Hugging Face's logo

Xenova
/
pythia-31m

Text Generation
Transformers.js
ONNX
gpt_neox
Model card Files Community
Xenova HF staff commited on
Commit
9d55a1b
•
1 Parent(s): 0d97305

Upload folder using huggingface_hub

Browse files
Files changed (12) hide show
  1. config.json +28 -0
  2. generation_config.json +6 -0
  3. onnx/decoder_model.onnx +3 -0
  4. onnx/decoder_model_merged.onnx +3 -0
  5. onnx/decoder_model_merged_quantized.onnx +3 -0
  6. onnx/decoder_model_quantized.onnx +3 -0
  7. onnx/decoder_with_past_model.onnx +3 -0
  8. onnx/decoder_with_past_model_quantized.onnx +3 -0
  9. quantize_config.json +103 -0
  10. special_tokens_map.json +5 -0
  11. tokenizer.json +0 -0
  12. tokenizer_config.json +9 -0
config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "EleutherAI/pythia-31m",
3
+ "architectures": [
4
+ "GPTNeoXForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": 0.1,
9
+ "eos_token_id": 0,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout": 0.0,
12
+ "hidden_size": 256,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 1024,
15
+ "layer_norm_eps": 1e-05,
16
+ "max_position_embeddings": 2048,
17
+ "model_type": "gpt_neox",
18
+ "num_attention_heads": 8,
19
+ "num_hidden_layers": 6,
20
+ "rope_scaling": null,
21
+ "rotary_emb_base": 10000,
22
+ "rotary_pct": 0.25,
23
+ "tie_word_embeddings": false,
24
+ "transformers_version": "4.34.0.dev0",
25
+ "use_cache": true,
26
+ "use_parallel_residual": true,
27
+ "vocab_size": 50304
28
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "eos_token_id": 0,
5
+ "transformers_version": "4.34.0.dev0"
6
+ }
onnx/decoder_model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83cf7f2b85db8ef2b0b26bd18ad5bc79f1d3dc48886badba75c7a39073981505
3
+ size 126637665
onnx/decoder_model_merged.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a534b8fea53ceabdf60137b6969ba1c840bbdb5d2704c8aef20ff4247fd2bb74
3
+ size 131304026
onnx/decoder_model_merged_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6910684747c8bc2af093c43cbff570138b9ffaf07384d4f257f19ea76742b2b3
3
+ size 40181145
onnx/decoder_model_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:60b26ca6823a9c0992cc21900e54306d04ab183178b0e734ce21bf3466266229
3
+ size 35363964
onnx/decoder_with_past_model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65428d89a8efef61d8d62c006b693ac59335aac61429929ee3af192b10e9d888
3
+ size 126645671
onnx/decoder_with_past_model_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c43a107123690d3ca183d40a92865d47268e8472ef7231365cca4e22cc1d672
3
+ size 35373974
quantize_config.json ADDED
@@ -0,0 +1,103 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "per_channel": false,
3
+ "reduce_range": false,
4
+ "per_model_config": {
5
+ "decoder_model": {
6
+ "op_types": [
7
+ "Add",
8
+ "Sqrt",
9
+ "Cast",
10
+ "Constant",
11
+ "ReduceMean",
12
+ "Squeeze",
13
+ "Transpose",
14
+ "Where",
15
+ "Unsqueeze",
16
+ "Concat",
17
+ "Expand",
18
+ "Sub",
19
+ "Reshape",
20
+ "Softmax",
21
+ "Pow",
22
+ "Neg",
23
+ "Div",
24
+ "Tile",
25
+ "Gather",
26
+ "Erf",
27
+ "Shape",
28
+ "Mul",
29
+ "Range",
30
+ "ConstantOfShape",
31
+ "MatMul",
32
+ "Slice",
33
+ "GatherElements"
34
+ ],
35
+ "weight_type": "QInt8"
36
+ },
37
+ "decoder_model_merged": {
38
+ "op_types": [
39
+ "Add",
40
+ "Sqrt",
41
+ "Cast",
42
+ "Constant",
43
+ "ReduceMean",
44
+ "Squeeze",
45
+ "Transpose",
46
+ "Where",
47
+ "Unsqueeze",
48
+ "If",
49
+ "Concat",
50
+ "Expand",
51
+ "Sub",
52
+ "Reshape",
53
+ "Softmax",
54
+ "Pow",
55
+ "Neg",
56
+ "Div",
57
+ "Tile",
58
+ "Gather",
59
+ "Erf",
60
+ "Shape",
61
+ "Mul",
62
+ "Range",
63
+ "ConstantOfShape",
64
+ "MatMul",
65
+ "Slice",
66
+ "GatherElements"
67
+ ],
68
+ "weight_type": "QInt8"
69
+ },
70
+ "decoder_with_past_model": {
71
+ "op_types": [
72
+ "Add",
73
+ "Sqrt",
74
+ "Cast",
75
+ "Constant",
76
+ "ReduceMean",
77
+ "Squeeze",
78
+ "Transpose",
79
+ "Where",
80
+ "Unsqueeze",
81
+ "Concat",
82
+ "Expand",
83
+ "Sub",
84
+ "Reshape",
85
+ "Softmax",
86
+ "Pow",
87
+ "Neg",
88
+ "Div",
89
+ "Tile",
90
+ "Gather",
91
+ "Erf",
92
+ "Shape",
93
+ "Mul",
94
+ "Range",
95
+ "ConstantOfShape",
96
+ "MatMul",
97
+ "Slice",
98
+ "GatherElements"
99
+ ],
100
+ "weight_type": "QInt8"
101
+ }
102
+ }
103
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<|endoftext|>",
3
+ "eos_token": "<|endoftext|>",
4
+ "unk_token": "<|endoftext|>"
5
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "bos_token": "<|endoftext|>",
4
+ "clean_up_tokenization_spaces": true,
5
+ "eos_token": "<|endoftext|>",
6
+ "model_max_length": 1000000000000000019884624838656,
7
+ "tokenizer_class": "GPTNeoXTokenizer",
8
+ "unk_token": "<|endoftext|>"
9
+ }