mwz commited on
Commit
c990746
1 Parent(s): 3cc409c

Upload 2 files

Browse files
Files changed (2) hide show
  1. config.json +187 -0
  2. generation_config.json +5 -0
config.json ADDED
@@ -0,0 +1,187 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_commit_hash": null,
3
+ "architectures": [
4
+ "EncoderDecoderModel"
5
+ ],
6
+ "decoder": {
7
+ "_name_or_path": "bert-base-multilingual-uncased",
8
+ "add_cross_attention": true,
9
+ "architectures": [
10
+ "BertForMaskedLM"
11
+ ],
12
+ "attention_probs_dropout_prob": 0.1,
13
+ "bad_words_ids": null,
14
+ "begin_suppress_tokens": null,
15
+ "bos_token_id": null,
16
+ "chunk_size_feed_forward": 0,
17
+ "classifier_dropout": null,
18
+ "cross_attention_hidden_size": null,
19
+ "decoder_start_token_id": null,
20
+ "directionality": "bidi",
21
+ "diversity_penalty": 0.0,
22
+ "do_sample": false,
23
+ "early_stopping": false,
24
+ "encoder_no_repeat_ngram_size": 0,
25
+ "eos_token_id": null,
26
+ "exponential_decay_length_penalty": null,
27
+ "finetuning_task": null,
28
+ "forced_bos_token_id": null,
29
+ "forced_eos_token_id": null,
30
+ "hidden_act": "gelu",
31
+ "hidden_dropout_prob": 0.1,
32
+ "hidden_size": 768,
33
+ "id2label": {
34
+ "0": "LABEL_0",
35
+ "1": "LABEL_1"
36
+ },
37
+ "initializer_range": 0.02,
38
+ "intermediate_size": 3072,
39
+ "is_decoder": true,
40
+ "is_encoder_decoder": false,
41
+ "label2id": {
42
+ "LABEL_0": 0,
43
+ "LABEL_1": 1
44
+ },
45
+ "layer_norm_eps": 1e-12,
46
+ "length_penalty": 1.0,
47
+ "max_length": 20,
48
+ "max_position_embeddings": 512,
49
+ "min_length": 0,
50
+ "model_type": "bert",
51
+ "no_repeat_ngram_size": 0,
52
+ "num_attention_heads": 12,
53
+ "num_beam_groups": 1,
54
+ "num_beams": 1,
55
+ "num_hidden_layers": 12,
56
+ "num_return_sequences": 1,
57
+ "output_attentions": false,
58
+ "output_hidden_states": false,
59
+ "output_scores": false,
60
+ "pad_token_id": 0,
61
+ "pooler_fc_size": 768,
62
+ "pooler_num_attention_heads": 12,
63
+ "pooler_num_fc_layers": 3,
64
+ "pooler_size_per_head": 128,
65
+ "pooler_type": "first_token_transform",
66
+ "position_embedding_type": "absolute",
67
+ "prefix": null,
68
+ "problem_type": null,
69
+ "pruned_heads": {},
70
+ "remove_invalid_values": false,
71
+ "repetition_penalty": 1.0,
72
+ "return_dict": true,
73
+ "return_dict_in_generate": false,
74
+ "sep_token_id": null,
75
+ "suppress_tokens": null,
76
+ "task_specific_params": null,
77
+ "temperature": 1.0,
78
+ "tf_legacy_loss": false,
79
+ "tie_encoder_decoder": false,
80
+ "tie_word_embeddings": true,
81
+ "tokenizer_class": null,
82
+ "top_k": 50,
83
+ "top_p": 1.0,
84
+ "torch_dtype": null,
85
+ "torchscript": false,
86
+ "transformers_version": "4.28.1",
87
+ "type_vocab_size": 2,
88
+ "typical_p": 1.0,
89
+ "use_bfloat16": false,
90
+ "use_cache": true,
91
+ "vocab_size": 105879
92
+ },
93
+ "decoder_start_token_id": 101,
94
+ "encoder": {
95
