Xenova HF staff commited on
Commit
330cd50
1 Parent(s): ab7316a

Upload folder using huggingface_hub

Browse files
config.json CHANGED
@@ -19,7 +19,7 @@
19
  "rope_scaling": null,
20
  "rope_theta": 10000.0,
21
  "tie_word_embeddings": true,
22
- "transformers_version": "4.34.0.dev0",
23
  "use_cache": true,
24
  "vocab_size": 32000
25
  }
 
19
  "rope_scaling": null,
20
  "rope_theta": 10000.0,
21
  "tie_word_embeddings": true,
22
+ "transformers_version": "4.33.2",
23
  "use_cache": true,
24
  "vocab_size": 32000
25
  }
generation_config.json CHANGED
@@ -2,5 +2,5 @@
2
  "_from_model_config": true,
3
  "bos_token_id": 1,
4
  "eos_token_id": 2,
5
- "transformers_version": "4.34.0.dev0"
6
  }
 
2
  "_from_model_config": true,
3
  "bos_token_id": 1,
4
  "eos_token_id": 2,
5
+ "transformers_version": "4.33.2"
6
  }
onnx/decoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9b4055de40adc6080300cb0a1a8986259c228b2447689651d7ce91ce340380b6
3
- size 439063717
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:793dc3c2b6d845a6954eec6c173cf952485e3af5fa80b22a6337a03a921c5606
3
+ size 439064264
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c27b7753ee0c8f83371ce28ebe01b94cddf93cc1045a9171623dc03d3b561c38
3
- size 440015873
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e899dcfd259d7f983e216fffef3756c7e6f72871790ff9fd9cd1eb16601a94ba
3
+ size 440016914
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:53a4a3b603720730b5d126fe18fb61f68ca137903881ca3d84db3bb0776d11d8
3
- size 111933644
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba2a5f1df565916ffeaf95948f9ed9bcf1d5b87c3887cd94a7691fff1458a91c
3
+ size 111935100
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b6c5f19b29fc636f5975f964289fc67d23039120b6479969746c7d51d36f2574
3
- size 110756863
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0aedbba316748e1a640f9be5d920497f8c0d9be6ca4e1c37c2e71dfdc3dc5693
3
+ size 110757634
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a4194c70db527ebd117f4e530341522136e8c6322c86d9ff9cbff875da26c577
3
- size 439070831
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04f8916b363cd9226c6fc019eb26b181f43572b3a87b4cd564a9e78e597100db
3
+ size 439071283
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5e4a9d9b0e56de9f578abc299639909b3cca9258b87a55d0ef3ef9ae5ede0a55
3
- size 110765314
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c7d4a097526bc1b211b8e5c7b4f4004f2b58c9d0076893afa9ca31c65507648
3
+ size 110765955
quantize_config.json CHANGED
@@ -4,101 +4,102 @@
4
  "per_model_config": {
5
  "decoder_with_past_model": {
6
  "op_types": [
7
- "Sub",
8
- "Slice",
9
- "Constant",
10
- "Identity",
11
- "Equal",
12
- "Sigmoid",
13
- "Div",
14
- "Mul",
15
- "MatMul",
16
- "Pow",
17
  "Where",
18
- "Neg",
19
- "ConstantOfShape",
 
20
  "Softmax",
21
- "Add",
22
  "Sqrt",
23
- "Unsqueeze",
24
- "Reshape",
25
- "Expand",
26
- "ReduceMean",
27
  "Concat",
28
- "Cast",
 
 
 
 
 
 
29
  "Gather",
30
- "Transpose",
31
- "Shape",
 
32
  "If",
33
- "Squeeze"
 
 
 
 
 
34
  ],
35
  "weight_type": "QInt8"
36
  },
37
  "decoder_model": {
38
  "op_types": [
39
- "Sub",
40
- "Slice",
41
- "Constant",
42
- "Identity",
43
- "Equal",
44
- "Sigmoid",
45
- "Div",
46
- "Mul",
47
- "MatMul",
48
- "Pow",
49
  "Where",
50
- "Neg",
51
- "ConstantOfShape",
 
52
  "Softmax",
53
- "Add",
54
  "Sqrt",
55
- "Unsqueeze",
56
- "Reshape",
57
- "Expand",
58
- "ReduceMean",
59
  "Concat",
 
 
 
 
60
  "Less",
61
- "Cast",
 
 
62
  "Gather",
63
- "Range",
64
- "Transpose",
65
- "Shape",
66
  "If",
67
- "Squeeze"
 
 
 
 
 
