Felix commited on
Commit
854ae2a
1 Parent(s): 0f955a0

Upload final model

Browse files
README.md ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: KBLab/bert-base-swedish-cased
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - ChangeIsKey/kubhist2
7
+ metrics:
8
+ - accuracy
9
+ model-index:
10
+ - name: test-mlm
11
+ results:
12
+ - task:
13
+ name: Masked Language Modeling
14
+ type: fill-mask
15
+ dataset:
16
+ name: ChangeIsKey/kubhist2
17
+ type: ChangeIsKey/kubhist2
18
+ config: all
19
+ split: train[:5%]
20
+ args: all
21
+ metrics:
22
+ - name: Accuracy
23
+ type: accuracy
24
+ value: 0.49963541184895327
25
+ ---
26
+
27
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
28
+ should probably proofread and complete it, then remove this comment. -->
29
+
30
+ # test-mlm
31
+
32
+ This model is a fine-tuned version of [KBLab/bert-base-swedish-cased](https://huggingface.co/KBLab/bert-base-swedish-cased) on the ChangeIsKey/kubhist2 dataset.
33
+ It achieves the following results on the evaluation set:
34
+ - Loss: 2.8883
35
+ - Accuracy: 0.4996
36
+
37
+ ## Model description
38
+
39
+ More information needed
40
+
41
+ ## Intended uses & limitations
42
+
43
+ More information needed
44
+
45
+ ## Training and evaluation data
46
+
47
+ More information needed
48
+
49
+ ## Training procedure
50
+
51
+ ### Training hyperparameters
52
+
53
+ The following hyperparameters were used during training:
54
+ - learning_rate: 5e-05
55
+ - train_batch_size: 128
56
+ - eval_batch_size: 32
57
+ - seed: 42
58
+ - gradient_accumulation_steps: 4
59
+ - total_train_batch_size: 512
60
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
61
+ - lr_scheduler_type: linear
62
+ - num_epochs: 3.0
63
+
64
+ ### Training results
65
+
66
+
67
+
68
+ ### Framework versions
69
+
70
+ - Transformers 4.33.1
71
+ - Pytorch 1.12.0
72
+ - Datasets 2.14.5
73
+ - Tokenizers 0.13.3
all_results.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.39,
3
+ "eval_accuracy": 0.49963541184895327,
4
+ "eval_loss": 2.888273239135742,
5
+ "eval_runtime": 4192.3617,
6
+ "eval_samples": 611273,
7
+ "eval_samples_per_second": 145.806,
8
+ "eval_steps_per_second": 4.557,
9
+ "perplexity": 17.962266268631573,
10
+ "train_loss": 0.0,
11
+ "train_runtime": 3.9321,
12
+ "train_samples": 9301750,
13
+ "train_samples_per_second": 7096699.346,
14
+ "train_steps_per_second": 13860.374
15
+ }
config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "KBLab/bert-base-swedish-cased",
3
+ "architectures": [
4
+ "BertForMaskedLM"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "gradient_checkpointing": false,
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 768,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3072,
14
+ "layer_norm_eps": 1e-12,
15
+ "max_position_embeddings": 512,
16
+ "model_type": "bert",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 12,
19
+ "output_past": true,
20
+ "pad_token_id": 0,
21
+ "position_embedding_type": "absolute",
22
+ "torch_dtype": "float32",
23
+ "transformers_version": "4.33.1",
24
+ "type_vocab_size": 2,
25
+ "use_cache": true,
26
+ "vocab_size": 50325
27
+ }
eval_results.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.39,
3
+ "eval_accuracy": 0.49963541184895327,
4
+ "eval_loss": 2.888273239135742,
5
+ "eval_runtime": 4192.3617,
6
+ "eval_samples": 611273,
7
+ "eval_samples_per_second": 145.806,
8
