Hugging Face's logo

Xenova
/
LaMini-Flan-T5-248M

Text2Text Generation
Transformers.js
ONNX
t5
Model card Files Community
Xenova HF staff commited on
Commit
7a42f28
•
1 Parent(s): 53dfcec

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": 2048,
7
  "d_kv": 64,
8
  "d_model": 768,
@@ -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": 2048,
8
  "d_kv": 64,
9
  "d_model": 768,
 
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:f7c7794543e29d2314d621b0ed7e9e56562d3d11fb2d0b049733c6882ea70541
3
- size 650849270
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ad02c232909c5e03e6c835780436b619915771ba0d65789071f2e43941f30897
3
+ size 650848961
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e3279511f470fa1b06365196edae2dcbda13f726e394aceab3d10942c1972423
3
- size 651183578
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f806a9a5d2689d8eaddebfb1d8c4eb9b0e6a8f4611044644912c5c6d8ff1e1b
3
+ size 651182887
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7188c2cc9557589fe50f949b0f1465a646792ee5a56fbe44d7c66b0c54bb1776
3
- size 163918689
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56bc844856b57fc0cdda8c112b303211a950ef64d76a4815cfc718c175d27498
3
+ size 164739632
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f59b07525e63713429821704789a78bd3430751463b0bab1a9b7246b8bb4027e
3
- size 163341718
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45a74d30a328f044cb213ed2ead9f5d48277e983f5182cf05b7988bd958654e0
3
+ size 164163118
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ed1dfec55c0ef2f24c3e038581192b5448c60bd460168f13d184e2e7823b59e9
3
- size 594197616
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f3fb070cbda801fe92a4a03d29d369eef216bf96108f64abc3d29ea27703789e
3
+ size 594197309
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f39b55dbb48da2a207e9575b6a9fcf459c711a3a544d9a32c983639084741991
3
- size 149125021
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7cc6b0ebc2289860c769f6295910c0afbe2dbf5d3a578319441472107dcf3fd
3
+ size 149854152
onnx/encoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ab30469eb23c5e2a4e5df6024d30e01d777833d9f615143fceeea31c8cf0d6f5
3
  size 438697388
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5bda698ea94b30e7d9fb417e0df3641243087f7812118adea9022e0260f85ce6
3
  size 438697388
onnx/encoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1e0a575d82b93b1ed6ae5e7725e0d734aa1489968594450a8886b6695e1281b4
3
- size 110025778
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4203b9de03bd3498215bc9f0a705fca1cb218a807a67983adff0c33ad7ef0571
3
+ size 110502358
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
+ "Pow",
8
+ "Range",
9
+ "Less",
10
+ "Log",
11
+ "Constant",
12
+ "Softmax",
13
+ "ReduceMean",
14
+ "Shape",
15
+ "Gather",
16
+ "Cast",
17
+ "Where",
18
+ "Mul",
19
+ "Tile",
20
+ "Concat",
21
+ "Div",
22
+ "Neg",
23
+ "Sqrt",
24
+ "Transpose",
25
+ "LessOrEqual",
26
+ "MatMul",
27
+ "Expand",
28
+ "ConstantOfShape",
29
+ "Sub",
30
+ "Tanh",
31
+ "Reshape",
32
+ "Add",
33
+ "Unsqueeze",
34
+ "Min"
35
+ ],
36
+ "weight_type": "QInt8"
37
+ },
38
+ "encoder_model": {
39
+ "op_types": [
40
+ "Pow",
41
+ "Range",
42
+ "Less",
43
+ "Abs",
44
+ "Log",
45
+ "Constant",
46
+ "Softmax",
47
+ "ReduceMean",
48
+ "Shape",
49
+ "Gather",
50
+ "Cast",
51
+ "Where",
52
+ "Mul",
53
+ "Concat",
54
+ "Div",
55
+ "Sqrt",
56
+ "Transpose",
57
+ "MatMul",
58
+ "Greater",
59
+ "ConstantOfShape",
60
+ "Sub",
61
+ "Tanh",
62
+ "Reshape",
63
+ "Add",
64
+ "Unsqueeze",
65
+ "Min"
66
+ ],
67
+ "weight_type": "QInt8"
68
+ },
69
+ "decoder_model_merged": {
70
+ "op_types": [
71
+ "Pow",
72
+ "Range",
73
+ "Less",
74
+ "Log",
75
+ "Constant",
76
+ "Softmax",
77
+ "ReduceMean",
78
+ "Shape",
79
+ "Gather",
80
+ "Cast",
81
+ "Where",
82
+ "Mul",
83
+ "Tile",
84
+ "Div",
85
+ "Unsqueeze",
86
+ "Neg",
87
+ "Sqrt",
88
+ "Slice",
89
+ "Transpose",
90
+ "LessOrEqual",
91
+ "MatMul",
92
+ "Expand",
93
+ "If",
94
+ "ConstantOfShape",
95
+ "Sub",
96
+ "Tanh",
97
+ "Reshape",
98
+ "Add",
99
+ "Concat",
100
+ "Min"
101
+ ],
102
+ "weight_type": "QInt8"
103
+ },
104
+ "decoder_with_past_model": {
105
+ "op_types": [
106
+ "Pow",
107
+ "Range",
108
+ "Less",
109
+ "Log",
110
+ "Constant",
111
+ "Softmax",
112
+ "ReduceMean",
113
+ "Shape",
114
+ "Gather",
115
+ "Cast",
116
+ "Where",
117
+ "Mul",
118
+ "Tile",
119
+ "Concat",
120
+ "Div",
121
+ "Neg",
122
+ "Sqrt",
123
+ "Slice",
124
+ "Transpose",
125
+ "LessOrEqual",
126
+ "MatMul",
127
+ "Expand",
128
+ "ConstantOfShape",
129
+ "Sub",
130
+ "Tanh",
131
+ "Reshape",
132
+ "Add",
133
+ "Unsqueeze",
134
+ "Min"
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
  }