Hugging Face's logo

Xenova
/
gpt-neo-125M

Text Generation
Transformers.js
ONNX
gpt_neo
Model card Files Community
Xenova HF staff commited on
Commit
4deb1ca
•
1 Parent(s): 4381c8f

Upload folder using huggingface_hub

Browse files
Files changed (3) hide show
  1. onnx/decoder_model_merged.onnx +1 -1
  2. onnx/decoder_model_merged_quantized.onnx +1 -1
  3. quantize_config.json +43 -43
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a93d4be13c01d658f7098461e96c67e896397c79fbbad2f512f2ad7ffdb65122
3
  size 518170806
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac8c9db792a77069251a5f4fae2865414e9e0d1efac916505b5aa523959e664b
3
  size 518170806
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c3934849b8b8f2c03bede375252f6d0a1f04b9e9464778ce85748011f0f1bf4b
3
  size 143637537
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac2acdb1665449ae0d70312dc1bedf08b42418bebfb24b9727bbde1ed5b20862
3
  size 143637537
quantize_config.json CHANGED
@@ -4,83 +4,83 @@
4
  "per_model_config": {
5
  "decoder_model": {
6
  "op_types": [
 
7
  "Squeeze",
8
- "Softmax",
9
- "Add",
10
  "Reshape",
11
- "Pow",
12
- "Concat",
13
- "Slice",
14
- "Mul",
15
  "Gather",
 
16
  "Unsqueeze",
17
- "ReduceMean",
 
18
  "Sub",
19
- "Cast",
20
- "Tanh",
21
- "Shape",
22
- "Constant",
23
  "Range",
24
  "MatMul",
25
- "Div",
26
- "Sqrt",
27
  "Where",
28
- "Transpose"
 
 
 
 
29
  ],
30
  "weight_type": "QInt8"
31
  },
32
  "decoder_model_merged": {
33
  "op_types": [
 
34
  "Squeeze",
35
- "Softmax",
36
- "Add",
37
  "Reshape",
38
- "Pow",
39
- "Concat",
40
- "Slice",
41
- "Mul",
42
  "Gather",
 
43
  "Unsqueeze",
44
- "ReduceMean",
 
45
  "Sub",
46
- "Cast",
47
- "Tanh",
48
- "Shape",
49
- "Transpose",
50
  "Range",
51
- "If",
52
  "MatMul",
53
- "Div",
54
- "Sqrt",
 
55
  "Where",
56
- "Constant"
 
 
 
 
57
  ],
58
  "weight_type": "QInt8"
59
  },
60
  "decoder_with_past_model": {
61
  "op_types": [
 
62
  "Squeeze",
63
- "Softmax",
64
- "Add",
65
  "Reshape",
66
- "Pow",
67
- "Concat",
68
- "Slice",
69
- "Mul",
70
  "Gather",
 
71
  "Unsqueeze",
72
- "ReduceMean",
 
73
  "Sub",
74
- "Cast",
75
- "Tanh",
76
- "Shape",
77
- "Constant",
78
  "Range",
79
  "MatMul",
80
- "Div",
81
- "Sqrt",
82
  "Where",
83
- "Transpose"
 
 
 
 
84
  ],
85
  "weight_type": "QInt8"
86
  }
 
4
  "per_model_config": {
5
  "decoder_model": {
6
  "op_types": [
7
+ "Div",
8
  "Squeeze",
 
 
9
  "Reshape",
10
+ "Transpose",
11
+ "Softmax",
 
 
12
  "Gather",
13
+ "Mul",
14
  "Unsqueeze",
15
+ "Slice",
16
+ "Sqrt",
17
  "Sub",
18
+ "Concat",
 
 
 
19
  "Range",
20
  "MatMul",
21
+ "ReduceMean",
22
+ "Shape",
23
  "Where",
24
+ "Cast",
25
+ "Add",
26
+ "Constant",
27
+ "Tanh",
28
+ "Pow"
29
  ],
30
  "weight_type": "QInt8"
31
  },
32
  "decoder_model_merged": {
33
  "op_types": [
34
+ "Div",
35
  "Squeeze",
 
 
36
  "Reshape",
37
+ "Transpose",
38
+ "Softmax",
 
 
39
  "Gather",
40
+ "Mul",
41
  "Unsqueeze",
42
+ "Slice",
43
+ "Sqrt",
44
  "Sub",
45
+ "Concat",
 
 
 
46
  "Range",
 
47
  "MatMul",
48
+ "If",
49
+ "ReduceMean",
50
+ "Shape",
51
  "Where",
52
+ "Cast",
53
+ "Add",
54
+ "Constant",
55
+ "Tanh",
56
+ "Pow"
57
  ],
58
  "weight_type": "QInt8"
59
  },
60
  "decoder_with_past_model": {
61
  "op_types": [
62
+ "Div",
63
  "Squeeze",
 
 
64
  "Reshape",
65
+ "Transpose",
66
+ "Softmax",
 
 
67
  "Gather",
68
+ "Mul",
69
  "Unsqueeze",
70
+ "Slice",
71
+ "Sqrt",
72
  "Sub",
73
+ "Concat",
 
 
 
74
  "Range",
75
  "MatMul",
76
+ "ReduceMean",
77
+ "Shape",
78
  "Where",
79
+ "Cast",
80
+ "Add",
81
+ "Constant",
82
+ "Tanh",
83
+ "Pow"
84
  ],
85
  "weight_type": "QInt8"
86
  }