Hugging Face's logo

Xenova
/
mt5-base

Transformers.js
ONNX
mt5
text2text-generation
Model card Files
xet
Community
3
Xenova HF Staff commited on
Commit
8e13e5f
·
1 Parent(s): f3142fb

Upload folder using huggingface_hub

Browse files
Files changed (12) hide show
  1. config.json +2 -1
  2. generation_config.json +1 -1
  3. onnx/decoder_model.onnx +2 -2
  4. onnx/decoder_model_merged.onnx +2 -2
  5. onnx/decoder_model_merged_quantized.onnx +2 -2
  6. onnx/decoder_model_quantized.onnx +2 -2
  7. onnx/decoder_with_past_model.onnx +2 -2
  8. onnx/decoder_with_past_model_quantized.onnx +2 -2
  9. onnx/encoder_model.onnx +1 -1
  10. onnx/encoder_model_quantized.onnx +2 -2
  11. quantize_config.json +139 -0
  12. tokenizer_config.json +1 -0
config.json CHANGED
@@ -3,6 +3,7 @@
3
  "architectures": [
4
  "MT5ForConditionalGeneration"
5
  ],
 
6
  "d_ff": 2048,
7
  "d_kv": 64,
8
  "d_model": 768,
@@ -25,7 +26,7 @@
25
  "relative_attention_num_buckets": 32,
26
  "tie_word_embeddings": false,
27
  "tokenizer_class": "T5Tokenizer",
28
- "transformers_version": "4.28.1",
29
  "use_cache": true,
30
  "vocab_size": 250112
31
  }
 
3
  "architectures": [
4
  "MT5ForConditionalGeneration"
5
  ],
6
+ "classifier_dropout": 0.0,
7
  "d_ff": 2048,
8
  "d_kv": 64,
9
  "d_model": 768,
 
26
  "relative_attention_num_buckets": 32,
27
  "tie_word_embeddings": false,
28
  "tokenizer_class": "T5Tokenizer",
29
+ "transformers_version": "4.33.0.dev0",
30
  "use_cache": true,
31
  "vocab_size": 250112
32
  }
generation_config.json CHANGED
@@ -3,5 +3,5 @@
3
  "decoder_start_token_id": 0,
4
  "eos_token_id": 1,
5
  "pad_token_id": 0,
6
- "transformers_version": "4.28.1"
7
  }
 