68
  ],
69
  "weight_type": "QInt8"
70
  },
71
  "decoder_model_merged": {
72
  "op_types": [
73
- "Sub",
74
- "Slice",
75
- "Constant",
76
- "Identity",
77
- "Equal",
78
- "Sigmoid",
79
- "Div",
80
- "Mul",
81
- "MatMul",
82
- "Pow",
83
  "Where",
84
- "Neg",
85
- "ConstantOfShape",
 
86
  "Softmax",
87
- "Add",
88
  "Sqrt",
89
- "Unsqueeze",
90
- "Reshape",
91
- "Expand",
92
- "ReduceMean",
93
  "Concat",
 
 
 
 
94
  "Less",
95
- "Cast",
 
 
96
  "Gather",
97
- "Range",
98
- "Transpose",
99
- "Shape",
100
  "If",
101
- "Squeeze"
 
 
 
 
 
102
  ],
103
  "weight_type": "QInt8"
104
  }
 
4
  "per_model_config": {
5
  "decoder_with_past_model": {
6
  "op_types": [
 
 
 
 
 
 
 
 
 
 
7
  "Where",
8
+ "Cast",
9
+ "Mul",
10
+ "Slice",
11
  "Softmax",
 
12
  "Sqrt",
13
+ "Shape",
14
+ "Add",
15
+ "Range",
 
16
  "Concat",
17
+ "ReduceMean",
18
+ "Sub",
19
+ "Unsqueeze",
20
+ "ConstantOfShape",
21
+ "Squeeze",
22
+ "Identity",
23
+ "Constant",
24
  "Gather",
25
+ "Expand",
26
+ "Sigmoid",
27
+ "Equal",
28
  "If",
29
+ "Div",
30
+ "Neg",
31
+ "Pow",
32
+ "Transpose",
33
+ "MatMul",
34
+ "Reshape"
35
  ],
36
  "weight_type": "QInt8"
37
  },
38
  "decoder_model": {
39
  "op_types": [
 
 
 
 
 
 
 
 
 
 
40
  "Where",
41
+ "Cast",
42
+ "Mul",
43
+ "Slice",
44
  "Softmax",
 
45
  "Sqrt",
46
+ "Shape",
47
+ "Add",
48
+ "Range",
 
49
  "Concat",
50
+ "ReduceMean",
51
+ "Sub",
52
+ "Unsqueeze",
53
+ "ConstantOfShape",
54
  "Less",
55
+ "Squeeze",
56
+ "Identity",
57
+ "Constant",
58
  "Gather",
59
+ "Expand",
60
+ "Sigmoid",
61
+ "Equal",
62
  "If",
63
+ "Div",
64
+ "Neg",
65
+ "Pow",
66
+ "Transpose",
67
+ "MatMul",
68
+ "Reshape"
69
  ],
70
  "weight_type": "QInt8"
71
  },
72
  "decoder_model_merged": {
73
  "op_types": [
 
 
 
 
 
 
 
 
 
 
74
  "Where",
75
+ "Cast",
76
+ "Mul",
77
+ "Slice",
78
  "Softmax",
 
79
  "Sqrt",
80
+ "Shape",
81
+ "Add",
82
+ "Range",
 
83
  "Concat",
84
+ "ReduceMean",
85
+ "Sub",
86
+ "Unsqueeze",
87
+ "ConstantOfShape",
88
  "Less",
89
+ "Squeeze",
90
+ "Identity",
91
+ "Constant",
92
  "Gather",
93
+ "Expand",
94
+ "Sigmoid",
95
+ "Equal",
96
  "If",
97
+ "Div",
98
+ "Neg",
99
+ "Pow",
100
+ "Transpose",
101
+ "MatMul",
102
+ "Reshape"
103
  ],
104
  "weight_type": "QInt8"
105
  }
tokenizer.json CHANGED
@@ -9,7 +9,7 @@
9
  "single_word": false,
10
  "lstrip": false,
11
  "rstrip": false,
12
- "normalized": true,
13
  "special": true
14
  },
15
  {
@@ -18,7 +18,7 @@
18
  "single_word": false,
19
  "lstrip": false,
20
  "rstrip": false,
21
- "normalized": true,
22
  "special": true
23
  },
24
  {
@@ -27,7 +27,7 @@
27
  "single_word": false,
28
  "lstrip": false,
29
  "rstrip": false,
30
- "normalized": true,
31
  "special": true
32
  }
33
  ],
 
9
  "single_word": false,
10
  "lstrip": false,
11
  "rstrip": false,
12
+ "normalized": false,
13
  "special": true
14
  },
15
  {
 
18
  "single_word": false,
19
  "lstrip": false,
20
  "rstrip": false,
21
+ "normalized": false,
22
  "special": true
23
  },
24
  {
 
27
  "single_word": false,
28
  "lstrip": false,
29
  "rstrip": false,
30
+ "normalized": false,
31
  "special": true
32
  }
33
  ],