+ "_name_or_path": "bert-base-multilingual-uncased",
96
+ "add_cross_attention": false,
97
+ "architectures": [
98
+ "BertForMaskedLM"
99
+ ],
100
+ "attention_probs_dropout_prob": 0.1,
101
+ "bad_words_ids": null,
102
+ "begin_suppress_tokens": null,
103
+ "bos_token_id": null,
104
+ "chunk_size_feed_forward": 0,
105
+ "classifier_dropout": null,
106
+ "cross_attention_hidden_size": null,
107
+ "decoder_start_token_id": null,
108
+ "directionality": "bidi",
109
+ "diversity_penalty": 0.0,
110
+ "do_sample": false,
111
+ "early_stopping": false,
112
+ "encoder_no_repeat_ngram_size": 0,
113
+ "eos_token_id": null,
114
+ "exponential_decay_length_penalty": null,
115
+ "finetuning_task": null,
116
+ "forced_bos_token_id": null,
117
+ "forced_eos_token_id": null,
118
+ "hidden_act": "gelu",
119
+ "hidden_dropout_prob": 0.1,
120
+ "hidden_size": 768,
121
+ "id2label": {
122
+ "0": "LABEL_0",
123
+ "1": "LABEL_1"
124
+ },
125
+ "initializer_range": 0.02,
126
+ "intermediate_size": 3072,
127
+ "is_decoder": false,
128
+ "is_encoder_decoder": false,
129
+ "label2id": {
130
+ "LABEL_0": 0,
131
+ "LABEL_1": 1
132
+ },
133
+ "layer_norm_eps": 1e-12,
134
+ "length_penalty": 1.0,
135
+ "max_length": 20,
136
+ "max_position_embeddings": 512,
137
+ "min_length": 0,
138
+ "model_type": "bert",
139
+ "no_repeat_ngram_size": 0,
140
+ "num_attention_heads": 12,
141
+ "num_beam_groups": 1,
142
+ "num_beams": 1,
143
+ "num_hidden_layers": 12,
144
+ "num_return_sequences": 1,
145
+ "output_attentions": false,
146
+ "output_hidden_states": false,
147
+ "output_scores": false,
148
+ "pad_token_id": 0,
149
+ "pooler_fc_size": 768,
150
+ "pooler_num_attention_heads": 12,
151
+ "pooler_num_fc_layers": 3,
152
+ "pooler_size_per_head": 128,
153
+ "pooler_type": "first_token_transform",
154
+ "position_embedding_type": "absolute",
155
+ "prefix": null,
156
+ "problem_type": null,
157
+ "pruned_heads": {},
158
+ "remove_invalid_values": false,
159
+ "repetition_penalty": 1.0,
160
+ "return_dict": true,
161
+ "return_dict_in_generate": false,
162
+ "sep_token_id": null,
163
+ "suppress_tokens": null,
164
+ "task_specific_params": null,
165
+ "temperature": 1.0,
166
+ "tf_legacy_loss": false,
167
+ "tie_encoder_decoder": false,
168
+ "tie_word_embeddings": true,
169
+ "tokenizer_class": null,
170
+ "top_k": 50,
171
+ "top_p": 1.0,
172
+ "torch_dtype": null,
173
+ "torchscript": false,
174
+ "transformers_version": "4.28.1",
175
+ "type_vocab_size": 2,
176
+ "typical_p": 1.0,
177
+ "use_bfloat16": false,
178
+ "use_cache": true,
179
+ "vocab_size": 105879
180
+ },
181
+ "eos_token_id": 102,
182
+ "is_encoder_decoder": true,
183
+ "model_type": "encoder-decoder",
184
+ "pad_token_id": 0,
185
+ "torch_dtype": "float32",
186
+ "transformers_version": null
187
+ }
generation_config.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "pad_token_id": 0,
4
+ "transformers_version": "4.28.1"
5
+ }