joelniklaus commited on
Commit
24eb1b5
1 Parent(s): db1f65d

Training in progress, step 50000

Browse files
.gitignore ADDED
@@ -0,0 +1 @@
 
 
1
+ checkpoint-*/
config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "data/plms/legal-swiss-roberta-large",
3
+ "architectures": [
4
+ "RobertaForMaskedLM"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 1,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 1024,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 4096,
15
+ "layer_norm_eps": 1e-05,
16
+ "max_position_embeddings": 514,
17
+ "model_type": "roberta",
18
+ "num_attention_heads": 16,
19
+ "num_hidden_layers": 24,
20
+ "pad_token_id": 0,
21
+ "position_embedding_type": "absolute",
22
+ "torch_dtype": "float32",
23
+ "transformers_version": "4.20.1",
24
+ "type_vocab_size": 1,
25
+ "use_cache": true,
26
+ "vocab_size": 128000
27
+ }
last-checkpoint/config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "data/plms/legal-swiss-roberta-large",
3
+ "architectures": [
4
+ "RobertaForMaskedLM"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 1,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 1024,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 4096,
15
+ "layer_norm_eps": 1e-05,
16
+ "max_position_embeddings": 514,
17
+ "model_type": "roberta",
18
+ "num_attention_heads": 16,
19
+ "num_hidden_layers": 24,
20
+ "pad_token_id": 0,
21
+ "position_embedding_type": "absolute",
22
+ "torch_dtype": "float32",
23
+ "transformers_version": "4.20.1",
24
+ "type_vocab_size": 1,
25
+ "use_cache": true,
26
+ "vocab_size": 128000
27
+ }
last-checkpoint/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a982e7b8c52e58016b548d0913c7a873f1fc173c4c773f447b8e02f753777220
3
+ size 3480941785
last-checkpoint/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1cb884b928f113ebd9559b6a1206824e01234abee82c00f58f1d575aa4ab30ab
3
+ size 1740493675
last-checkpoint/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0cba4c4280fe2708987ba830760c10347d94e259759f59f6c8b17c2643d54d1d
3
+ size 13611
last-checkpoint/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0cba4c4280fe2708987ba830760c10347d94e259759f59f6c8b17c2643d54d1d
3
+ size 13611
last-checkpoint/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0cba4c4280fe2708987ba830760c10347d94e259759f59f6c8b17c2643d54d1d
3
+ size 13611
last-checkpoint/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0cba4c4280fe2708987ba830760c10347d94e259759f59f6c8b17c2643d54d1d
3
+ size 13611
last-checkpoint/rng_state_4.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0cba4c4280fe2708987ba830760c10347d94e259759f59f6c8b17c2643d54d1d
3
+ size 13611
last-checkpoint/rng_state_5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0cba4c4280fe2708987ba830760c10347d94e259759f59f6c8b17c2643d54d1d
3
+ size 13611
last-checkpoint/rng_state_6.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0cba4c4280fe2708987ba830760c10347d94e259759f59f6c8b17c2643d54d1d
3
+ size 13611
last-checkpoint/rng_state_7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0cba4c4280fe2708987ba830760c10347d94e259759f59f6c8b17c2643d54d1d
3
+ size 13611
last-checkpoint/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4d3c7b27bde8b1f5a5595802aeb17178a4cf1850adf7366fbf6ace3edc933ed8
3
+ size 623
last-checkpoint/special_tokens_map.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": "<mask>",
