ONNX
t5
rolf-mozilla commited on
Commit
56bb4af
·
unverified ·
1 Parent(s): e658dc5

pious-butterfly-170 remove layers and fine tune from flan-t5-small model

Browse files
config.json CHANGED
@@ -1,29 +1,60 @@
1
  {
2
- "_name_or_path": "dulcet-durian-136",
3
  "architectures": [
4
  "T5ForConditionalGeneration"
5
  ],
6
  "classifier_dropout": 0.0,
7
  "d_ff": 1024,
8
  "d_kv": 64,
9
- "d_model": 256,
10
  "decoder_start_token_id": 0,
11
- "dense_act_fn": "relu",
12
  "dropout_rate": 0.1,
13
  "eos_token_id": 1,
14
- "feed_forward_proj": "relu",
15
  "initializer_factor": 1.0,
16
  "is_encoder_decoder": true,
17
- "is_gated_act": false,
18
  "layer_norm_epsilon": 1e-06,
19
  "model_type": "t5",
20
  "n_positions": 512,
21
- "num_decoder_layers": 4,
22
- "num_heads": 4,
23
- "num_layers": 4,
 
24
  "pad_token_id": 0,
25
  "relative_attention_max_distance": 128,
26
  "relative_attention_num_buckets": 32,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
27
  "transformers_version": "4.43.4",
28
  "use_cache": true,
29
  "vocab_size": 32128
 
1
  {
2
+ "_name_or_path": "pious-butterfly-170",
3
  "architectures": [
4
  "T5ForConditionalGeneration"
5
  ],
6
  "classifier_dropout": 0.0,
7
  "d_ff": 1024,
8
  "d_kv": 64,
9
+ "d_model": 512,
10
  "decoder_start_token_id": 0,
11
+ "dense_act_fn": "gelu_new",
12
  "dropout_rate": 0.1,
13
  "eos_token_id": 1,
14
+ "feed_forward_proj": "gated-gelu",
15
  "initializer_factor": 1.0,
16
  "is_encoder_decoder": true,
17
+ "is_gated_act": true,
18
  "layer_norm_epsilon": 1e-06,
19
  "model_type": "t5",
20
  "n_positions": 512,
21
+ "num_decoder_layers": 3,
22
+ "num_heads": 6,
23
+ "num_layers": 3,
24
+ "output_past": true,
25
  "pad_token_id": 0,
26
  "relative_attention_max_distance": 128,
27
  "relative_attention_num_buckets": 32,
28
+ "task_specific_params": {
29
+ "summarization": {
30
+ "early_stopping": true,
31
+ "length_penalty": 2.0,
32
+ "max_length": 200,
33
+ "min_length": 30,
34
+ "no_repeat_ngram_size": 3,
35
+ "num_beams": 4,
36
+ "prefix": "summarize: "
37
+ },
38
+ "translation_en_to_de": {
39
+ "early_stopping": true,
40
+ "max_length": 300,
41
+ "num_beams": 4,
42
+ "prefix": "translate English to German: "
43
+ },
44
+ "translation_en_to_fr": {
45
+ "early_stopping": true,
46
+ "max_length": 300,
47
+ "num_beams": 4,
48
+ "prefix": "translate English to French: "
49
+ },
50
+ "translation_en_to_ro": {
51
+ "early_stopping": true,
52
+ "max_length": 300,
53
+ "num_beams": 4,
54
+ "prefix": "translate English to Romanian: "
55
+ }
56
+ },
57
+ "tie_word_embeddings": false,
58
  "transformers_version": "4.43.4",
59
  "use_cache": true,
60
  "vocab_size": 32128
generation_config.json CHANGED
@@ -1,5 +1,275 @@
1
  {
2
  "_from_model_config": true,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3
  "decoder_start_token_id": 0,
4
  "eos_token_id": 1,
5
  "pad_token_id": 0,
 
