Hugging Face's logo

Shivam098
/
opt-translation

Text2Text Generation
Transformers
PyTorch
Safetensors
mbart
Generated from Trainer
Inference Endpoints
4-bit precision
gptq
Model card Files Community
1
Shivam098 commited on
Commit
18ec7db
•
1 Parent(s): 54693af

Upload MBartForConditionalGeneration

Browse files
Files changed (3) hide show
  1. config.json +220 -0
  2. generation_config.json +12 -0
  3. pytorch_model.bin +3 -0
config.json ADDED
@@ -0,0 +1,220 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Shivam098/Translation",
3
+ "_num_labels": 3,
4
+ "activation_dropout": 0.0,
5
+ "activation_function": "relu",
6
+ "add_bias_logits": false,
7
+ "add_final_layer_norm": true,
8
+ "architectures": [
9
+ "MBartForConditionalGeneration"
10
+ ],
11
+ "attention_dropout": 0.0,
12
+ "bos_token_id": 0,
13
+ "classif_dropout": 0.0,
14
+ "classifier_dropout": 0.0,
15
+ "d_model": 1024,
16
+ "decoder_attention_heads": 16,
17
+ "decoder_ffn_dim": 4096,
18
+ "decoder_layerdrop": 0.0,
19
+ "decoder_layers": 12,
20
+ "decoder_start_token_id": 2,
21
+ "dropout": 0.1,
22
+ "early_stopping": true,
23
+ "encoder_attention_heads": 16,
24
+ "encoder_ffn_dim": 4096,
25
+ "encoder_layerdrop": 0.0,
26
+ "encoder_layers": 12,
27
+ "eos_token_id": 2,
28
+ "forced_eos_token_id": 2,
29
+ "gradient_checkpointing": false,
30
+ "id2label": {
31
+ "0": "LABEL_0",
32
+ "1": "LABEL_1",
33
+ "2": "LABEL_2"
34
+ },
35
+ "init_std": 0.02,
36
+ "is_encoder_decoder": true,
37
+ "label2id": {
38
+ "LABEL_0": 0,
39
+ "LABEL_1": 1,
40
+ "LABEL_2": 2
41
+ },
42
+ "max_length": 200,
43
+ "max_position_embeddings": 1024,
44
+ "model_type": "mbart",
45
+ "normalize_before": true,
46
+ "normalize_embedding": true,
47
+ "num_beams": 5,
48
+ "num_hidden_layers": 12,
49
+ "output_past": true,
50
+ "pad_token_id": 1,
51
+ "quantization_config": {
52
+ "batch_size": 1,
53
+ "bits": 4,
54
+ "block_name_to_quantize": "model.decoder.layers",
55
+ "damp_percent": 0.1,
56
+ "dataset": [
57
+ "auto-gptq is an easy-to-use model quantization library with user-friendly apis, based on GPTQ algorithm."
58
+ ],
59
+ "desc_act": false,
60
+ "disable_exllama": false,
61
+ "group_size": 128,
62
+ "model_seqlen": 1024,
63
+ "module_name_preceding_first_block": [
64
+ "model.shared",
65
+ "model.encoder.embed_tokens",
66
+ "model.encoder.embed_positions",
67
+ "model.encoder.layers.0.self_attn.k_proj",
68
+ "model.encoder.layers.0.self_attn.v_proj",
69
+ "model.encoder.layers.0.self_attn.q_proj",
70
+ "model.encoder.layers.0.self_attn.out_proj",
71
+ "model.encoder.layers.0.self_attn",
72
+ "model.encoder.layers.0.self_attn_layer_norm",
73
+ "model.encoder.layers.0.activation_fn",
74
+ "model.encoder.layers.0.fc1",
75
+ "model.encoder.layers.0.fc2",
76
+ "model.encoder.layers.0.final_layer_norm",
77
+ "model.encoder.layers.0",
78
+ "model.encoder.layers.1.self_attn.k_proj",
79
+ "model.encoder.layers.1.self_attn.v_proj",
80
+ "model.encoder.layers.1.self_attn.q_proj",
81
+ "model.encoder.layers.1.self_attn.out_proj",
82
+ "model.encoder.layers.1.self_attn",
83
+ "model.encoder.layers.1.self_attn_layer_norm",