6
+ "pad_token": "<pad>",
7
+ "sep_token": "</s>",
8
+ "unk_token": "<unk>"
9
+ }
last-checkpoint/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
last-checkpoint/tokenizer_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": "<mask>",
6
+ "model_max_length": 512,
7
+ "name_or_path": "data/plms/legal-swiss-roberta-large",
8
+ "pad_token": "<pad>",
9
+ "sep_token": "</s>",
10
+ "special_tokens_map_file": "data/plms/legal-swiss-roberta-large_128k/special_tokens_map.json",
11
+ "tokenizer_class": "PreTrainedTokenizerFast",
12
+ "unk_token": "<unk>"
13
+ }
last-checkpoint/trainer_state.json ADDED
@@ -0,0 +1,324 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.05,
5
+ "global_step": 50000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.0,
12
+ "learning_rate": 2.0000000000000003e-06,
13
+ "loss": 6.2493,
14
+ "step": 1000
15
+ },
16
+ {
17
+ "epoch": 0.0,
18
+ "learning_rate": 4.000000000000001e-06,
19
+ "loss": 4.6137,
20
+ "step": 2000
21
+ },
22
+ {
23
+ "epoch": 0.0,
24
+ "learning_rate": 6e-06,
25
+ "loss": 3.7251,
26
+ "step": 3000
27
+ },
28
+ {
29
+ "epoch": 0.0,
30
+ "learning_rate": 8.000000000000001e-06,
31
+ "loss": 3.5112,
32
+ "step": 4000
33
+ },
34
+ {
35
+ "epoch": 0.01,
36
+ "learning_rate": 1e-05,
37
+ "loss": 3.3686,
38
+ "step": 5000
39
+ },
40
+ {
41
+ "epoch": 0.01,
42
+ "learning_rate": 1.2e-05,
43
+ "loss": 3.3207,
44
+ "step": 6000
45
+ },
46
+ {
47
+ "epoch": 0.01,
48
+ "learning_rate": 1.4000000000000001e-05,
49
+ "loss": 3.0354,
50
+ "step": 7000
51
+ },
52
+ {
53
+ "epoch": 0.01,
54
+ "learning_rate": 1.6000000000000003e-05,
55
+ "loss": 2.9748,
56
+ "step": 8000
57
+ },
58
+ {
59
+ "epoch": 0.01,
60
+ "learning_rate": 1.8e-05,
61
+ "loss": 3.0796,
62
+ "step": 9000
63
+ },
64
+ {
65
+ "epoch": 0.01,
66
+ "learning_rate": 2e-05,
67
+ "loss": 2.8979,
68
+ "step": 10000
69
+ },
70
+ {
71
+ "epoch": 0.01,
72
+ "learning_rate": 2.2000000000000003e-05,
73
+ "loss": 2.7057,
74
+ "step": 11000
75
+ },
76
+ {
77
+ "epoch": 0.01,
78
+ "learning_rate": 2.4e-05,
79
+ "loss": 2.5434,
80
+ "step": 12000
81
+ },
82
+ {
83
+ "epoch": 0.01,
84
+ "learning_rate": 2.6000000000000002e-05,
85
+ "loss": 2.4562,
86
+ "step": 13000
87
+ },
88
+ {
89
+ "epoch": 0.01,
90
+ "learning_rate": 2.8000000000000003e-05,
91
+ "loss": 2.4019,
92
+ "step": 14000
93
+ },
94
+ {
95
+ "epoch": 0.01,
96
+ "learning_rate": 3e-05,
97
+ "loss": 2.3888,
98
+ "step": 15000
99
+ },
100
+ {
101
+ "epoch": 0.02,
102
+ "learning_rate": 3.2000000000000005e-05,
103
+ "loss": 2.2318,
104
+ "step": 16000
105
+ },
106
+ {
107
+ "epoch": 0.02,
108
+ "learning_rate": 3.4000000000000007e-05,
109
+ "loss": 2.2318,
110
+ "step": 17000
111
+ },
112
+ {
113
+ "epoch": 0.02,
114
+ "learning_rate": 3.6e-05,
115
+ "loss": 2.2482,
116
+ "step": 18000
117
+ },
118
+ {
119
+ "epoch": 0.02,
120
+ "learning_rate": 3.8e-05,
121
+ "loss": 2.24,
122
+ "step": 19000
123
+ },
124
+ {
125
+ "epoch": 0.02,
126
+ "learning_rate": 4e-05,
127
+ "loss": 2.2051,
128
+ "step": 20000
129
+ },
130
+ {
131
+ "epoch": 0.02,
132
+ "learning_rate": 4.2e-05,
133
+ "loss": 2.1275,
134
+ "step": 21000
135
+ },
136
+ {
137
+ "epoch": 0.02,
138
+ "learning_rate": 4.4000000000000006e-05,
139
+ "loss": 2.0528,
140
+ "step": 22000
141
+ },
142
+ {
143
+ "epoch": 0.02,
