Hugging Face's logo

Xenova
/
t5-v1_1-small

Text2Text Generation
Transformers.js
ONNX
t5
Model card Files Community
1
Xenova HF staff commited on
Commit
1130de9
•
1 Parent(s): bb99df7

Upload folder using huggingface_hub

Browse files
Files changed (10) 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. quantize_config.json +139 -0
  10. tokenizer_config.json +1 -0
config.json CHANGED
@@ -3,6 +3,7 @@
3
  "architectures": [
4
  "T5ForConditionalGeneration"
5
  ],
 
6
  "d_ff": 1024,
7
  "d_kv": 64,
8
  "d_model": 512,
@@ -24,7 +25,7 @@
24
  "relative_attention_max_distance": 128,
25
  "relative_attention_num_buckets": 32,
26
  "tie_word_embeddings": false,
27
- "transformers_version": "4.29.2",
28
  "use_cache": true,
29
  "vocab_size": 32128
30
  }
 
3
  "architectures": [
4
  "T5ForConditionalGeneration"
5
  ],
6
+ "classifier_dropout": 0.0,
7
  "d_ff": 1024,
8
  "d_kv": 64,
9
  "d_model": 512,
 
25
  "relative_attention_max_distance": 128,
26
  "relative_attention_num_buckets": 32,
27
  "tie_word_embeddings": false,
28
+ "transformers_version": "4.33.0.dev0",
29
  "use_cache": true,
30
  "vocab_size": 32128
31
  }
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": "4.29.2"
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:d127e71a74ef0ce05f055942680cd888c4ff919148000d8bec72d70d28d52fcf
3
- size 232553947
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:52d42a1456684fa728a4a00719feda8da997e7f914949bb44dc0aad7d29a376d
3
+ size 232553640
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:81fc9fb634ada96ec24acbe496107e65c29d83001b90ca7bfa7d51d9145c9b1c
3
- size 232785013
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e626548c9f7354ba077c355ff4412a07dc455b35719e4f7cd467ff4a61cb7320
3
+ size 232784326
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d8d182742637a76d2224975f60201b91e4410e75f10961e70abac2b174a77da5
3
- size 59340172
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b79a93296dc081de06e9225e5268ccb4f22977f9e8aa92bd6a4623fd7c100e1a
3
+ size 59339331
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3a3961aee00cca80a6be942ef2dbcd8ead9d4bb1900dc486fece99c6a0ecfdc8
3
- size 58943706
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a29c9a46fecc1546ecadbc071cdd16de023b4154cb5b09749b08995849db71b8
3
+ size 58943318
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c79a621efa007d70baed6e3d3348296c8327d09f8dec0f8f7b03d26d8d9b6939
3
- size 219954259
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0487eff4fd959d7f85751178e18b663cd8855d16d8d8d6296bd95ac56bc1b1ed
3
+ size 219953954
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:10cdf3fb63a4629c0b5fc1aa8ed71af258c91ec80989156cc626d866ec30a6c4
3
- size 55729558
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:00c765a02883f4655abb9fdf2c190fabe152d11b9ed716c5351f9781c9082435
3
+ size 55729181
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
+ "Log",
8
+ "Gather",
9
+ "MatMul",
10
+ "Sqrt",
11
+ "Pow",
12
+ "Neg",
13
+ "ConstantOfShape",
14
+ "Cast",
15
+ "Less",
16
+ "Reshape",
17
+ "Tile",
18
+ "LessOrEqual",
19
+ "Range",
20
+ "ReduceMean",
21
+ "Where",
22
+ "Expand",
23
+ "Mul",
24
+ "Shape",
25
+ "Add",
26
+ "Concat",
27
+ "Transpose",
28
+ "Constant",
29
+ "Sub",
30
+ "Tanh",
31
+ "Softmax",
32
+ "Unsqueeze",
33
+ "Min",
34
+ "Div"
35
+ ],
36
+ "weight_type": "QInt8"
37
+ },
38
+ "encoder_model": {
39
+ "op_types": [
40
+ "Log",
41
+ "Gather",
42
+ "MatMul",
43
+ "Sqrt",
44
+ "Pow",
45
+ "Greater",
46
+ "Cast",
47
+ "ConstantOfShape",
48
+ "Less",
49
+ "Reshape",
50
+ "Range",
51
+ "ReduceMean",
52
+ "Where",
53
+ "Mul",
54
+ "Abs",
55
+ "Shape",
56
+ "Add",
57
+ "Concat",
58
+ "Transpose",
59
+ "Constant",
60
+ "Sub",
61
+ "Tanh",
62
+ "Softmax",
63
+ "Unsqueeze",
64
+ "Min",
65
+ "Div"
66
+ ],
67
+ "weight_type": "QInt8"
68
+ },
69
+ "decoder_model_merged": {
70
+ "op_types": [
71
+ "Log",
72
+ "Gather",
73
+ "MatMul",
74
+ "Sqrt",
75
+ "Pow",
76
+ "Neg",
77
+ "ConstantOfShape",
78
+ "Cast",
79
+ "Less",
80
+ "Reshape",
81
+ "Tile",
82
+ "LessOrEqual",
83
+ "Range",
84
+ "If",
85
+ "ReduceMean",
86
+ "Where",
87
+ "Expand",
88
+ "Mul",
89
+ "Shape",
90
+ "Add",
91
+ "Concat",
92
+ "Transpose",
93
+ "Constant",
94
+ "Sub",
95
+ "Tanh",
96
+ "Softmax",
97
+ "Unsqueeze",
98
+ "Min",
99
+ "Slice",
100
+ "Div"
101
+ ],
102
+ "weight_type": "QInt8"
103
+ },
104
+ "decoder_with_past_model": {
105
+ "op_types": [
106
+ "Log",
107
+ "Gather",
108
+ "MatMul",
109
+ "Sqrt",
110
+ "Pow",
111
+ "Neg",
112
+ "ConstantOfShape",
113
+ "Cast",
114
+ "Less",
115
+ "Reshape",
116
+ "Tile",
117
+ "LessOrEqual",
118
+ "Range",
119
+ "ReduceMean",
120
+ "Where",
121
+ "Expand",
122
+ "Mul",
123
+ "Shape",
124
+ "Add",
125
+ "Concat",
126
+ "Transpose",
127
+ "Constant",
128
+ "Sub",
129
+ "Tanh",
130
+ "Softmax",
131
+ "Unsqueeze",
132
+ "Min",
133
+ "Slice",
134
+ "Div"
135
+ ],
136
+ "weight_type": "QInt8"
137
+ }
138
+ }
139
+ }
tokenizer_config.json CHANGED
@@ -104,6 +104,7 @@
104
  "clean_up_tokenization_spaces": true,
105
  "eos_token": "</s>",
106
  "extra_ids": 100,
 
107
  "model_max_length": 512,
108
  "pad_token": "<pad>",
109
  "sp_model_kwargs": {},
 
104
  "clean_up_tokenization_spaces": true,
105
  "eos_token": "</s>",
106
  "extra_ids": 100,
107
+ "legacy": true,
108
  "model_max_length": 512,
109
  "pad_token": "<pad>",
110
  "sp_model_kwargs": {},