Hugging Face's logo

Xenova
/
LaMini-Flan-T5-77M

Text2Text Generation
Transformers.js
ONNX
t5
Model card Files Community
1
Xenova HF staff commited on
Commit
bc91a8c
•
1 Parent(s): eb4948a

Upload folder using huggingface_hub

Browse files
Files changed (12) hide show
  1. config.json +2 -2
  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 +4 -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,
@@ -54,8 +55,7 @@
54
  }
55
  },
56
  "tie_word_embeddings": false,
57
- "torch_dtype": "float32",
58
- "transformers_version": "4.28.1",
59
  "use_cache": true,
60
  "vocab_size": 32128
61
  }
 
3
  "architectures": [
4
  "T5ForConditionalGeneration"
5
  ],
6
+ "classifier_dropout": 0.0,
7
  "d_ff": 1024,
8
  "d_kv": 64,
9
  "d_model": 512,
 
55
  }
56
  },
57
  "tie_word_embeddings": false,
58
+ "transformers_version": "4.34.0.dev0",
 
59
  "use_cache": true,
60
  "vocab_size": 32128
61
  }
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": "4.34.0.dev0"
7
  }
onnx/decoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ac555f68055f812730a99fc89ec613915430bb2190358d9e4d913af55de3c421
3
- size 232553947
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a233d8d3e8e0f427a7e1c710e43f24bb14840d6c9f081adae4b063f6087df8d4
3
+ size 232553640
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:316835cc9cd3689cefc56989299152998789ea4e3e0d5d45c4509456dc4cdac2
3
- size 232785013
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd4110bfcce60a6289ab6c87a80d2d2b792de988c626625d107ff7d6a6827a76
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 sha256:785216dabd0dc6505beb043866b9815ad0526dcbc38fdfc02026141c6334f33f
3
- size 58943560
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e2cc3c6a18ba1952567754366c7a76c0f266afc5d071b5488138b7f684d63c28
3
+ size 59339330
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:450cd59ce03e07c623dede0b5029b3c6fa091770830e03ce9a4746b600a18886
3
- size 58547094
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:924e82748d8c18411bf0fd44d231049679c0a5a790df80ce5aaf7e23fb5a8cf7
3
+ size 58943317
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cbc1ec900adc17c19cba148cfd617f402d25807b532784583fc7ef494000d284
3
- size 219954259
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:152c3daae461a9ebd3049894e8c5903b341c641063e49b7f8dfc5fcf4a4d5133
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 sha256:92a4ce22a1aa8aefdd14863f1cf23528cae6dcccaf3127242e14fda0ef50679f
3
- size 55363746
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b92b469595120f1e1e682396795c1dc8958106f9cd56c1ab4f84935e13694fb
3
+ size 55729180
onnx/encoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:38e46b7cea40a4a13bc90c790949cbd635727d0faaad17b30a0ce3572ec4def2
3
  size 141456352
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5693b8e24d0d9ecdbdc1ae8c35b1a3c2908bdb6bc5ceb22c60246fc70f442f23
3
  size 141456352
onnx/encoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:911514fe167787f3c9ef2e8e1f19e61cb20b1b455ade26603cd28d3d06ae0c8f
3
- size 35589888
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f390de450b7d6f23817795484e37ccb0f6c2bbe618f4bad414712c664cfe07d
3
+ size 35759128
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
+ "Mul",
8
+ "Transpose",
9
+ "Softmax",
10
+ "Gather",
11
+ "Constant",
12
+ "Neg",
13
+ "Sub",
14
+ "ConstantOfShape",
15
+ "Concat",
16
+ "MatMul",
17
+ "Div",
18
+ "LessOrEqual",
19
+ "Shape",
20
+ "Log",
21
+ "Min",
22
+ "Reshape",
23
+ "Tanh",
24
+ "Tile",
25
+ "Add",
26
+ "Cast",
27
+ "Range",
28
+ "ReduceMean",
29
+ "Where",
30
+ "Sqrt",
31
+ "Unsqueeze",
32
+ "Pow",
33
+ "Less",
34
+ "Expand"
35
+ ],
36
+ "weight_type": "QInt8"
37
+ },
38
+ "encoder_model": {
39
+ "op_types": [
40
+ "Mul",
41
+ "Transpose",
42
+ "Softmax",
43
+ "Gather",
44
+ "Constant",
45
+ "Sub",
46
+ "ConstantOfShape",
47
+ "Concat",
48
+ "MatMul",
49
+ "Div",
50
+ "Shape",
51
+ "Log",
52
+ "Min",
53
+ "Reshape",
54
+ "Tanh",
55
+ "Add",
56
+ "Cast",
57
+ "Range",
58
+ "ReduceMean",
59
+ "Where",
60
+ "Sqrt",
61
+ "Greater",
62
+ "Unsqueeze",
63
+ "Pow",
64
+ "Less",
65
+ "Abs"
66
+ ],
67
+ "weight_type": "QInt8"
68
+ },
69
+ "decoder_model_merged": {
70
+ "op_types": [
71
+ "Mul",
72
+ "Transpose",
73
+ "Softmax",
74
+ "Gather",
75
+ "Constant",
76
+ "Neg",
77
+ "Sub",
78
+ "ConstantOfShape",
79
+ "Slice",
80
+ "Concat",
81
+ "MatMul",
82
+ "Div",
83
+ "LessOrEqual",
84
+ "Shape",
85
+ "Log",
86
+ "Min",
87
+ "Reshape",
88
+ "Tanh",
89
+ "Tile",
90
+ "Add",
91
+ "Cast",
92
+ "Range",
93
+ "ReduceMean",
94
+ "Where",
95
+ "Sqrt",
96
+ "Unsqueeze",
97
+ "Pow",
98
+ "If",
99
+ "Less",
100
+ "Expand"
101
+ ],
102
+ "weight_type": "QInt8"
103
+ },
104
+ "decoder_with_past_model": {
105
+ "op_types": [
106
+ "Mul",
107
+ "Transpose",
108
+ "Softmax",
109
+ "Gather",
110
+ "Constant",
111
+ "Neg",
112
+ "Sub",
113
+ "ConstantOfShape",
114
+ "Slice",
115
+ "Concat",
116
+ "MatMul",
117
+ "Div",
118
+ "LessOrEqual",
119
+ "Shape",
120
+ "Log",
121
+ "Min",
122
+ "Reshape",
123
+ "Tanh",
124
+ "Tile",
125
+ "Add",
126
+ "Cast",
127
+ "Range",
128
+ "ReduceMean",
129
+ "Where",
130
+ "Sqrt",
131
+ "Unsqueeze",
132
+ "Pow",
133
+ "Less",
134
+ "Expand"
135
+ ],
136
+ "weight_type": "QInt8"
137
+ }
138
+ }
139
+ }
tokenizer_config.json CHANGED
@@ -104,9 +104,13 @@
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": {},
 
110
  "tokenizer_class": "T5Tokenizer",
 
 
111
  "unk_token": "<unk>"
112
  }
 
104
  "clean_up_tokenization_spaces": true,
105
  "eos_token": "</s>",
106
  "extra_ids": 100,
107
+ "max_length": 512,
108
  "model_max_length": 512,
109
  "pad_token": "<pad>",
110
  "sp_model_kwargs": {},
111
+ "stride": 0,
112
  "tokenizer_class": "T5Tokenizer",
113
+ "truncation_side": "right",
114
+ "truncation_strategy": "longest_first",
115
  "unk_token": "<unk>"
116
  }