84
+ "model.encoder.layers.1.activation_fn",
85
+ "model.encoder.layers.1.fc1",
86
+ "model.encoder.layers.1.fc2",
87
+ "model.encoder.layers.1.final_layer_norm",
88
+ "model.encoder.layers.1",
89
+ "model.encoder.layers.2.self_attn.k_proj",
90
+ "model.encoder.layers.2.self_attn.v_proj",
91
+ "model.encoder.layers.2.self_attn.q_proj",
92
+ "model.encoder.layers.2.self_attn.out_proj",
93
+ "model.encoder.layers.2.self_attn",
94
+ "model.encoder.layers.2.self_attn_layer_norm",
95
+ "model.encoder.layers.2.activation_fn",
96
+ "model.encoder.layers.2.fc1",
97
+ "model.encoder.layers.2.fc2",
98
+ "model.encoder.layers.2.final_layer_norm",
99
+ "model.encoder.layers.2",
100
+ "model.encoder.layers.3.self_attn.k_proj",
101
+ "model.encoder.layers.3.self_attn.v_proj",
102
+ "model.encoder.layers.3.self_attn.q_proj",
103
+ "model.encoder.layers.3.self_attn.out_proj",
104
+ "model.encoder.layers.3.self_attn",
105
+ "model.encoder.layers.3.self_attn_layer_norm",
106
+ "model.encoder.layers.3.activation_fn",
107
+ "model.encoder.layers.3.fc1",
108
+ "model.encoder.layers.3.fc2",
109
+ "model.encoder.layers.3.final_layer_norm",
110
+ "model.encoder.layers.3",
111
+ "model.encoder.layers.4.self_attn.k_proj",
112
+ "model.encoder.layers.4.self_attn.v_proj",
113
+ "model.encoder.layers.4.self_attn.q_proj",
114
+ "model.encoder.layers.4.self_attn.out_proj",
115
+ "model.encoder.layers.4.self_attn",
116
+ "model.encoder.layers.4.self_attn_layer_norm",
117
+ "model.encoder.layers.4.activation_fn",
118
+ "model.encoder.layers.4.fc1",
119
+ "model.encoder.layers.4.fc2",
120
+ "model.encoder.layers.4.final_layer_norm",
121
+ "model.encoder.layers.4",
122
+ "model.encoder.layers.5.self_attn.k_proj",
123
+ "model.encoder.layers.5.self_attn.v_proj",
124
+ "model.encoder.layers.5.self_attn.q_proj",
125
+ "model.encoder.layers.5.self_attn.out_proj",
126
+ "model.encoder.layers.5.self_attn",
127
+ "model.encoder.layers.5.self_attn_layer_norm",
128
+ "model.encoder.layers.5.activation_fn",
129
+ "model.encoder.layers.5.fc1",
130
+ "model.encoder.layers.5.fc2",
131
+ "model.encoder.layers.5.final_layer_norm",
132
+ "model.encoder.layers.5",
133
+ "model.encoder.layers.6.self_attn.k_proj",
134
+ "model.encoder.layers.6.self_attn.v_proj",
135
+ "model.encoder.layers.6.self_attn.q_proj",
136
+ "model.encoder.layers.6.self_attn.out_proj",
137
+ "model.encoder.layers.6.self_attn",
138
+ "model.encoder.layers.6.self_attn_layer_norm",
139
+ "model.encoder.layers.6.activation_fn",
140
+ "model.encoder.layers.6.fc1",
141
+ "model.encoder.layers.6.fc2",
142
+ "model.encoder.layers.6.final_layer_norm",
143
+ "model.encoder.layers.6",
144
+ "model.encoder.layers.7.self_attn.k_proj",
145
+ "model.encoder.layers.7.self_attn.v_proj",
146
+ "model.encoder.layers.7.self_attn.q_proj",
147
+ "model.encoder.layers.7.self_attn.out_proj",
148
+ "model.encoder.layers.7.self_attn",
149
+ "model.encoder.layers.7.self_attn_layer_norm",
150
+ "model.encoder.layers.7.activation_fn",
151
+ "model.encoder.layers.7.fc1",
152
+ "model.encoder.layers.7.fc2",
153
+ "model.encoder.layers.7.final_layer_norm",