+ "eval_steps_per_second": 4.557,
9
+ "perplexity": 17.962266268631573
10
+ }
generation_config.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "pad_token_id": 0,
4
+ "transformers_version": "4.33.1"
5
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0fa76221690065c236b6addcd86e75e48390a744e41bea31c753213881fc07a1
3
+ size 499038381
special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "clean_up_tokenization_spaces": true,
3
+ "cls_token": "[CLS]",
4
+ "do_basic_tokenize": true,
5
+ "do_lower_case": false,
6
+ "mask_token": "[MASK]",
7
+ "model_max_length": 1000000000000000019884624838656,
8
+ "never_split": null,
9
+ "pad_token": "[PAD]",
10
+ "sep_token": "[SEP]",
11
+ "strip_accents": false,
12
+ "tokenize_chinese_chars": true,
13
+ "tokenizer_class": "BertTokenizer",
14
+ "unk_token": "[UNK]"
15
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.39,
3
+ "train_loss": 0.0,
4
+ "train_runtime": 3.9321,
5
+ "train_samples": 9301750,
6
+ "train_samples_per_second": 7096699.346,
7
+ "train_steps_per_second": 13860.374
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1072 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.3943855786431816,
5
+ "eval_steps": 500,
6
+ "global_step": 87000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 4.977065272235219e-05,
14
+ "loss": 3.2765,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.03,
19
+ "learning_rate": 4.954130544470437e-05,
20
+ "loss": 2.8884,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.04,
25
+ "learning_rate": 4.931195816705656e-05,
26
+ "loss": 2.7582,
27
+ "step": 1500
28
+ },
29
+ {
30
+ "epoch": 0.06,
31
+ "learning_rate": 4.908261088940875e-05,
32
+ "loss": 2.6689,
33
+ "step": 2000
34
+ },
35
+ {
36
+ "epoch": 0.07,
37
+ "learning_rate": 4.885326361176093e-05,
38
+ "loss": 2.6055,
39
+ "step": 2500
40
+ },
41
+ {
42
+ "epoch": 0.08,
43
+ "learning_rate": 4.862391633411312e-05,
44
+ "loss": 2.5549,
45
+ "step": 3000
46
+ },
47
+ {
48
+ "epoch": 0.1,
49
+ "learning_rate": 4.83945690564653e-05,
50
+ "loss": 2.5199,
51
+ "step": 3500
52
+ },
53
+ {
54
+ "epoch": 0.11,
55
+ "learning_rate": 4.816522177881749e-05,
56
+ "loss": 2.4834,
57
+ "step": 4000
58
+ },
59
+ {
60
+ "epoch": 0.12,
61
+ "learning_rate": 4.7935874501169674e-05,
62
+ "loss": 2.4538,
63
+ "step": 4500
64
+ },
65
+ {
66
+ "epoch": 0.14,
67
+ "learning_rate": 4.770652722352186e-05,
68
+ "loss": 2.4273,
69
+ "step": 5000
70
+ },
71
+ {
72
+ "epoch": 0.15,
73
+ "learning_rate": 4.7477179945874044e-05,
74
+ "loss": 2.4007,
75
+ "step": 5500
76
+ },
77
+ {
78
+ "epoch": 0.17,
79
+ "learning_rate": 4.7247832668226236e-05,
80
+ "loss": 2.3784,
81
+ "step": 6000
82
+ },
83
+ {
84
+ "epoch": 0.18,
85
+ "learning_rate": 4.7018485390578414e-05,
86
+ "loss": 2.3597,
87
+ "step": 6500
88
+ },
89
+ {
90
+ "epoch": 0.19,
91
+ "learning_rate": 4.67891381129306e-05,
92
+ "loss": 2.3433,
93
+ "step": 7000
94
+ },
95
+ {
96
+ "epoch": 0.21,
97
+ "learning_rate": 4.6559790835282785e-05,
98
+ "loss": 2.3291,
99
+ "step": 7500
100
+ },
101
+ {
102
+ "epoch": 0.22,
103
+ "learning_rate": 4.633044355763497e-05,
104
+ "loss": 2.3127,
105
+ "step": 8000
106
+ },
107
+ {
108
+ "epoch": 0.23,
109
+ "learning_rate": 4.610109627998716e-05,
110
+ "loss": 2.3013,
111
+ "step": 8500
112
+ },
113
+ {
114
+ "epoch": 0.25,
115
+ "learning_rate": 4.587174900233934e-05,
116
+ "loss": 2.2863,
117
+ "step": 9000
118
+ },
