dehio commited on
Commit
254eb3c
1 Parent(s): 9923a71

Initial upload

Browse files
README.md ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - german_dpr
7
+ model-index:
8
+ - name: german-qg-t5-quad
9
+ results: []
10
+ ---
11
+
12
+ # t5-base-pretrained-qg
13
+
14
+ This model is fine-tuned in question generation in German. The expected answer must be highlighted with `<hl>` token.
15
+
16
+ ## Task example
17
+
18
+ #### Input
19
+ ```
20
+ generate question: Obwohl die Vereinigten Staaten wie auch viele Staaten des Commonwealth Erben des <hl>britischen Common Laws<hl> sind, setzt sich das amerikanische Recht bedeutend davon ab. Dies rührt größtenteils von dem langen Zeitraum her, [...]
21
+ ```
22
+ #### Expected output
23
+ ```
24
+ Von welchem Gesetzt stammt das Amerikanische ab?
25
+ ```
26
+
27
+ ## Model description
28
+
29
+ This model is a fine-tuned version of [valhalla/t5-base-qg-hl](https://huggingface.co/valhalla/t5-base-qg-hl) on the [GermanQUAD](https://www.deepset.ai/germanquad) dataset.
30
+
31
+ ## Training and evaluation data
32
+
33
+ The training script can be accessed [here](https://github.com/d-e-h-i-o/german-qg).
34
+
35
+ ### Evaluation
36
+
37
+ The model achieves a BLEU-4 score of **11.30** on the GermanQuAD test set (n=2204).
38
+
39
+ ### Training hyperparameters
40
+
41
+ The following hyperparameters were used during training:
42
+ - learning_rate: 0.0001
43
+ - train_batch_size: 2
44
+ - eval_batch_size: 2
45
+ - seed: 100
46
+ - gradient_accumulation_steps: 8
47
+ - total_train_batch_size: 16
48
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
+ - lr_scheduler_type: linear
50
+ - num_epochs: 10
51
+
52
+ ### Framework versions
53
+
54
+ - Transformers 4.13.0.dev0
55
+ - Pytorch 1.10.0+cu102
56
+ - Datasets 1.16.1
57
+ - Tokenizers 0.10.3
added_tokens.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"<hl>": 32100}
all_results.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "predict_bleu": 13.4716,
4
+ "predict_gen_len": 16.7945,
5
+ "predict_loss": 1.5842393636703491,
6
+ "predict_runtime": 384.6342,
7
+ "predict_samples": 2204,
8
+ "predict_samples_per_second": 5.73,
9
+ "predict_steps_per_second": 2.865,
10
+ "train_loss": 0.9689414988638463,
11
+ "train_runtime": 6282.5026,
12
+ "train_samples": 11518,
13
+ "train_samples_per_second": 18.333,
14
+ "train_steps_per_second": 1.144
15
+ }
config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "dehio/german-qg-t5-quad",
3
+ "architectures": [
4
+ "T5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 3072,
7
+ "d_kv": 64,
8
+ "d_model": 768,
9
+ "decoder_start_token_id": 0,
10
+ "dropout_rate": 0.1,
11
+ "eos_token_id": 1,
12
+ "feed_forward_proj": "relu",
13
+ "initializer_factor": 1.0,
14
+ "is_encoder_decoder": true,
15
+ "layer_norm_epsilon": 1e-06,
16
+ "model_type": "t5",
17
+ "n_positions": 512,
18
+ "num_decoder_layers": 12,
19
+ "num_heads": 12,
20
