system HF staff commited on
Commit
c63fa6e
1 Parent(s): e91c67e

Update config.json

Browse files
Files changed (1) hide show
  1. config.json +136 -0
config.json ADDED
@@ -0,0 +1,136 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "EncoderDecoderModel"
4
+ ],
5
+ "decoder": {
6
+ "add_cross_attention": true,
7
+ "architectures": [
8
+ "BertForSeqGenerationDecoder"
9
+ ],
10
+ "attention_probs_dropout_prob": 0.1,
11
+ "bad_words_ids": null,
12
+ "bos_token_id": null,
13
+ "chunk_size_feed_forward": 0,
14
+ "decoder_start_token_id": null,
15
+ "directionality": "bidi",
16
+ "do_sample": false,
17
+ "early_stopping": false,
18
+ "eos_token_id": null,
19
+ "finetuning_task": null,
20
+ "gradient_checkpointing": false,
21
+ "hidden_act": "gelu",
22
+ "hidden_dropout_prob": 0.1,
23
+ "hidden_size": 1024,
24
+ "id2label": {
25
+ "0": "LABEL_0",
26
+ "1": "LABEL_1"
27
+ },
28
+ "initializer_range": 0.02,
29
+ "intermediate_size": 4096,
30
+ "is_decoder": true,
31
+ "is_encoder_decoder": false,
32
+ "label2id": {
33
+ "LABEL_0": 0,
34
+ "LABEL_1": 1
35
+ },
36
+ "layer_norm_eps": 1e-12,
37
+ "length_penalty": 1.0,
38
+ "max_length": 20,
39
+ "max_position_embeddings": 512,
40
+ "min_length": 0,
41
+ "model_type": "bert-for-seq-generation",
42
+ "no_repeat_ngram_size": 0,
43
+ "num_attention_heads": 16,
44
+ "num_beams": 1,
45
+ "num_hidden_layers": 24,
46
+ "num_return_sequences": 1,
47
+ "output_attentions": false,
48
+ "output_hidden_states": false,
49
+ "pad_token_id": 0,
50
+ "prefix": null,
51
+ "pruned_heads": {},
52
+ "repetition_penalty": 1.0,
53
+ "return_dict": false,
54
+ "task_specific_params": null,
55
+ "temperature": 1.0,
56
+ "tie_encoder_decoder": false,
57
+ "tie_word_embeddings": true,
58
+ "top_k": 50,
59
+ "top_p": 1.0,
60
+ "torchscript": false,
61
+ "use_bfloat16": false,
62
+ "use_cache": true,
63
+ "vocab_size": 50358,
64
+ "xla_device": null
65
+ },
66
+ "encoder": {
67
+ "add_cross_attention": false,
68
+ "architectures": [
69
+ "BertForSeqGenerationDecoder"
70
+ ],
71
+ "attention_probs_dropout_prob": 0.1,
72
+ "bad_words_ids": null,
73
+ "bos_token_id": null,
74
+ "chunk_size_feed_forward": 0,
75
+ "decoder_start_token_id": null,
76
+ "directionality": "bidi",
77
+ "do_sample": false,
78
+ "early_stopping": false,
79
+ "eos_token_id": null,
80
+ "finetuning_task": null,
81
+ "gradient_checkpointing": false,
82
+ "hidden_act": "gelu",
83
+ "hidden_dropout_prob": 0.1,
84
+ "hidden_size": 1024,
85
+ "id2label": {
86
+ "0": "LABEL_0",
87
+ "1": "LABEL_1"
88
+ },
89
+ "initializer_range": 0.02,
90
+ "intermediate_size": 4096,
91
+ "is_decoder": false,
92
+ "is_encoder_decoder": false,
93
+ "label2id": {
94
+ "LABEL_0": 0,
95
+ "LABEL_1": 1
96
+ },
97
+ "layer_norm_eps": 1e-12,
98
+ "length_penalty": 1.0,
99
+ "max_length": 20,
100
+ "max_position_embeddings": 512,
101
+ "min_length": 0,
102
+ "model_type": "bert-for-seq-generation",
103
+ "no_repeat_ngram_size": 0,
104
+ "num_attention_heads": 16,
105
+ "num_beams": 1,
106
+ "num_hidden_layers": 24,
107
+ "num_return_sequences": 1,
108
+ "output_attentions": false,
109
+ "output_hidden_states": false,
110
+ "pad_token_id": 0,
111
+ "prefix": null,
112
+ "pruned_heads": {},
113
+ "repetition_penalty": 1.0,
114
+ "return_dict": false,
115
+ "task_specific_params": null,
116
+ "temperature": 1.0,
117
+ "tie_encoder_decoder": false,
118
+ "tie_word_embeddings": true,
119
+ "top_k": 50,
120
+ "top_p": 1.0,
121
+ "torchscript": false,
122
+ "use_bfloat16": false,
123
+ "use_cache": true,
124
+ "vocab_size": 50358,
125
+ "xla_device": null
126
+ },
127
+ "is_encoder_decoder": true,
128
+ "model_type": "encoder_decoder",
129
+ "tie_encoder_decoder": true,
130
+ "bos_token_id": 2,
131
+ "pad_token_id": 0,
132
+ "eos_token_id": 1,
133
+ "decoder_start_token_id": 2,
134
+ "num_beams": 4,
135
+ "max_length": 128
136
+ }