154
+ "model.encoder.layers.7",
155
+ "model.encoder.layers.8.self_attn.k_proj",
156
+ "model.encoder.layers.8.self_attn.v_proj",
157
+ "model.encoder.layers.8.self_attn.q_proj",
158
+ "model.encoder.layers.8.self_attn.out_proj",
159
+ "model.encoder.layers.8.self_attn",
160
+ "model.encoder.layers.8.self_attn_layer_norm",
161
+ "model.encoder.layers.8.activation_fn",
162
+ "model.encoder.layers.8.fc1",
163
+ "model.encoder.layers.8.fc2",
164
+ "model.encoder.layers.8.final_layer_norm",
165
+ "model.encoder.layers.8",
166
+ "model.encoder.layers.9.self_attn.k_proj",
167
+ "model.encoder.layers.9.self_attn.v_proj",
168
+ "model.encoder.layers.9.self_attn.q_proj",
169
+ "model.encoder.layers.9.self_attn.out_proj",
170
+ "model.encoder.layers.9.self_attn",
171
+ "model.encoder.layers.9.self_attn_layer_norm",
172
+ "model.encoder.layers.9.activation_fn",
173
+ "model.encoder.layers.9.fc1",
174
+ "model.encoder.layers.9.fc2",
175
+ "model.encoder.layers.9.final_layer_norm",
176
+ "model.encoder.layers.9",
177
+ "model.encoder.layers.10.self_attn.k_proj",
178
+ "model.encoder.layers.10.self_attn.v_proj",
179
+ "model.encoder.layers.10.self_attn.q_proj",
180
+ "model.encoder.layers.10.self_attn.out_proj",
181
+ "model.encoder.layers.10.self_attn",
182
+ "model.encoder.layers.10.self_attn_layer_norm",
183
+ "model.encoder.layers.10.activation_fn",
184
+ "model.encoder.layers.10.fc1",
185
+ "model.encoder.layers.10.fc2",
186
+ "model.encoder.layers.10.final_layer_norm",
187
+ "model.encoder.layers.10",
188
+ "model.encoder.layers.11.self_attn.k_proj",
189
+ "model.encoder.layers.11.self_attn.v_proj",
190
+ "model.encoder.layers.11.self_attn.q_proj",
191
+ "model.encoder.layers.11.self_attn.out_proj",
192
+ "model.encoder.layers.11.self_attn",
193
+ "model.encoder.layers.11.self_attn_layer_norm",
194
+ "model.encoder.layers.11.activation_fn",
195
+ "model.encoder.layers.11.fc1",
196
+ "model.encoder.layers.11.fc2",
197
+ "model.encoder.layers.11.final_layer_norm",
198
+ "model.encoder.layers.11",
199
+ "model.encoder.layers",
200
+ "model.encoder.layernorm_embedding",
201
+ "model.encoder.layer_norm",
202
+ "model.encoder",
203
+ "model.decoder.embed_tokens",
204
+ "model.decoder.embed_positions"
205
+ ],
206
+ "pad_token_id": null,
207
+ "quant_method": "gptq",
208
+ "sym": true,
209
+ "tokenizer": null,
210
+ "true_sequential": true,
211
+ "use_cuda_fp16": true
212
+ },
213
+ "scale_embedding": true,
214
+ "static_position_embeddings": false,
215
+ "tokenizer_class": "MBart50Tokenizer",
216
+ "torch_dtype": "float16",
217
+ "transformers_version": "4.33.0.dev0",
218
+ "use_cache": true,
219
+ "vocab_size": 250054
220
+ }
generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "decoder_start_token_id": 2,
5
+ "early_stopping": true,
6
+ "eos_token_id": 2,
7
+ "forced_eos_token_id": 2,
8
+ "max_length": 200,
9
+ "num_beams": 5,
10
+ "pad_token_id": 1,
11
+ "transformers_version": "4.33.0.dev0"
12
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87a19dcde73bf510b5d841460dc0b5c014107630406c5cce30a8c3758ac18923
3
+ size 1949361861