+ "num_layers": 12,
21
+ "output_past": true,
22
+ "pad_token_id": 0,
23
+ "relative_attention_num_buckets": 32,
24
+ "task_specific_params": {
25
+ "summarization": {
26
+ "early_stopping": true,
27
+ "max_length": 32,
28
+ "num_beams": 4,
29
+ "prefix": ""
30
+ },
31
+ "translation_en_to_de": {
32
+ "early_stopping": true,
33
+ "max_length": 32,
34
+ "num_beams": 4,
35
+ "prefix": ""
36
+ },
37
+ "translation_en_to_fr": {
38
+ "early_stopping": true,
39
+ "max_length": 32,
40
+ "num_beams": 4,
41
+ "prefix": ""
42
+ },
43
+ "translation_en_to_ro": {
44
+ "early_stopping": true,
45
+ "max_length": 32,
46
+ "num_beams": 4,
47
+ "prefix": ""
48
+ }
49
+ },
50
+ "torch_dtype": "float32",
51
+ "transformers_version": "4.13.0.dev0",
52
+ "use_cache": true,
53
+ "vocab_size": 32101
54
+ }
generated_predictions.txt ADDED
The diff for this file is too large to render. See raw diff
 
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:79b661c5955b1c3ecaada3bb43f35502b8107bf7313e97a177006f6244d14fc6
3
+ size 891647935
sources.txt ADDED
The diff for this file is too large to render. See raw diff
 
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"]}
spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
3
+ size 791656
targets.txt ADDED
The diff for this file is too large to render. See raw diff
 
test_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "predict_bleu": 13.4716,
3
+ "predict_gen_len": 16.7945,
4
+ "predict_loss": 1.5842393636703491,
5
+ "predict_runtime": 384.6342,
6
+ "predict_samples": 2204,
7
+ "predict_samples_per_second": 5.73,
8
+ "predict_steps_per_second": 2.865
9
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 100, "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"], "model_max_length": 512, "special_tokens_map_file": "models/german-qg-t5/special_tokens_map.json", "full_tokenizer_file": null, "name_or_path": "dehio/german-qg-t5-quad", "sp_model_kwargs": {}, "tokenizer_class": "T5Tokenizer"}
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "train_loss": 0.9689414988638463,
4
+ "train_runtime": 6282.5026,
5
+ "train_samples": 11518,
6
+ "train_samples_per_second": 18.333,
7
+ "train_steps_per_second": 1.144
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,451 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 9.998784511199862,
5
+ "global_step": 7190,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.14,
12
+ "learning_rate": 9.860917941585536e-05,
13
+ "loss": 1.7018,
14
+ "step": 100
15
+ },
16
+ {
17
+ "epoch": 0.28,
18
+ "learning_rate": 9.721835883171071e-05,
19
+ "loss": 1.5678,
20
+ "step": 200
21
+ },
22
+ {
23
+ "epoch": 0.42,
24
+ "learning_rate": 9.582753824756607e-05,
25
+ "loss": 1.5333,
26
+ "step": 300
27
+ },
28
+ {
29
+ "epoch": 0.56,
30
+ "learning_rate": 9.443671766342142e-05,
31
+ "loss": 1.5235,
32
+ "step": 400
33
+ },
34
+ {
35
+ "epoch": 0.69,
36
+ "learning_rate": 9.304589707927678e-05,
37
+ "loss": 1.4892,