119
+ {
120
+ "epoch": 0.26,
121
+ "learning_rate": 4.564240172469153e-05,
122
+ "loss": 2.2772,
123
+ "step": 9500
124
+ },
125
+ {
126
+ "epoch": 0.28,
127
+ "learning_rate": 4.5413054447043716e-05,
128
+ "loss": 2.2645,
129
+ "step": 10000
130
+ },
131
+ {
132
+ "epoch": 0.29,
133
+ "learning_rate": 4.51837071693959e-05,
134
+ "loss": 2.2523,
135
+ "step": 10500
136
+ },
137
+ {
138
+ "epoch": 0.3,
139
+ "learning_rate": 4.4954359891748087e-05,
140
+ "loss": 2.2424,
141
+ "step": 11000
142
+ },
143
+ {
144
+ "epoch": 0.32,
145
+ "learning_rate": 4.472501261410027e-05,
146
+ "loss": 2.2344,
147
+ "step": 11500
148
+ },
149
+ {
150
+ "epoch": 0.33,
151
+ "learning_rate": 4.4495665336452457e-05,
152
+ "loss": 2.2204,
153
+ "step": 12000
154
+ },
155
+ {
156
+ "epoch": 0.34,
157
+ "learning_rate": 4.426631805880465e-05,
158
+ "loss": 2.2142,
159
+ "step": 12500
160
+ },
161
+ {
162
+ "epoch": 0.36,
163
+ "learning_rate": 4.4036970781156827e-05,
164
+ "loss": 2.2074,
165
+ "step": 13000
166
+ },
167
+ {
168
+ "epoch": 0.37,
169
+ "learning_rate": 4.380762350350902e-05,
170
+ "loss": 2.1972,
171
+ "step": 13500
172
+ },
173
+ {
174
+ "epoch": 0.39,
175
+ "learning_rate": 4.3578276225861203e-05,
176
+ "loss": 2.1857,
177
+ "step": 14000
178
+ },
179
+ {
180
+ "epoch": 0.4,
181
+ "learning_rate": 4.334892894821339e-05,
182
+ "loss": 2.1819,
183
+ "step": 14500
184
+ },
185
+ {
186
+ "epoch": 0.41,
187
+ "learning_rate": 4.3119581670565573e-05,
188
+ "loss": 2.1736,
189
+ "step": 15000
190
+ },
191
+ {
192
+ "epoch": 0.43,
193
+ "learning_rate": 4.289023439291776e-05,
194
+ "loss": 2.165,
195
+ "step": 15500
196
+ },
197
+ {
198
+ "epoch": 0.44,
199
+ "learning_rate": 4.2660887115269944e-05,
200
+ "loss": 2.1599,
201
+ "step": 16000
202
+ },
203
+ {
204
+ "epoch": 0.45,
205
+ "learning_rate": 4.243153983762213e-05,
206
+ "loss": 2.1566,
207
+ "step": 16500
208
+ },
209
+ {
210
+ "epoch": 0.47,
211
+ "learning_rate": 4.2202192559974314e-05,
212
+ "loss": 2.1496,
213
+ "step": 17000
214
+ },
215
+ {
216
+ "epoch": 0.48,
217
+ "learning_rate": 4.19728452823265e-05,
218
+ "loss": 2.1421,
219
+ "step": 17500
220
+ },
221
+ {
222
+ "epoch": 0.5,
223
+ "learning_rate": 4.174349800467869e-05,
224
+ "loss": 2.138,
225
+ "step": 18000
226
+ },
227
+ {
228
+ "epoch": 0.51,
229
+ "learning_rate": 4.151415072703087e-05,
230
+ "loss": 2.1319,
231
+ "step": 18500
232
+ },
233
+ {
234
+ "epoch": 0.52,
235
+ "learning_rate": 4.128480344938306e-05,
236
+ "loss": 2.1262,
237
+ "step": 19000
238
+ },
239
+ {
240
+ "epoch": 0.54,
241
+ "learning_rate": 4.1055456171735245e-05,
242
+ "loss": 2.1194,
243
+ "step": 19500
244
+ },
245
+ {
246
+ "epoch": 0.55,
247
+ "learning_rate": 4.082610889408743e-05,
248
+ "loss": 2.1162,
249
+ "step": 20000
250
+ },
251
+ {
252
+ "epoch": 0.56,
253
+ "learning_rate": 4.0596761616439616e-05,
254
+ "loss": 2.1084,
255
+ "step": 20500
256
+ },
257
+ {
258
+ "epoch": 0.58,
259
+ "learning_rate": 4.03674143387918e-05,
260
+ "loss": 2.1045,
261
+ "step": 21000
262
+ },
263
+ {
264
+ "epoch": 0.59,
265
+ "learning_rate": 4.0138067061143986e-05,
266
+ "loss": 2.0962,
267
+ "step": 21500
268
+ },
269
+ {
270
+ "epoch": 0.61,
271
+ "learning_rate": 3.990871978349617e-05,
272
+ "loss": 2.0967,
273
+ "step": 22000
274
+ },
275
+ {
276
+ "epoch": 0.62,
277
+ "learning_rate": 3.9679372505848356e-05,
278
+ "loss": 2.0875,
279
+ "step": 22500
