Hugging Face's logo

Xenova
/
blenderbot_small-90M

Text2Text Generation
Transformers.js
ONNX
blenderbot-small
Model card Files Community
Xenova HF staff commited on
Commit
a890c69
•
1 Parent(s): d03add6

Upload folder using huggingface_hub

Browse files
Files changed (15) hide show
  1. config.json +61 -0
  2. generation_config.json +14 -0
  3. merges.txt +0 -0
  4. onnx/decoder_model.onnx +3 -0
  5. onnx/decoder_model_merged.onnx +3 -0
  6. onnx/decoder_model_merged_quantized.onnx +3 -0
  7. onnx/decoder_model_quantized.onnx +3 -0
  8. onnx/decoder_with_past_model.onnx +3 -0
  9. onnx/decoder_with_past_model_quantized.onnx +3 -0
  10. onnx/encoder_model.onnx +3 -0
  11. onnx/encoder_model_quantized.onnx +3 -0
  12. quantize_config.json +125 -0
  13. special_tokens_map.json +6 -0
  14. tokenizer_config.json +10 -0
  15. vocab.json +0 -0
config.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/blenderbot_small-90M",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "gelu",
5
+ "add_bias_logits": false,
6
+ "add_final_layer_norm": false,
7
+ "architectures": [
8
+ "BlenderbotSmallForConditionalGeneration"
9
+ ],
10
+ "attention_dropout": 0.0,
11
+ "bos_token_id": 1,
12
+ "classif_dropout": 0.0,
13
+ "classifier_dropout": 0.0,
14
+ "d_model": 512,
15
+ "decoder_attention_heads": 16,
16
+ "decoder_ffn_dim": 2048,
17
+ "decoder_layerdrop": 0.0,
18
+ "decoder_layers": 8,
19
+ "decoder_start_token_id": 1,
20
+ "do_blenderbot_90_layernorm": true,
21
+ "dropout": 0.1,
22
+ "encoder_attention_heads": 16,
23
+ "encoder_ffn_dim": 2048,
24
+ "encoder_layerdrop": 0.0,
25
+ "encoder_layers": 8,
26
+ "eos_token_id": 2,
27
+ "extra_pos_embeddings": 0,
28
+ "force_bos_token_to_be_generated": false,
29
+ "forced_eos_token_id": 2,
30
+ "gradient_checkpointing": false,
31
+ "id2label": {
32
+ "0": "LABEL_0",
33
+ "1": "LABEL_1",
34
+ "2": "LABEL_2"
35
+ },
36
+ "init_std": 0.02,
37
+ "is_encoder_decoder": true,
38
+ "label2id": {
39
+ "LABEL_0": 0,
40
+ "LABEL_1": 1,
41
+ "LABEL_2": 2
42
+ },
43
+ "layernorm_variant": "xlm",
44
+ "length_penalty": 0.65,
45
+ "max_length": 128,
46
+ "max_position_embeddings": 512,
47
+ "min_length": 20,
48
+ "model_type": "blenderbot-small",
49
+ "no_repeat_ngram_size": 3,
50
+ "normalize_before": false,
51
+ "normalize_embedding": true,
52
+ "num_beams": 10,
53
+ "num_hidden_layers": 8,
54
+ "pad_token_id": 0,
55
+ "scale_embedding": true,
56
+ "static_position_embeddings": false,
57
+ "transformers_version": "4.34.0.dev0",
58
+ "unk_token_id": 3,
59
+ "use_cache": true,
60
+ "vocab_size": 54944
61
+ }
generation_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "decoder_start_token_id": 1,
5
+ "eos_token_id": 2,
6
+ "forced_eos_token_id": 2,
7
+ "length_penalty": 0.65,
8
+ "max_length": 128,
9
+ "min_length": 20,
10
+ "no_repeat_ngram_size": 3,
11
+ "num_beams": 10,
12
+ "pad_token_id": 0,
13
+ "transformers_version": "4.34.0.dev0"
14
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
onnx/decoder_model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:90e07af04bec03120915340d3cf2e9edd4c2d9b25737e49d67e6e39bf3904826
3
+ size 248701447
onnx/decoder_model_merged.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e7385d8a799449515b71ac64ed0f9e4c508430299e363f838ea026252a913977
3
+ size 249023782
onnx/decoder_model_merged_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:908938426e6ae33ba9f53788b2af6b1ae4d60a2c80e062aa29c9d977e1bab3d6