1
  {
2
  "_from_model_config": true,
3
+ "bad_words_ids": [
4
+ [
5
+ 8581
6
+ ],
7
+ [
8
+ 815
9
+ ],
10
+ [
11
+ 1216
12
+ ],
13
+ [
14
+ 3,
15
+ 2781
16
+ ],
17
+ [
18
+ 2886
19
+ ],
20
+ [
21
+ 815
22
+ ],
23
+ [
24
+ 3640
25
+ ],
26
+ [
27
+ 3,
28
+ 4994
29
+ ],
30
+ [
31
+ 1216
32
+ ],
33
+ [
34
+ 5781
35
+ ],
36
+ [
37
+ 1216
38
+ ],
39
+ [
40
+ 2886
41
+ ],
42
+ [
43
+ 3,
44
+ 6855
45
+ ],
46
+ [
47
+ 6949
48
+ ],
49
+ [
50
+ 3,
51
+ 7315
52
+ ],
53
+ [
54
+ 7706
55
+ ],
56
+ [
57
+ 7738
58
+ ],
59
+ [
60
+ 7927
61
+ ],
62
+ [
63
+ 815
64
+ ],
65
+ [
66
+ 8759
67
+ ],
68
+ [
69
+ 3,
70
+ 8578
71
+ ],
72
+ [
73
+ 8581
74
+ ],
75
+ [
76
+ 8759
77
+ ],
78
+ [
79
+ 9529
80
+ ],
81
+ [
82
+ 3,
83
+ 9715
84
+ ],
85
+ [
86
+ 3,
87
+ 10008
88
+ ],
89
+ [
90
+ 10203
91
+ ],
92
+ [
93
+ 10490
94
+ ],
95
+ [
96
+ 5781
97
+ ],
98
+ [
99
+ 11112
100
+ ],
101
+ [
102
+ 3640
103
+ ],
104
+ [
105
+ 11178
106
+ ],
107
+ [
108
+ 7927
109
+ ],
110
+ [
111
+ 2886
112
+ ],
113
+ [
114
+ 3,
115
+ 12030
116
+ ],
117
+ [
118
+ 3,
119
+ 12105
120
+ ],
121
+ [
122
+ 13150
123
+ ],
124
+ [
125
+ 13292
126
+ ],
127
+ [
128
+ 13503
129
+ ],
130
+ [
131
+ 7102,
132
+ 208
133
+ ],
134
+ [
135
+ 13721
136
+ ],
137
+ [
138
+ 3,
139
+ 4994
140
+ ],
141
+ [
142
+ 5781
143
+ ],
144
+ [
145
+ 9858,
146
+ 23
147
+ ],
148
+ [
149
+ 15484
150
+ ],
151
+ [
152
+ 7706
153
+ ],
154
+ [
155
+ 24556
156
+ ],
157
+ [
158
+ 16998
159
+ ],
160
+ [
161
+ 17081
162
+ ],
163
+ [
164
+ 17227
165
+ ],
166
+ [
167
+ 3,
168
+ 2781
169
+ ],
170
+ [
171
+ 28577,
172
+ 23
173
+ ],
174
+ [
175
+ 3,
176
+ 17789
177
+ ],
178
+ [
179
+ 3,
180
+ 17789
181
+ ],
182
+ [
183
+ 3640
184
+ ],
185
+ [
186
+ 18364
187
+ ],
188
+ [
189
+ 18754
190
+ ],
191
+ [
192
+ 3,
193
+ 19230
194
+ ],
195
+ [
196
+ 19329
197
+ ],
198
+ [
199
+ 19981
200
+ ],
201
+ [
202
+ 16998
203
+ ],
204
+ [
205
+ 3,
206
+ 26,
207
+ 3142
208
+ ],
209
+ [
210
+ 3,
211
+ 10008
212
+ ],
213
+ [
214
+ 1560,
215
+ 1171
216
+ ],
217
+ [
218
+ 22658
219
+ ],
220
+ [
221
+ 23568
222
+ ],
223
+ [
224
+ 13292
225
+ ],
226
+ [
227
+ 24556
228
+ ],
229
+ [
230
+ 9529
231
+ ],
232
+ [
233
+ 24974
234
+ ],
235
+ [
236
+ 25039
237
+ ],
238
+ [
239
+ 25164
240
+ ],
241
+ [
242
+ 25547
243
+ ],
244
+ [
245
+ 6949
246
+ ],
247
+ [
248
+ 25723
249
+ ],
250
+ [
251
+ 26113
252
+ ],
253
+ [
254
+ 26806
255
+ ],
256
+ [
257
+ 30320
258
+ ],
259
+ [
260
+ 6949
261
+ ],
262
+ [
263
+ 3,
264
+ 31145
265
+ ],
266
+ [
267
+ 31648
268
+ ],
269
+ [
270
+ 11178
271
+ ]
272
+ ],
273
  "decoder_start_token_id": 0,
274
  "eos_token_id": 1,
275
  "pad_token_id": 0,
onnx/decoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6b3f924580ed1fb94175f1f4520b3f1d723cf626016a5ec2cb206ca3ff3f0bda
3
- size 49810798
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:568444304b19da5c4a8eb3c7282fe53e4c08c57991e5812ea19fbd846e25e9a6
3
+ size 169472577
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9f179daee6bbc93ec07ebcbd4d69f6e51413f293d44fd6a3a45deb3df801aaf7
3
- size 49934579
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9fba6c06c12046fa4b9c7b680f9fc6093e1210bcaa7c345456f335d791dffb50
3
+ size 169578954
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f8b2b3f7c2438219ef677763e088d89f440ae1fbc2649d314d8cf410ab3fcd73
3
- size 45733795
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe83a419e30af7f28066f20c6178956963dae41cd9da67781d03398e18bad42e
3
+ size 42895009
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:eb32c0f150b93b71443445b481465eac48ed36e67da9be2b54cc12598e3e249f
3
- size 45568756
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3aeaa9f321ebd94bf26537f423fe31a817d85d0d48378a1fe0751e18ca4adf6
3
+ size 42753828
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:84be9fe3fd001ecae01a5b866ff46469e4decc715819bcac7af4fbabf6539794
3
- size 47692795
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:40a0ee7a12e84d445f2fd76f9aa779130d77391cf257f1df27a85519c4e814f9
3
+ size 164738287
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c5125b3c19badeb55148498ecc58501dda392b82c64aae2f5717127e1e8f4b7d
3
- size 45005314
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af0ab350e4137be691706d89bb9d5dea92ee871620361481be583afdb2dcd96e
3
+ size 41540983
onnx/encoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:039a737b275da31cfc474020e81514774adf871c44411ee548443554402accf8
3
- size 45553423
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:61c4716849d3d0fe971611531f33a1cef3e7cad9b839cf9cc90112ef6b62002f
3
+ size 94183774
onnx/encoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:927a680f7cb4bbcdd7d718a95c2c658731c27d0f204f02e947910b54ddf6f3fa
3
- size 11515728
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1dffba65048f543e7598826fe6c5a86f1d10cf41b13e1e03420995e3d03cf81b
3
+ size 23688378
quantize_config.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "modes": [
3
- "fp16"
4
  ],
5
  "per_channel": true,
6
  "reduce_range": true,
 