280
+ },
281
+ {
282
+ "epoch": 0.63,
283
+ "learning_rate": 3.945002522820054e-05,
284
+ "loss": 2.0823,
285
+ "step": 23000
286
+ },
287
+ {
288
+ "epoch": 0.65,
289
+ "learning_rate": 3.922067795055273e-05,
290
+ "loss": 2.0831,
291
+ "step": 23500
292
+ },
293
+ {
294
+ "epoch": 0.66,
295
+ "learning_rate": 3.899133067290491e-05,
296
+ "loss": 2.0766,
297
+ "step": 24000
298
+ },
299
+ {
300
+ "epoch": 0.67,
301
+ "learning_rate": 3.87619833952571e-05,
302
+ "loss": 2.0725,
303
+ "step": 24500
304
+ },
305
+ {
306
+ "epoch": 0.69,
307
+ "learning_rate": 3.853263611760928e-05,
308
+ "loss": 2.0637,
309
+ "step": 25000
310
+ },
311
+ {
312
+ "epoch": 0.7,
313
+ "learning_rate": 3.830328883996147e-05,
314
+ "loss": 2.064,
315
+ "step": 25500
316
+ },
317
+ {
318
+ "epoch": 0.72,
319
+ "learning_rate": 3.807394156231366e-05,
320
+ "loss": 2.0582,
321
+ "step": 26000
322
+ },
323
+ {
324
+ "epoch": 0.73,
325
+ "learning_rate": 3.784459428466584e-05,
326
+ "loss": 2.0575,
327
+ "step": 26500
328
+ },
329
+ {
330
+ "epoch": 0.74,
331
+ "learning_rate": 3.761524700701803e-05,
332
+ "loss": 2.0531,
333
+ "step": 27000
334
+ },
335
+ {
336
+ "epoch": 0.76,
337
+ "learning_rate": 3.738589972937022e-05,
338
+ "loss": 2.0473,
339
+ "step": 27500
340
+ },
341
+ {
342
+ "epoch": 0.77,
343
+ "learning_rate": 3.71565524517224e-05,
344
+ "loss": 2.0474,
345
+ "step": 28000
346
+ },
347
+ {
348
+ "epoch": 0.78,
349
+ "learning_rate": 3.692720517407459e-05,
350
+ "loss": 2.0426,
351
+ "step": 28500
352
+ },
353
+ {
354
+ "epoch": 0.8,
355
+ "learning_rate": 3.6697857896426775e-05,
356
+ "loss": 2.0421,
357
+ "step": 29000
358
+ },
359
+ {
360
+ "epoch": 0.81,
361
+ "learning_rate": 3.646851061877895e-05,
362
+ "loss": 2.0366,
363
+ "step": 29500
364
+ },
365
+ {
366
+ "epoch": 0.83,
367
+ "learning_rate": 3.6239163341131145e-05,
368
+ "loss": 2.0312,
369
+ "step": 30000
370
+ },
371
+ {
372
+ "epoch": 0.84,
373
+ "learning_rate": 3.600981606348332e-05,
374
+ "loss": 2.0291,
375
+ "step": 30500
376
+ },
377
+ {
378
+ "epoch": 0.85,
379
+ "learning_rate": 3.5780468785835515e-05,
380
+ "loss": 2.0279,
381
+ "step": 31000
382
+ },
383
+ {
384
+ "epoch": 0.87,
385
+ "learning_rate": 3.55511215081877e-05,
386
+ "loss": 2.028,
387
+ "step": 31500
388
+ },
389
+ {
390
+ "epoch": 0.88,
391
+ "learning_rate": 3.5321774230539885e-05,
392
+ "loss": 2.023,
393
+ "step": 32000
394
+ },
395
+ {
396
+ "epoch": 0.89,
397
+ "learning_rate": 3.509242695289207e-05,
398
+ "loss": 2.0178,
399
+ "step": 32500
400
+ },
401
+ {
402
+ "epoch": 0.91,
403
+ "learning_rate": 3.486307967524426e-05,
404
+ "loss": 2.015,
405
+ "step": 33000
406
+ },
407
+ {
408
+ "epoch": 0.92,
409
+ "learning_rate": 3.463373239759644e-05,
410
+ "loss": 2.0145,
411
+ "step": 33500
412
+ },
413
+ {
414
+ "epoch": 0.94,
415
+ "learning_rate": 3.440438511994863e-05,
416
+ "loss": 2.0159,
417
+ "step": 34000
418
+ },
419
+ {
420
+ "epoch": 0.95,
421
+ "learning_rate": 3.417503784230081e-05,
422
+ "loss": 2.0089,
423
+ "step": 34500
424
+ },
425
+ {
426
+ "epoch": 0.96,
427
+ "learning_rate": 3.3945690564653e-05,
428
+ "loss": 2.004,
429
+ "step": 35000
430
+ },
431
+ {
432
+ "epoch": 0.98,
433
+ "learning_rate": 3.371634328700519e-05,
434
+ "loss": 2.0045,
435
+ "step": 35500
436
+ },
437
+ {
438
+ "epoch": 0.99,
439
+ "learning_rate": 3.348699600935737e-05,
440
+ "loss": 2.0026,
441
+ "step": 36000