38
+ "step": 500
39
+ },
40
+ {
41
+ "epoch": 0.83,
42
+ "learning_rate": 9.165507649513213e-05,
43
+ "loss": 1.4781,
44
+ "step": 600
45
+ },
46
+ {
47
+ "epoch": 0.97,
48
+ "learning_rate": 9.026425591098748e-05,
49
+ "loss": 1.4319,
50
+ "step": 700
51
+ },
52
+ {
53
+ "epoch": 1.11,
54
+ "learning_rate": 8.887343532684285e-05,
55
+ "loss": 1.3011,
56
+ "step": 800
57
+ },
58
+ {
59
+ "epoch": 1.25,
60
+ "learning_rate": 8.74826147426982e-05,
61
+ "loss": 1.2885,
62
+ "step": 900
63
+ },
64
+ {
65
+ "epoch": 1.39,
66
+ "learning_rate": 8.609179415855355e-05,
67
+ "loss": 1.2524,
68
+ "step": 1000
69
+ },
70
+ {
71
+ "epoch": 1.53,
72
+ "learning_rate": 8.47009735744089e-05,
73
+ "loss": 1.2674,
74
+ "step": 1100
75
+ },
76
+ {
77
+ "epoch": 1.67,
78
+ "learning_rate": 8.331015299026426e-05,
79
+ "loss": 1.2533,
80
+ "step": 1200
81
+ },
82
+ {
83
+ "epoch": 1.81,
84
+ "learning_rate": 8.191933240611962e-05,
85
+ "loss": 1.268,
86
+ "step": 1300
87
+ },
88
+ {
89
+ "epoch": 1.95,
90
+ "learning_rate": 8.052851182197498e-05,
91
+ "loss": 1.2499,
92
+ "step": 1400
93
+ },
94
+ {
95
+ "epoch": 2.09,
96
+ "learning_rate": 7.913769123783032e-05,
97
+ "loss": 1.1782,
98
+ "step": 1500
99
+ },
100
+ {
101
+ "epoch": 2.23,
102
+ "learning_rate": 7.774687065368567e-05,
103
+ "loss": 1.1113,
104
+ "step": 1600
105
+ },
106
+ {
107
+ "epoch": 2.36,
108
+ "learning_rate": 7.635605006954103e-05,
109
+ "loss": 1.1167,
110
+ "step": 1700
111
+ },
112
+ {
113
+ "epoch": 2.5,
114
+ "learning_rate": 7.496522948539638e-05,
115
+ "loss": 1.1422,
116
+ "step": 1800
117
+ },
118
+ {
119
+ "epoch": 2.64,
120
+ "learning_rate": 7.357440890125175e-05,
121
+ "loss": 1.118,
122
+ "step": 1900
123
+ },
124
+ {
125
+ "epoch": 2.78,
126
+ "learning_rate": 7.21835883171071e-05,
127
+ "loss": 1.1133,
128
+ "step": 2000
129
+ },
130
+ {
131
+ "epoch": 2.92,
132
+ "learning_rate": 7.079276773296244e-05,
133
+ "loss": 1.1058,
134
+ "step": 2100
135
+ },
136
+ {
137
+ "epoch": 3.06,
138
+ "learning_rate": 6.94019471488178e-05,
139
+ "loss": 1.0747,
140
+ "step": 2200
141
+ },
142
+ {
143
+ "epoch": 3.2,
144
+ "learning_rate": 6.801112656467315e-05,
145
+ "loss": 1.0059,
146
+ "step": 2300
147
+ },
148
+ {
149
+ "epoch": 3.34,
150
+ "learning_rate": 6.662030598052852e-05,
151
+ "loss": 0.9988,
152
+ "step": 2400
153
+ },
154
+ {
155
+ "epoch": 3.48,
156
+ "learning_rate": 6.522948539638388e-05,
157
+ "loss": 1.0111,
158
+ "step": 2500
159
+ },
160
+ {
161
+ "epoch": 3.62,
162
+ "learning_rate": 6.383866481223923e-05,
163
+ "loss": 1.0237,
164
+ "step": 2600
165
+ },
166
+ {
167
+ "epoch": 3.75,
168
+ "learning_rate": 6.244784422809457e-05,
169
+ "loss": 1.0011,
170
+ "step": 2700
171
+ },
172
+ {
173
+ "epoch": 3.89,
174
+ "learning_rate": 6.105702364394992e-05,
175
+ "loss": 1.0133,
176
+ "step": 2800
177
+ },
178
+ {
179
+ "epoch": 4.03,
180
+ "learning_rate": 5.966620305980529e-05,
181
