Hugging Face's logo

Xenova
/
LaMini-Neo-125M

Text Generation
Transformers.js
ONNX
gpt_neo
Model card Files Community
Xenova HF staff commited on
Commit
2a56393
•
1 Parent(s): b629947

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:a9be8abd3ce0240b3b0dd9cdaa378f4f00ea1427abf12a10b1478ba68bc19684
3
  size 518173878
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76689120f980a494806c310b7262c6eeb8db4753915dbbba3d77bbb6b33917d2
3
  size 518173878
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7c738a0bebd9cae8408ed1941192bd7cce35140997c8f42fca632dda543d1465
3
  size 143638305
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91a5452ab7e6a8cd75d708f498f721eabb30b664e630cf8ebf755659d365f24d
3
  size 143638305
quantize_config.json CHANGED
@@ -4,83 +4,83 @@
4
  "per_model_config": {
5
  "decoder_model": {
6
  "op_types": [
 
 
 
 
7
  "Tanh",
8
- "Slice",
9
- "MatMul",
10
- "Div",
11
  "Shape",
12
- "ReduceMean",
13
- "Concat",
14
  "Squeeze",
 
15
  "Constant",
 
 
16
  "Reshape",
17
  "Mul",
18
- "Where",
19
- "Range",
 
20
  "Gather",
21
- "Sqrt",
22
- "Pow",
23
- "Transpose",
24
- "Cast",
25
  "Sub",
26
- "Unsqueeze",
27
- "Add",
28
- "Softmax"
29
  ],
30
  "weight_type": "QInt8"
31
  },
32
  "decoder_model_merged": {
33
  "op_types": [
 
 
 
 
34
  "Tanh",
35
- "Slice",
36
- "MatMul",
37
- "Div",
38
  "Shape",
39
- "ReduceMean",
40
- "Concat",
41
  "Squeeze",
 
42
  "Constant",
 
 
43
  "Reshape",
44
  "Mul",
45
- "Where",
46
- "Range",
47
- "Gather",
48
- "Sqrt",
49
- "Pow",
50
- "Transpose",
51
- "Cast",
52
- "Sub",
53
- "Unsqueeze",
54
  "Add",
 
55
  "If",
56
- "Softmax"
 
 
 
 
57
  ],
58
  "weight_type": "QInt8"
59
  },
60
  "decoder_with_past_model": {
61
  "op_types": [
 
 
 
 
62
  "Tanh",
63
- "Slice",
64
- "MatMul",
65
- "Div",
66
  "Shape",
67
- "ReduceMean",
68
- "Concat",
69
  "Squeeze",
 
70
  "Constant",
 
 
71
  "Reshape",
72
  "Mul",
73
- "Where",
74
- "Range",
 
75
  "Gather",
76
- "Sqrt",
77
- "Pow",
78
- "Transpose",
79
- "Cast",
80
  "Sub",
81
- "Unsqueeze",
82
- "Add",
83
- "Softmax"
84
  ],
85
  "weight_type": "QInt8"
86
  }
 
4
  "per_model_config": {
5
  "decoder_model": {
6
  "op_types": [
7
+ "Sqrt",
8
+ "Softmax",
9
+ "Pow",
10
+ "Transpose",
11
  "Tanh",
12
+ "Where",
 
 
13
  "Shape",
14
+ "Range",
 
15
  "Squeeze",
16
+ "Div",
17
  "Constant",
18
+ "Unsqueeze",
19
+ "Concat",
20
  "Reshape",
21
  "Mul",
22
+ "Add",
23
+ "Slice",
24
+ "ReduceMean",
25
  "Gather",
 
 
 
 
26
  "Sub",
27
+ "MatMul",
28
+ "Cast"
 
29
  ],
30
  "weight_type": "QInt8"
31
  },
32
  "decoder_model_merged": {
33
  "op_types": [
34
+ "Sqrt",
35
+ "Softmax",
36
+ "Pow",
37
+ "Transpose",
38
  "Tanh",
39
+ "Where",
 
 
40
  "Shape",
41
+ "Range",
 
42
  "Squeeze",
43
+ "Div",
44
  "Constant",
45
+ "Unsqueeze",
46
+ "Concat",
47
  "Reshape",
48
  "Mul",
 
 
 
 
 
 
 
 
 
49
  "Add",
50
+ "Slice",
51
  "If",
52
+ "ReduceMean",
53
+ "Gather",
54
+ "Sub",
55
+ "MatMul",
56
+ "Cast"
57
  ],
58
  "weight_type": "QInt8"
59
  },
60
  "decoder_with_past_model": {
61
  "op_types": [
62
+ "Sqrt",
63
+ "Softmax",
64
+ "Pow",
65
+ "Transpose",
66
  "Tanh",
67
+ "Where",
 
 
68
  "Shape",
69
+ "Range",
 
70
  "Squeeze",
71
+ "Div",
72
  "Constant",
73
+ "Unsqueeze",
74
+ "Concat",
75
  "Reshape",
76
  "Mul",
77
+ "Add",
78
+ "Slice",
79
+ "ReduceMean",
80
  "Gather",
 
 
 
 
81
  "Sub",
82
+ "MatMul",
83
+ "Cast"
 
84
  ],
85
  "weight_type": "QInt8"
86
  }