442
+ },
443
+ {
444
+ "epoch": 1.0,
445
+ "learning_rate": 3.325764873170956e-05,
446
+ "loss": 1.9939,
447
+ "step": 36500
448
+ },
449
+ {
450
+ "epoch": 1.02,
451
+ "learning_rate": 3.302830145406174e-05,
452
+ "loss": 1.9928,
453
+ "step": 37000
454
+ },
455
+ {
456
+ "epoch": 1.03,
457
+ "learning_rate": 3.279895417641393e-05,
458
+ "loss": 1.9924,
459
+ "step": 37500
460
+ },
461
+ {
462
+ "epoch": 1.05,
463
+ "learning_rate": 3.256960689876611e-05,
464
+ "loss": 1.9886,
465
+ "step": 38000
466
+ },
467
+ {
468
+ "epoch": 1.06,
469
+ "learning_rate": 3.23402596211183e-05,
470
+ "loss": 1.9853,
471
+ "step": 38500
472
+ },
473
+ {
474
+ "epoch": 1.07,
475
+ "learning_rate": 3.211091234347048e-05,
476
+ "loss": 1.9814,
477
+ "step": 39000
478
+ },
479
+ {
480
+ "epoch": 1.09,
481
+ "learning_rate": 3.1881565065822674e-05,
482
+ "loss": 1.9826,
483
+ "step": 39500
484
+ },
485
+ {
486
+ "epoch": 1.1,
487
+ "learning_rate": 3.165221778817485e-05,
488
+ "loss": 1.98,
489
+ "step": 40000
490
+ },
491
+ {
492
+ "epoch": 1.11,
493
+ "learning_rate": 3.1422870510527044e-05,
494
+ "loss": 1.9785,
495
+ "step": 40500
496
+ },
497
+ {
498
+ "epoch": 1.13,
499
+ "learning_rate": 3.119352323287923e-05,
500
+ "loss": 1.9761,
501
+ "step": 41000
502
+ },
503
+ {
504
+ "epoch": 1.14,
505
+ "learning_rate": 3.0964175955231414e-05,
506
+ "loss": 1.9724,
507
+ "step": 41500
508
+ },
509
+ {
510
+ "epoch": 1.16,
511
+ "learning_rate": 3.07348286775836e-05,
512
+ "loss": 1.9746,
513
+ "step": 42000
514
+ },
515
+ {
516
+ "epoch": 1.17,
517
+ "learning_rate": 3.0505481399935787e-05,
518
+ "loss": 1.972,
519
+ "step": 42500
520
+ },
521
+ {
522
+ "epoch": 1.18,
523
+ "learning_rate": 3.027613412228797e-05,
524
+ "loss": 1.9707,
525
+ "step": 43000
526
+ },
527
+ {
528
+ "epoch": 1.2,
529
+ "learning_rate": 3.0046786844640157e-05,
530
+ "loss": 1.9685,
531
+ "step": 43500
532
+ },
533
+ {
534
+ "epoch": 1.21,
535
+ "learning_rate": 2.981743956699234e-05,
536
+ "loss": 1.9605,
537
+ "step": 44000
538
+ },
539
+ {
540
+ "epoch": 1.22,
541
+ "learning_rate": 2.9588092289344527e-05,
542
+ "loss": 1.9596,
543
+ "step": 44500
544
+ },
545
+ {
546
+ "epoch": 1.24,
547
+ "learning_rate": 2.9358745011696716e-05,
548
+ "loss": 1.9608,
549
+ "step": 45000
550
+ },
551
+ {
552
+ "epoch": 1.25,
553
+ "learning_rate": 2.9129397734048897e-05,
554
+ "loss": 1.9598,
555
+ "step": 45500
556
+ },
557
+ {
558
+ "epoch": 1.27,
559
+ "learning_rate": 2.8900050456401086e-05,
560
+ "loss": 1.9564,
561
+ "step": 46000
562
+ },
563
+ {
564
+ "epoch": 1.28,
565
+ "learning_rate": 2.867070317875327e-05,
566
+ "loss": 1.9557,
567
+ "step": 46500
568
+ },
569
+ {
570
+ "epoch": 1.29,
571
+ "learning_rate": 2.8441355901105456e-05,
572
+ "loss": 1.9526,
573
+ "step": 47000
574
+ },
575
+ {
576
+ "epoch": 1.31,
577
+ "learning_rate": 2.821200862345764e-05,
578
+ "loss": 1.9472,
579
+ "step": 47500
580
+ },
581
+ {
582
+ "epoch": 1.32,
583
+ "learning_rate": 2.7982661345809826e-05,
584
+ "loss": 1.9466,
585
+ "step": 48000
586
+ },
587
+ {
588
+ "epoch": 1.33,
589
+ "learning_rate": 2.775331406816201e-05,
590
+ "loss": 1.9479,
591
+ "step": 48500
592
+ },
593
+ {
594
+ "epoch": 1.35,
595
+ "learning_rate": 2.75239667905142e-05,
596
+ "loss": 1.9448,
597
+ "step": 49000
598
+ },
599
+ {
600
+ "epoch": 1.36,
601
+ "learning_rate": 2.729461951286638e-05,
602
+ "loss": 1.9451,
603