3
+ size 63849989
onnx/decoder_model_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3dee3b0cd7d343a77b420c3c8f16761c58547ad5386c5101bd6b3fefe838904
3
+ size 63344767
onnx/decoder_with_past_model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:75500f107b2ba0ec8d90767a0860090dca70e855c0e045aeb069996895d59d15
3
+ size 231836114
onnx/decoder_with_past_model_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:12a7a8e571e7997af8998e0309d784ec48e24e2af3fa854ecabf69c9f9dba507
3
+ size 58985639
onnx/encoder_model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4fd62f4847aa765fc131f1693af7632f64c5d517d466afd65ff6b321743b72e
3
+ size 214608300
onnx/encoder_model_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:442034c7372de5077289ace18124bf315e32fb4a8d3b5c13f29f71397bd7b161
3
+ size 54216554
quantize_config.json ADDED
@@ -0,0 +1,125 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "per_channel": true,
3
+ "reduce_range": true,
4
+ "per_model_config": {
5
+ "encoder_model": {
6
+ "op_types": [
7
+ "Unsqueeze",
8
+ "Sqrt",
9
+ "Expand",
10
+ "Erf",
11
+ "Sub",
12
+ "Div",
13
+ "Cast",
14
+ "MatMul",
15
+ "Where",
16
+ "Reshape",
17
+ "Gather",
18
+ "Constant",
19
+ "ConstantOfShape",
20
+ "Mul",
21
+ "Add",
22
+ "ReduceMean",
23
+ "Softmax",
24
+ "Shape",
25
+ "Concat",
26
+ "Range",
27
+ "Equal",
28
+ "Pow",
29
+ "Transpose"
30
+ ],
31
+ "weight_type": "QInt8"
32
+ },
33
+ "decoder_model_merged": {
34
+ "op_types": [
35
+ "Unsqueeze",
36
+ "Sqrt",
37
+ "Expand",
38
+ "Erf",
39
+ "Squeeze",
40
+ "Sub",
41
+ "Div",
42
+ "Cast",
43
+ "MatMul",
44
+ "Where",
45
+ "Reshape",
46
+ "Gather",
47
+ "Constant",
48
+ "ConstantOfShape",
49
+ "Mul",
50
+ "Add",
51
+ "ReduceMean",
52
+ "Softmax",
53
+ "Less",
54
+ "Shape",
55
+ "Concat",
56
+ "Range",
57
+ "Equal",
58
+ "If",
59
+ "Slice",
60
+ "Pow",
61
+ "Transpose"
62
+ ],
63
+ "weight_type": "QInt8"
64
+ },
65
+ "decoder_with_past_model": {
66
+ "op_types": [
67
+ "Unsqueeze",
68
+ "Sqrt",
69
+ "Expand",
70
+ "Erf",
71
+ "Sub",
72
+ "Div",
73
+ "Cast",
74
+ "MatMul",
75
+ "Where",
76
+ "Reshape",
77
+ "Gather",
78
+ "Constant",
79
+ "ConstantOfShape",
80
+ "Mul",
81
+ "Add",
82
+ "ReduceMean",
83
+ "Softmax",
84
+ "Shape",
85
+ "Concat",
86
+ "Equal",
87
+ "Range",
88
+ "Pow",
89
+ "Transpose"
90
+ ],
91
+ "weight_type": "QInt8"
92
+ },
93
+ "decoder_model": {
94
+ "op_types": [
95
+ "Unsqueeze",
96
+ "Sqrt",
97
+ "Expand",
98
+ "Erf",
99
+ "Squeeze",
100
+ "Sub",
101
+ "Div",
102
+ "Cast",
103
+ "MatMul",
104
+ "Where",
105
+ "Reshape",
106
+ "Gather",
107
+ "Constant",
108
+ "ConstantOfShape",
109
+ "Mul",
110
+ "Add",
111
+ "ReduceMean",
112
+ "Softmax",
113
+ "Less",
114
+ "Shape",
115
+ "Concat",
116
+ "Range",
117
+ "Equal",
118
+ "Slice",
119
+ "Pow",
120
+ "Transpose"
121
+ ],
122
+ "weight_type": "QInt8"
123
+ }
124
+ }
125
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "__start__",
3
+ "eos_token": "__end__",
4
+ "pad_token": "__null__",
5
+ "unk_token": "__unk__"
6
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "__start__",
3
+ "clean_up_tokenization_spaces": true,
4
+ "eos_token": "__end__",
5
+ "model_max_length": 512,
6
+ "pad_token": "__null__",
7
+ "tokenizer_class": "BlenderbotSmallTokenizer",
8
+ "tokenizer_file": null,
9
+ "unk_token": "__unk__"
10
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff