype": "bfloat16",
29
+ "transformers_version": "4.37.0.dev0",
30
+ "use_cache": true,
31
+ "vocab_size": 32000,
32
+ "quantization_config": {
33
+ "bits": 4,
34
+ "group_size": 128,
35
+ "damp_percent": 0.1,
36
+ "desc_act": true,
37
+ "static_groups": false,
38
+ "sym": true,
39
+ "true_sequential": true,
40
+ "model_name_or_path": null,
41
+ "model_file_base_name": "model",
42
+ "quant_method": "gptq",
43
+ "modules_in_block_to_quantize": [
44
+ [
45
+ "self_attn.k_proj",
46
+ "self_attn.v_proj",
47
+ "self_attn.q_proj"
48
+ ],
49
+ [
50
+ "self_attn.o_proj"
51
+ ],
52
+ [
53
+ "block_sparse_moe.experts.0.w1",
54
+ "block_sparse_moe.experts.1.w1",
55
+ "block_sparse_moe.experts.2.w1",
56
+ "block_sparse_moe.experts.3.w1",
57
+ "block_sparse_moe.experts.4.w1",
58
+ "block_sparse_moe.experts.5.w1",
59
+ "block_sparse_moe.experts.6.w1",
60
+ "block_sparse_moe.experts.7.w1",
61
+ "block_sparse_moe.experts.0.w3",
62
+ "block_sparse_moe.experts.1.w3",
63
+ "block_sparse_moe.experts.2.w3",
64
+ "block_sparse_moe.experts.3.w3",
65
+ "block_sparse_moe.experts.4.w3",
66
+ "block_sparse_moe.experts.5.w3",
67
+ "block_sparse_moe.experts.6.w3",
68
+ "block_sparse_moe.experts.7.w3"
69
+ ],
70
+ [
71
+ "block_sparse_moe.experts.0.w2",
72
+ "block_sparse_moe.experts.1.w2",
73
+ "block_sparse_moe.experts.2.w2",
74
+ "block_sparse_moe.experts.3.w2",
75
+ "block_sparse_moe.experts.4.w2",
76
+ "block_sparse_moe.experts.5.w2",
77
+ "block_sparse_moe.experts.6.w2",
78
+ "block_sparse_moe.experts.7.w2"
79
+ ]
80
+ ]
81
+ }
82
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6da8efe37e2272198d15a25e8c2362d254b328c0aec445d8e1fca9acb1458b29
3
+ size 7092912896
quantize_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bits": 4,
3
+ "group_size": 128,
4
+ "damp_percent": 0.1,
5
+ "desc_act": true,
6
+ "static_groups": false,
7
+ "sym": true,
8
+ "true_sequential": true,
9
+ "model_name_or_path": null,
10
+ "model_file_base_name": "model"
11
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<unk>",
4
+ "<s>",
5
+ "</s>"
6
+ ],
7
+ "bos_token": {
8
+ "content": "<s>",
9
+ "lstrip": false,
10
+ "normalized": false,
11
+ "rstrip": false,
12
+ "single_word": false
13
+ },
14
+ "eos_token": {
15
+ "content": "</s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false
20
+ },
21
+ "pad_token": "<s>",
22
+ "unk_token": {
23
+ "content": "<unk>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false
28
+ }
29
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "additional_special_tokens": [
31
+ "<unk>",
32
+ "<s>",
33
+ "</s>"
34
+ ],
35
+ "bos_token": "<s>",
36
+ "clean_up_tokenization_spaces": false,
37
+ "eos_token": "</s>",
38
+ "legacy": true,
39
+ "model_max_length": 1000000000000000019884624838656,
40
+ "pad_token": "<s>",
41
+ "padding_side": "left",
42
+ "sp_model_kwargs": {},
43
+ "spaces_between_special_tokens": false,
44
+ "split_special_tokens": false,
45
+ "tokenizer_class": "LlamaTokenizer",
46
+ "unk_token": "<unk>",
47
+ "use_default_system_prompt": true
48
+ }