1
  {
2
  "modes": [
3
+ "q8"
4
  ],
5
  "per_channel": true,
6
  "reduce_range": true,
special_tokens_map.json CHANGED
@@ -101,7 +101,25 @@
101
  "<extra_id_98>",
102
  "<extra_id_99>"
103
  ],
104
- "eos_token": "</s>",
105
- "pad_token": "<pad>",
106
- "unk_token": "<unk>"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
107
  }
 
101
  "<extra_id_98>",
102
  "<extra_id_99>"
103
  ],
104
+ "eos_token": {
105
+ "content": "</s>",
106
+ "lstrip": false,
107
+ "normalized": false,
108
+ "rstrip": false,
109
+ "single_word": false
110
+ },
111
+ "pad_token": {
112
+ "content": "<pad>",
113
+ "lstrip": false,
114
+ "normalized": false,
115
+ "rstrip": false,
116
+ "single_word": false
117
+ },
118
+ "unk_token": {
119
+ "content": "<unk>",
120
+ "lstrip": false,
121
+ "normalized": false,
122
+ "rstrip": false,
123
+ "single_word": false
124
+ }
125
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json CHANGED
@@ -1,4 +1,5 @@
1
  {
 
2
  "added_tokens_decoder": {
3
  "0": {
4
  "content": "<pad>",
@@ -930,8 +931,10 @@
930
  "clean_up_tokenization_spaces": true,
931
  "eos_token": "</s>",
932
  "extra_ids": 100,
933
- "model_max_length": 1000000000000000019884624838656,
 
934
  "pad_token": "<pad>",
 
935
  "tokenizer_class": "T5Tokenizer",
936
  "unk_token": "<unk>"
937
  }
 
1
  {
2
+ "add_prefix_space": true,
3
  "added_tokens_decoder": {
4
  "0": {
5
  "content": "<pad>",
 
931
  "clean_up_tokenization_spaces": true,
932
  "eos_token": "</s>",
933
  "extra_ids": 100,
934
+ "legacy": true,
935
+ "model_max_length": 512,
936
  "pad_token": "<pad>",
937
+ "sp_model_kwargs": {},
938
  "tokenizer_class": "T5Tokenizer",
939
  "unk_token": "<unk>"
940
  }