+ "loss": 0.9724,
182
+ "step": 2900
183
+ },
184
+ {
185
+ "epoch": 4.17,
186
+ "learning_rate": 5.827538247566065e-05,
187
+ "loss": 0.9005,
188
+ "step": 3000
189
+ },
190
+ {
191
+ "epoch": 4.31,
192
+ "learning_rate": 5.6884561891515995e-05,
193
+ "loss": 0.9264,
194
+ "step": 3100
195
+ },
196
+ {
197
+ "epoch": 4.45,
198
+ "learning_rate": 5.549374130737135e-05,
199
+ "loss": 0.933,
200
+ "step": 3200
201
+ },
202
+ {
203
+ "epoch": 4.59,
204
+ "learning_rate": 5.41029207232267e-05,
205
+ "loss": 0.9283,
206
+ "step": 3300
207
+ },
208
+ {
209
+ "epoch": 4.73,
210
+ "learning_rate": 5.2712100139082064e-05,
211
+ "loss": 0.9033,
212
+ "step": 3400
213
+ },
214
+ {
215
+ "epoch": 4.87,
216
+ "learning_rate": 5.132127955493742e-05,
217
+ "loss": 0.9139,
218
+ "step": 3500
219
+ },
220
+ {
221
+ "epoch": 5.01,
222
+ "learning_rate": 4.993045897079277e-05,
223
+ "loss": 0.9383,
224
+ "step": 3600
225
+ },
226
+ {
227
+ "epoch": 5.15,
228
+ "learning_rate": 4.853963838664812e-05,
229
+ "loss": 0.8268,
230
+ "step": 3700
231
+ },
232
+ {
233
+ "epoch": 5.28,
234
+ "learning_rate": 4.714881780250348e-05,
235
+ "loss": 0.841,
236
+ "step": 3800
237
+ },
238
+ {
239
+ "epoch": 5.42,
240
+ "learning_rate": 4.5757997218358836e-05,
241
+ "loss": 0.8594,
242
+ "step": 3900
243
+ },
244
+ {
245
+ "epoch": 5.56,
246
+ "learning_rate": 4.436717663421418e-05,
247
+ "loss": 0.8346,
248
+ "step": 4000
249
+ },
250
+ {
251
+ "epoch": 5.7,
252
+ "learning_rate": 4.2976356050069544e-05,
253
+ "loss": 0.8626,
254
+ "step": 4100
255
+ },
256
+ {
257
+ "epoch": 5.84,
258
+ "learning_rate": 4.15855354659249e-05,
259
+ "loss": 0.8691,
260
+ "step": 4200
261
+ },
262
+ {
263
+ "epoch": 5.98,
264
+ "learning_rate": 4.019471488178025e-05,
265
+ "loss": 0.8424,
266
+ "step": 4300
267
+ },
268
+ {
269
+ "epoch": 6.12,
270
+ "learning_rate": 3.880389429763561e-05,
271
+ "loss": 0.793,
272
+ "step": 4400
273
+ },
274
+ {
275
+ "epoch": 6.26,
276
+ "learning_rate": 3.741307371349096e-05,
277
+ "loss": 0.8025,
278
+ "step": 4500
279
+ },
280
+ {
281
+ "epoch": 6.4,
282
+ "learning_rate": 3.6022253129346316e-05,
283
+ "loss": 0.8245,
284
+ "step": 4600
285
+ },
286
+ {
287
+ "epoch": 6.54,
288
+ "learning_rate": 3.463143254520167e-05,
289
+ "loss": 0.797,
290
+ "step": 4700
291
+ },
292
+ {
293
+ "epoch": 6.68,
294
+ "learning_rate": 3.3240611961057024e-05,
295
+ "loss": 0.8063,
296
+ "step": 4800
297
+ },
298
+ {
299
+ "epoch": 6.81,
300
+ "learning_rate": 3.184979137691238e-05,
301
+ "loss": 0.8042,
302
+ "step": 4900
303
+ },
304
+ {
305
+ "epoch": 6.95,
306
+ "learning_rate": 3.0458970792767733e-05,
307
+ "loss": 0.8068,
308
+ "step": 5000
309
+ },
310
+ {
311
+ "epoch": 7.09,
312
+ "learning_rate": 2.906815020862309e-05,
313
+ "loss": 0.7873,
314
+ "step": 5100
315
+ },
316
+ {
317
+ "epoch": 7.23,
318
+ "learning_rate": 2.767732962447844e-05,
319
+ "loss": 0.7495,
320
+ "step": 5200
321
+ },
322
+ {