+ "step": 49500
604
+ },
605
+ {
606
+ "epoch": 1.38,
607
+ "learning_rate": 2.706527223521857e-05,
608
+ "loss": 1.9458,
609
+ "step": 50000
610
+ },
611
+ {
612
+ "epoch": 1.39,
613
+ "learning_rate": 2.6835924957570758e-05,
614
+ "loss": 1.9373,
615
+ "step": 50500
616
+ },
617
+ {
618
+ "epoch": 1.4,
619
+ "learning_rate": 2.660657767992294e-05,
620
+ "loss": 1.9364,
621
+ "step": 51000
622
+ },
623
+ {
624
+ "epoch": 1.42,
625
+ "learning_rate": 2.6377230402275128e-05,
626
+ "loss": 1.9345,
627
+ "step": 51500
628
+ },
629
+ {
630
+ "epoch": 1.43,
631
+ "learning_rate": 2.614788312462731e-05,
632
+ "loss": 1.9366,
633
+ "step": 52000
634
+ },
635
+ {
636
+ "epoch": 1.44,
637
+ "learning_rate": 2.5918535846979498e-05,
638
+ "loss": 1.9348,
639
+ "step": 52500
640
+ },
641
+ {
642
+ "epoch": 1.46,
643
+ "learning_rate": 2.5689188569331686e-05,
644
+ "loss": 1.9336,
645
+ "step": 53000
646
+ },
647
+ {
648
+ "epoch": 1.47,
649
+ "learning_rate": 2.5459841291683868e-05,
650
+ "loss": 1.932,
651
+ "step": 53500
652
+ },
653
+ {
654
+ "epoch": 1.49,
655
+ "learning_rate": 2.5230494014036056e-05,
656
+ "loss": 1.9333,
657
+ "step": 54000
658
+ },
659
+ {
660
+ "epoch": 1.5,
661
+ "learning_rate": 2.500114673638824e-05,
662
+ "loss": 1.9303,
663
+ "step": 54500
664
+ },
665
+ {
666
+ "epoch": 1.51,
667
+ "learning_rate": 2.4771799458740426e-05,
668
+ "loss": 1.9276,
669
+ "step": 55000
670
+ },
671
+ {
672
+ "epoch": 1.53,
673
+ "learning_rate": 2.454245218109261e-05,
674
+ "loss": 1.9268,
675
+ "step": 55500
676
+ },
677
+ {
678
+ "epoch": 1.54,
679
+ "learning_rate": 2.4313104903444796e-05,
680
+ "loss": 1.9244,
681
+ "step": 56000
682
+ },
683
+ {
684
+ "epoch": 1.55,
685
+ "learning_rate": 2.408375762579698e-05,
686
+ "loss": 1.9256,
687
+ "step": 56500
688
+ },
689
+ {
690
+ "epoch": 1.57,
691
+ "learning_rate": 2.385441034814917e-05,
692
+ "loss": 1.9191,
693
+ "step": 57000
694
+ },
695
+ {
696
+ "epoch": 1.58,
697
+ "learning_rate": 2.3625063070501355e-05,
698
+ "loss": 1.9186,
699
+ "step": 57500
700
+ },
701
+ {
702
+ "epoch": 1.6,
703
+ "learning_rate": 2.339571579285354e-05,
704
+ "loss": 1.9188,
705
+ "step": 58000
706
+ },
707
+ {
708
+ "epoch": 1.61,
709
+ "learning_rate": 2.3166368515205725e-05,
710
+ "loss": 1.9179,
711
+ "step": 58500
712
+ },
713
+ {
714
+ "epoch": 1.62,
715
+ "learning_rate": 2.2937021237557913e-05,
716
+ "loss": 1.9176,
717
+ "step": 59000
718
+ },
719
+ {
720
+ "epoch": 1.64,
721
+ "learning_rate": 2.27076739599101e-05,
722
+ "loss": 1.9154,
723
+ "step": 59500
724
+ },
725
+ {
726
+ "epoch": 1.65,
727
+ "learning_rate": 2.2478326682262283e-05,
728
+ "loss": 1.9119,
729
+ "step": 60000
730
+ },
731
+ {
732
+ "epoch": 1.67,
733
+ "learning_rate": 2.224897940461447e-05,
734
+ "loss": 1.9138,
735
+ "step": 60500
736
+ },
737
+ {
738
+ "epoch": 1.68,
739
+ "learning_rate": 2.2019632126966653e-05,
740
+ "loss": 1.9129,
741
+ "step": 61000
742
+ },
743
+ {
744
+ "epoch": 1.69,
745
+ "learning_rate": 2.179028484931884e-05,
746
+ "loss": 1.9114,
747
+ "step": 61500
748
+ },
749
+ {
750
+ "epoch": 1.71,
751
+ "learning_rate": 2.1560937571671023e-05,
752
+ "loss": 1.909,
753
+ "step": 62000
754
+ },
755
+ {
756
+ "epoch": 1.72,
757
+ "learning_rate": 2.133159029402321e-05,
758
+ "loss": 1.9054,
759
+ "step": 62500
760
+ },
761
+ {
762
+ "epoch": 1.73,
763
+ "learning_rate": 2.1102243016375397e-05,
764