3
  "decoder_start_token_id": 0,
4
  "eos_token_id": 1,
5
  "pad_token_id": 0,
6
+ "transformers_version": "4.33.0.dev0"
7
  }
onnx/decoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:446c4420b33340663588efa44f4edd9a80cd52efb7605afe7d198e3667027b27
3
- size 1990142970
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c48668239d5c97066b00a0dd7c78d618f45ba9e1911daced0ad2cc9a5c915783
3
+ size 1990142661
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a30cc06a6ada3c522caeb065280348eadadda31e7e94db38feb792a6eae83763
3
- size 1990477278
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4823de9d05f2cdcf8743e29f9b7d72f5374d6899a6240144a53c48009d90a0ef
3
+ size 1990476587
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4a5ad041f1f348d8654671e0e88f119f00fd38dec9c5de8719538ed13ce64927
3
- size 498742113
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c456eabc1724752ff14267e84f6d3b7b5e7757873cf6ae82c916a49ccc4efce0
3
+ size 500652976
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7cc5d67b457cb18a5b59361839f9ae9185ac120693d97f1370e436645a80e20d
3
- size 498165142
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07cde49811938805c3e59fc87608aba75c408a2d9a4827e73e5fe33b87057911
3
+ size 500076462
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:edb08cc4d61361168917ce3d099ba53546e7097ddf465fa0bfeb0eedd7ba356c
3
- size 1933491316
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed18f6622a1c609c54ba3850ff93a39574192733469eaa143d799244007ad0cd
3
+ size 1933491009
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b9b489c732aae00276bf020389d4abdaecf00d4f58019e035ecffc5667541aba
3
- size 483948445
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb11ef4ba08e9ed766c1f0ffdc468fa8666da08b6e01a4be337555eae769bd5b
3
+ size 485767496
onnx/encoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0c9d15e3a303f7059a0ddb3e5af4ebaf74343f7c0f81fd9de174f5666c07df0c
3
  size 1108344238
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76b26f34b6bd3834064d5ac9b6ac2695367b6db43816b9ffea3ccb2938d6552c
3
  size 1108344238
onnx/encoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dccaebf4dec740d215e310dc7dc86540fcc0ced03d2a64fb7aff96b4a42b7f91
3
- size 277437491
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:96011ec0b2e27757156d4d2d933b74a317d59558b6415387834fca9eab516a7d
3
+ size 277914071
quantize_config.json ADDED
@@ -0,0 +1,139 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "per_channel": true,
3
+ "reduce_range": true,
4
+ "per_model_config": {
5
+ "decoder_model": {
6
+ "op_types": [
7
+ "Gather",
8
+ "Where",
9
+ "Concat",
10
+ "Tanh",
11
+ "Min",
12
+ "Tile",
13
+ "Cast",
14
+ "Shape",
15
+ "Pow",
16
+ "Log",
17
+ "Mul",
18
+ "Sqrt",
19
+ "Less",
20
+ "Softmax",
21
+ "Range",
22
+ "MatMul",
23
+ "Add",
24
+ "Transpose",
25
+ "Reshape",
26
+ "ConstantOfShape",
27
+ "LessOrEqual",
28
+ "ReduceMean",
29
+ "Expand",
30
+ "Sub",
31
+ "Unsqueeze",
32
+ "Constant",
33
+ "Div",
34
+ "Neg"
35
+ ],
36
+ "weight_type": "QInt8"
37
+ },
38
+ "encoder_model": {
39
+ "op_types": [
40
+ "Gather",
41
+ "Where",
42
+ "Concat",
43
+ "Tanh",
44
+ "Min",
45
+ "Cast",
46
+ "Shape",
47
+ "Pow",
48
+ "Greater",
49
+ "Log",
50
+ "Mul",
51
+ "Sqrt",
52
+ "Less",
53
+ "Softmax",
54
+ "Range",
55
+ "MatMul",
56
+ "Add",
57
+ "Transpose",
58
+ "Reshape",
59
+ "ConstantOfShape",
60
+ "ReduceMean",
61
+ "Sub",
62
+ "Abs",
63
+ "Unsqueeze",
64
+ "Constant",
65
+ "Div"
66
+ ],
67
+ "weight_type": "QInt8"
68
+ },
69
+ "decoder_model_merged": {
70
+ "op_types": [
71
+ "Gather",
72
+ "Where",
73
+ "Concat",
74
+ "Tanh",
75
+ "Min",
76
+ "Tile",
77
+ "Cast",
78
+ "Shape",
79
+ "Pow",
80
+ "Log",
81
+ "Mul",
82
+ "Sqrt",
83
+ "Less",
84
+ "Softmax",
85
+ "Range",
86
+ "MatMul",
87
+ "Add",
88
+ "Transpose",
89
+ "Reshape",
90
+ "Slice",
91
+ "ConstantOfShape",
92
+ "If",
93
+ "LessOrEqual",
94
+ "ReduceMean",
95
+ "Expand",
96
+ "Sub",
97
+ "Unsqueeze",
98
+ "Constant",
99
+ "Div",
100
+ "Neg"
101
+ ],
102
+ "weight_type": "QInt8"
103
+ },
104
+ "decoder_with_past_model": {
105
+ "op_types": [
106
+ "Gather",
107
+ "Where",
108
+ "Concat",
109
+ "Tanh",
110
+ "Min",
111
+ "Tile",
112
+ "Cast",
113
+ "Shape",
114
+ "Pow",
115
+ "Log",
116
+ "Mul",
117
+ "Sqrt",
118
+ "Less",
119
+ "Softmax",
120
+ "Range",
121
+ "MatMul",
122
+ "Add",
123
+ "Transpose",
124
+ "Reshape",
125
+ "Slice",
126
+ "ConstantOfShape",
127
+ "LessOrEqual",
128
+ "ReduceMean",
129
+ "Expand",
130
+ "Sub",
131
+ "Unsqueeze",
132
+ "Constant",
133
+ "Div",
134
+ "Neg"
135
+ ],
136
+ "weight_type": "QInt8"
137
+ }
138
+ }
139
+ }
tokenizer_config.json CHANGED
@@ -3,6 +3,7 @@
3
  "clean_up_tokenization_spaces": true,
4
  "eos_token": "</s>",
5
  "extra_ids": 0,
 
6
  "model_max_length": 1000000000000000019884624838656,
7
  "pad_token": "<pad>",
8
  "sp_model_kwargs": {},
 
3
  "clean_up_tokenization_spaces": true,
4
  "eos_token": "</s>",
5
  "extra_ids": 0,
6
+ "legacy": true,
7
  "model_max_length": 1000000000000000019884624838656,
8
  "pad_token": "<pad>",
9
  "sp_model_kwargs": {},