323
+ "epoch": 7.37,
324
+ "learning_rate": 2.6286509040333796e-05,
325
+ "loss": 0.7538,
326
+ "step": 5300
327
+ },
328
+ {
329
+ "epoch": 7.51,
330
+ "learning_rate": 2.4895688456189153e-05,
331
+ "loss": 0.7631,
332
+ "step": 5400
333
+ },
334
+ {
335
+ "epoch": 7.65,
336
+ "learning_rate": 2.3504867872044508e-05,
337
+ "loss": 0.7532,
338
+ "step": 5500
339
+ },
340
+ {
341
+ "epoch": 7.79,
342
+ "learning_rate": 2.2114047287899862e-05,
343
+ "loss": 0.7772,
344
+ "step": 5600
345
+ },
346
+ {
347
+ "epoch": 7.93,
348
+ "learning_rate": 2.0723226703755216e-05,
349
+ "loss": 0.7691,
350
+ "step": 5700
351
+ },
352
+ {
353
+ "epoch": 8.07,
354
+ "learning_rate": 1.933240611961057e-05,
355
+ "loss": 0.7544,
356
+ "step": 5800
357
+ },
358
+ {
359
+ "epoch": 8.21,
360
+ "learning_rate": 1.7941585535465928e-05,
361
+ "loss": 0.7243,
362
+ "step": 5900
363
+ },
364
+ {
365
+ "epoch": 8.34,
366
+ "learning_rate": 1.655076495132128e-05,
367
+ "loss": 0.7308,
368
+ "step": 6000
369
+ },
370
+ {
371
+ "epoch": 8.48,
372
+ "learning_rate": 1.5159944367176635e-05,
373
+ "loss": 0.7115,
374
+ "step": 6100
375
+ },
376
+ {
377
+ "epoch": 8.62,
378
+ "learning_rate": 1.376912378303199e-05,
379
+ "loss": 0.7275,
380
+ "step": 6200
381
+ },
382
+ {
383
+ "epoch": 8.76,
384
+ "learning_rate": 1.2378303198887344e-05,
385
+ "loss": 0.7279,
386
+ "step": 6300
387
+ },
388
+ {
389
+ "epoch": 8.9,
390
+ "learning_rate": 1.0987482614742698e-05,
391
+ "loss": 0.7497,
392
+ "step": 6400
393
+ },
394
+ {
395
+ "epoch": 9.04,
396
+ "learning_rate": 9.596662030598054e-06,
397
+ "loss": 0.7163,
398
+ "step": 6500
399
+ },
400
+ {
401
+ "epoch": 9.18,
402
+ "learning_rate": 8.205841446453408e-06,
403
+ "loss": 0.7047,
404
+ "step": 6600
405
+ },
406
+ {
407
+ "epoch": 9.32,
408
+ "learning_rate": 6.815020862308763e-06,
409
+ "loss": 0.7089,
410
+ "step": 6700
411
+ },
412
+ {
413
+ "epoch": 9.46,
414
+ "learning_rate": 5.424200278164117e-06,
415
+ "loss": 0.7078,
416
+ "step": 6800
417
+ },
418
+ {
419
+ "epoch": 9.6,
420
+ "learning_rate": 4.033379694019471e-06,
421
+ "loss": 0.7065,
422
+ "step": 6900
423
+ },
424
+ {
425
+ "epoch": 9.73,
426
+ "learning_rate": 2.6425591098748263e-06,
427
+ "loss": 0.7017,
428
+ "step": 7000
429
+ },
430
+ {
431
+ "epoch": 9.87,
432
+ "learning_rate": 1.2517385257301808e-06,
433
+ "loss": 0.6855,
434
+ "step": 7100
435
+ },
436
+ {
437
+ "epoch": 10.0,
438
+ "step": 7190,
439
+ "total_flos": 6.20936877252096e+16,
440
+ "train_loss": 0.9689414988638463,
441
+ "train_runtime": 6282.5026,
442
+ "train_samples_per_second": 18.333,
443
+ "train_steps_per_second": 1.144
444
+ }
445
+ ],
446
+ "max_steps": 7190,
447
+ "num_train_epochs": 10,
448
+ "total_flos": 6.20936877252096e+16,
449
+ "trial_name": null,
450
+ "trial_params": null
451
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc3e1f465cde2661ed9562251787718f1ce3e2112920cf5a88b7c91a5d00224e
3
+ size 3055