+ "loss": 1.905,
765
+ "step": 63000
766
+ },
767
+ {
768
+ "epoch": 1.75,
769
+ "learning_rate": 2.0872895738727582e-05,
770
+ "loss": 1.9037,
771
+ "step": 63500
772
+ },
773
+ {
774
+ "epoch": 1.76,
775
+ "learning_rate": 2.0643548461079767e-05,
776
+ "loss": 1.9061,
777
+ "step": 64000
778
+ },
779
+ {
780
+ "epoch": 1.78,
781
+ "learning_rate": 2.0414201183431952e-05,
782
+ "loss": 1.902,
783
+ "step": 64500
784
+ },
785
+ {
786
+ "epoch": 1.79,
787
+ "learning_rate": 2.018485390578414e-05,
788
+ "loss": 1.8995,
789
+ "step": 65000
790
+ },
791
+ {
792
+ "epoch": 1.8,
793
+ "learning_rate": 1.9955506628136325e-05,
794
+ "loss": 1.8987,
795
+ "step": 65500
796
+ },
797
+ {
798
+ "epoch": 1.82,
799
+ "learning_rate": 1.972615935048851e-05,
800
+ "loss": 1.8988,
801
+ "step": 66000
802
+ },
803
+ {
804
+ "epoch": 1.83,
805
+ "learning_rate": 1.9496812072840695e-05,
806
+ "loss": 1.9013,
807
+ "step": 66500
808
+ },
809
+ {
810
+ "epoch": 1.84,
811
+ "learning_rate": 1.9267464795192884e-05,
812
+ "loss": 1.8992,
813
+ "step": 67000
814
+ },
815
+ {
816
+ "epoch": 1.86,
817
+ "learning_rate": 1.903811751754507e-05,
818
+ "loss": 1.8974,
819
+ "step": 67500
820
+ },
821
+ {
822
+ "epoch": 1.87,
823
+ "learning_rate": 1.8808770239897254e-05,
824
+ "loss": 1.8943,
825
+ "step": 68000
826
+ },
827
+ {
828
+ "epoch": 1.89,
829
+ "learning_rate": 1.857942296224944e-05,
830
+ "loss": 1.8954,
831
+ "step": 68500
832
+ },
833
+ {
834
+ "epoch": 1.9,
835
+ "learning_rate": 1.8350075684601624e-05,
836
+ "loss": 1.8942,
837
+ "step": 69000
838
+ },
839
+ {
840
+ "epoch": 1.91,
841
+ "learning_rate": 1.812072840695381e-05,
842
+ "loss": 1.8904,
843
+ "step": 69500
844
+ },
845
+ {
846
+ "epoch": 1.93,
847
+ "learning_rate": 1.7891381129305994e-05,
848
+ "loss": 1.8915,
849
+ "step": 70000
850
+ },
851
+ {
852
+ "epoch": 1.94,
853
+ "learning_rate": 1.7662033851658182e-05,
854
+ "loss": 1.8871,
855
+ "step": 70500
856
+ },
857
+ {
858
+ "epoch": 1.95,
859
+ "learning_rate": 1.7432686574010367e-05,
860
+ "loss": 1.8872,
861
+ "step": 71000
862
+ },
863
+ {
864
+ "epoch": 1.97,
865
+ "learning_rate": 1.7203339296362552e-05,
866
+ "loss": 1.8885,
867
+ "step": 71500
868
+ },
869
+ {
870
+ "epoch": 1.98,
871
+ "learning_rate": 1.6973992018714737e-05,
872
+ "loss": 1.8835,
873
+ "step": 72000
874
+ },
875
+ {
876
+ "epoch": 2.0,
877
+ "learning_rate": 1.6744644741066926e-05,
878
+ "loss": 1.8854,
879
+ "step": 72500
880
+ },
881
+ {
882
+ "epoch": 2.01,
883
+ "learning_rate": 1.651529746341911e-05,
884
+ "loss": 1.8856,
885
+ "step": 73000
886
+ },
887
+ {
888
+ "epoch": 2.02,
889
+ "learning_rate": 1.6285950185771296e-05,
890
+ "loss": 1.8836,
891
+ "step": 73500
892
+ },
893
+ {
894
+ "epoch": 2.04,
895
+ "learning_rate": 1.605660290812348e-05,
896
+ "loss": 1.8834,
897
+ "step": 74000
898
+ },
899
+ {
900
+ "epoch": 2.05,
901
+ "learning_rate": 1.582725563047567e-05,
902
+ "loss": 1.8814,
903
+ "step": 74500
904
+ },
905
+ {
906
+ "epoch": 2.06,
907
+ "learning_rate": 1.5597908352827854e-05,
908
+ "loss": 1.8795,
909
+ "step": 75000
910
+ },
911
+ {
912
+ "epoch": 2.08,
913
+ "learning_rate": 1.536856107518004e-05,
914
+ "loss": 1.8823,
915
+ "step": 75500
916
+ },
917
+ {
918
+ "epoch": 2.09,
919
+ "learning_rate": 1.5139213797532223e-05,
920
+ "loss": 1.8776,
921
+ "step": 76000
922
+ },
923
+ {
924
+ "epoch": 2.11,
925
+ "learning_rate": 1.4909866519884411e-05,
