Hugging Face's logo

Xenova
/
pythia-160m-deduped

Text Generation
Transformers.js
ONNX
gpt_neox
Model card Files Community
Xenova HF staff commited on
Commit
5e578b5
•
1 Parent(s): dc2a299

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-160m-deduped",
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": 768,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 3072,
15
+ "layer_norm_eps": 1e-05,
16
+ "max_position_embeddings": 2048,
17
+ "model_type": "gpt_neox",
18
+ "num_attention_heads": 12,
19
+ "num_hidden_layers": 12,
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 sha256:4a998679daf59a3cc166952b886c07ba11103c45095c0e4a60d8fb5543c288bc
3
+ size 654410647
onnx/decoder_model_merged.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7781f5547942165364283de47a2cfa22c9e7ad69588008e1d5b5237708c6d87
3
+ size 659543749
onnx/decoder_model_merged_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b2ca8dffa5721772bf0283ae542e94782812b2fb2b868edde33f5a726e694580
3
+ size 173530346
onnx/decoder_model_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f838d77dfb7cef71a65b30059f3d9b6f2a825ce97060a00413300c2ac09b0b4a
3
+ size 168099893
onnx/decoder_with_past_model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a4356a07601892777bdba1c68767ca82c3a334bb568618d154e8082e1ea4413f
3
+ size 654425869
onnx/decoder_with_past_model_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c0e20397f02d9f70b9fc5fe1eef058252c78c20d5c50e344d1bcf8d88671bc3
3
+ size 168119246
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
+ "Tile",
8
+ "Neg",
9
+ "Gather",
10
+ "Mul",
11
+ "Div",
12
+ "Sub",
13
+ "Pow",
14
+ "Constant",
15
+ "Range",
16
+ "ConstantOfShape",
17
+ "Where",
18
+ "Sqrt",
19
+ "Add",
20
+ "Unsqueeze",
21
+ "ReduceMean",
22
+ "Concat",
23
+ "Squeeze",
24
+ "GatherElements",
25
+ "Cast",
26
+ "Softmax",
27
+ "Transpose",
28
+ "Erf",
29
+ "Shape",
30
+ "MatMul",
31
+ "Expand",
32
+ "Slice",
33
+ "Reshape"
34
+ ],
35
+ "weight_type": "QInt8"
36
+ },
37
+ "decoder_model_merged": {
38
+ "op_types": [
39
+ "Tile",
40
+ "Neg",
41
+ "Gather",
42
+ "Mul",
43
+ "Div",
44
+ "Sub",
45
+ "Pow",
46
+ "Constant",
47
+ "Range",
48
+ "ConstantOfShape",
49
+ "Where",
50
+ "Sqrt",
51
+ "Add",
52
+ "Unsqueeze",
53
+ "ReduceMean",
54
+ "Concat",
55
+ "Squeeze",
56
+ "GatherElements",
57
+ "Cast",
58
+ "Softmax",
59
+ "Transpose",
60
+ "If",
61
+ "Erf",
62
+ "Shape",
63
+ "MatMul",
64
+ "Expand",
65
+ "Slice",
66
+ "Reshape"
67
+ ],
68
+ "weight_type": "QInt8"
69
+ },
70
+ "decoder_with_past_model": {
71
+ "op_types": [
72
+ "Tile",
73
+ "Neg",
74
+ "Gather",
75
+ "Mul",
76
+ "Div",
77
+ "Sub",
78
+ "Pow",
79
+ "Constant",
80
+ "Range",
81
+ "ConstantOfShape",
82
+ "Where",
83
+ "Sqrt",
84
+ "Add",
85
+ "Unsqueeze",
86
+ "ReduceMean",
87
+ "Concat",
88
+ "Squeeze",
89
+ "GatherElements",
90
+ "Cast",
91
+ "Softmax",
92
+ "Transpose",
93
+ "Erf",
94
+ "Shape",
95
+ "MatMul",
96
+ "Expand",
97
+ "Slice",
98
+ "Reshape"
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
+ }