926
+ "loss": 1.8741,
927
+ "step": 76500
928
+ },
929
+ {
930
+ "epoch": 2.12,
931
+ "learning_rate": 1.4680519242236596e-05,
932
+ "loss": 1.8779,
933
+ "step": 77000
934
+ },
935
+ {
936
+ "epoch": 2.13,
937
+ "learning_rate": 1.4451171964588781e-05,
938
+ "loss": 1.8771,
939
+ "step": 77500
940
+ },
941
+ {
942
+ "epoch": 2.15,
943
+ "learning_rate": 1.4221824686940966e-05,
944
+ "loss": 1.8791,
945
+ "step": 78000
946
+ },
947
+ {
948
+ "epoch": 2.16,
949
+ "learning_rate": 1.3992477409293153e-05,
950
+ "loss": 1.873,
951
+ "step": 78500
952
+ },
953
+ {
954
+ "epoch": 2.17,
955
+ "learning_rate": 1.3763130131645338e-05,
956
+ "loss": 1.8713,
957
+ "step": 79000
958
+ },
959
+ {
960
+ "epoch": 2.19,
961
+ "learning_rate": 1.3533782853997523e-05,
962
+ "loss": 1.8717,
963
+ "step": 79500
964
+ },
965
+ {
966
+ "epoch": 2.2,
967
+ "learning_rate": 1.3304435576349708e-05,
968
+ "loss": 1.8699,
969
+ "step": 80000
970
+ },
971
+ {
972
+ "epoch": 2.22,
973
+ "learning_rate": 1.3075088298701896e-05,
974
+ "loss": 1.8729,
975
+ "step": 80500
976
+ },
977
+ {
978
+ "epoch": 2.23,
979
+ "learning_rate": 1.2845741021054082e-05,
980
+ "loss": 1.8708,
981
+ "step": 81000
982
+ },
983
+ {
984
+ "epoch": 2.24,
985
+ "learning_rate": 1.2616393743406265e-05,
986
+ "loss": 1.869,
987
+ "step": 81500
988
+ },
989
+ {
990
+ "epoch": 2.26,
991
+ "learning_rate": 1.2387046465758452e-05,
992
+ "loss": 1.8693,
993
+ "step": 82000
994
+ },
995
+ {
996
+ "epoch": 2.27,
997
+ "learning_rate": 1.2157699188110638e-05,
998
+ "loss": 1.8697,
999
+ "step": 82500
1000
+ },
1001
+ {
1002
+ "epoch": 2.28,
1003
+ "learning_rate": 1.1928351910462823e-05,
1004
+ "loss": 1.8694,
1005
+ "step": 83000
1006
+ },
1007
+ {
1008
+ "epoch": 2.3,
1009
+ "learning_rate": 1.169900463281501e-05,
1010
+ "loss": 1.8692,
1011
+ "step": 83500
1012
+ },
1013
+ {
1014
+ "epoch": 2.31,
1015
+ "learning_rate": 1.1469657355167195e-05,
1016
+ "loss": 1.871,
1017
+ "step": 84000
1018
+ },
1019
+ {
1020
+ "epoch": 2.33,
1021
+ "learning_rate": 1.1240310077519382e-05,
1022
+ "loss": 1.8667,
1023
+ "step": 84500
1024
+ },
1025
+ {
1026
+ "epoch": 2.34,
1027
+ "learning_rate": 1.1010962799871565e-05,
1028
+ "loss": 1.8603,
1029
+ "step": 85000
1030
+ },
1031
+ {
1032
+ "epoch": 2.35,
1033
+ "learning_rate": 1.0781615522223752e-05,
1034
+ "loss": 1.8638,
1035
+ "step": 85500
1036
+ },
1037
+ {
1038
+ "epoch": 2.37,
1039
+ "learning_rate": 1.0552268244575937e-05,
1040
+ "loss": 1.8632,
1041
+ "step": 86000
1042
+ },
1043
+ {
1044
+ "epoch": 2.38,
1045
+ "learning_rate": 1.0322920966928124e-05,
1046
+ "loss": 1.8652,
1047
+ "step": 86500
1048
+ },
1049
+ {
1050
+ "epoch": 2.39,
1051
+ "learning_rate": 1.0093573689280309e-05,
1052
+ "loss": 1.8638,
1053
+ "step": 87000
1054
+ },
1055
+ {
1056
+ "epoch": 2.39,
1057
+ "step": 87000,
1058
+ "total_flos": 5.8634472383906e+18,
1059
+ "train_loss": 0.0,
1060
+ "train_runtime": 3.9321,
1061
+ "train_samples_per_second": 7096699.346,
1062
+ "train_steps_per_second": 13860.374
1063
+ }
1064
+ ],
1065
+ "logging_steps": 500,
1066
+ "max_steps": 54501,
1067
+ "num_train_epochs": 3,
1068
+ "save_steps": 500,
1069
+ "total_flos": 5.8634472383906e+18,
1070
+ "trial_name": null,
1071
+ "trial_params": null
1072
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e7a1526f22ab763ee64a6a4bed2dacb2aa81fbe55d4d1816e6fa4ecb51db8799
3
+ size 3951