144
+ "learning_rate": 4.600000000000001e-05,
145
+ "loss": 2.0807,
146
+ "step": 23000
147
+ },
148
+ {
149
+ "epoch": 0.02,
150
+ "learning_rate": 4.8e-05,
151
+ "loss": 1.9779,
152
+ "step": 24000
153
+ },
154
+ {
155
+ "epoch": 0.03,
156
+ "learning_rate": 5e-05,
157
+ "loss": 1.8728,
158
+ "step": 25000
159
+ },
160
+ {
161
+ "epoch": 0.03,
162
+ "learning_rate": 5.2000000000000004e-05,
163
+ "loss": 1.6435,
164
+ "step": 26000
165
+ },
166
+ {
167
+ "epoch": 0.03,
168
+ "learning_rate": 5.4000000000000005e-05,
169
+ "loss": 1.4678,
170
+ "step": 27000
171
+ },
172
+ {
173
+ "epoch": 0.03,
174
+ "learning_rate": 5.6000000000000006e-05,
175
+ "loss": 1.4546,
176
+ "step": 28000
177
+ },
178
+ {
179
+ "epoch": 0.03,
180
+ "learning_rate": 5.8e-05,
181
+ "loss": 1.4257,
182
+ "step": 29000
183
+ },
184
+ {
185
+ "epoch": 0.03,
186
+ "learning_rate": 6e-05,
187
+ "loss": 1.3665,
188
+ "step": 30000
189
+ },
190
+ {
191
+ "epoch": 0.03,
192
+ "learning_rate": 6.2e-05,
193
+ "loss": 1.2821,
194
+ "step": 31000
195
+ },
196
+ {
197
+ "epoch": 0.03,
198
+ "learning_rate": 6.400000000000001e-05,
199
+ "loss": 1.3203,
200
+ "step": 32000
201
+ },
202
+ {
203
+ "epoch": 0.03,
204
+ "learning_rate": 6.6e-05,
205
+ "loss": 1.3713,
206
+ "step": 33000
207
+ },
208
+ {
209
+ "epoch": 0.03,
210
+ "learning_rate": 6.800000000000001e-05,
211
+ "loss": 1.338,
212
+ "step": 34000
213
+ },
214
+ {
215
+ "epoch": 0.04,
216
+ "learning_rate": 7e-05,
217
+ "loss": 1.2829,
218
+ "step": 35000
219
+ },
220
+ {
221
+ "epoch": 0.04,
222
+ "learning_rate": 7.2e-05,
223
+ "loss": 1.1656,
224
+ "step": 36000
225
+ },
226
+ {
227
+ "epoch": 0.04,
228
+ "learning_rate": 7.4e-05,
229
+ "loss": 1.2479,
230
+ "step": 37000
231
+ },
232
+ {
233
+ "epoch": 0.04,
234
+ "learning_rate": 7.6e-05,
235
+ "loss": 1.2553,
236
+ "step": 38000
237
+ },
238
+ {
239
+ "epoch": 0.04,
240
+ "learning_rate": 7.800000000000001e-05,
241
+ "loss": 1.2514,
242
+ "step": 39000
243
+ },
244
+ {
245
+ "epoch": 0.04,
246
+ "learning_rate": 8e-05,
247
+ "loss": 1.2509,
248
+ "step": 40000
249
+ },
250
+ {
251
+ "epoch": 0.04,
252
+ "learning_rate": 8.2e-05,
253
+ "loss": 1.2178,
254
+ "step": 41000
255
+ },
256
+ {
257
+ "epoch": 0.04,
258
+ "learning_rate": 8.4e-05,
259
+ "loss": 1.2506,
260
+ "step": 42000
261
+ },
262
+ {
263
+ "epoch": 0.04,
264
+ "learning_rate": 8.6e-05,
265
+ "loss": 1.2735,
266
+ "step": 43000
267
+ },
268
+ {
269
+ "epoch": 0.04,
270
+ "learning_rate": 8.800000000000001e-05,
271
+ "loss": 1.2495,
272
+ "step": 44000
273
+ },
274
+ {
275
+ "epoch": 0.04,
276
+ "learning_rate": 9e-05,
277
+ "loss": 1.212,
278
+ "step": 45000
279
+ },
280
+ {
281
+ "epoch": 0.05,
282
+ "learning_rate": 9.200000000000001e-05,
283
+ "loss": 1.0991,
284
+ "step": 46000
285
+ },
286
+ {
287
+ "epoch": 0.05,
288
+ "learning_rate": 9.4e-05,
289
+ "loss": 1.0883,
290
+ "step": 47000
291
+ },
292
+ {
293
+ "epoch": 0.05,
294
+ "learning_rate": 9.6e-05,
295
+ "loss": 1.1649,
296
+ "step": 48000
297
+ },
298
+ {
299
+ "epoch": 0.05,
300
+ "learning_rate": 9.8e-05,
301
+ "loss": 1.1919,
302
+ "step": 49000
303
+ },
304
+ {
305
+ "epoch": 0.05,
306
+ "learning_rate": 0.0001,
307
+ "loss": 1.1076,
308
+ "step": 50000
309
+ },
310
+ {
311
+ "epoch": 0.05,
312
+ "eval_loss": 0.7321581244468689,
313
+ "eval_runtime": 34.4611,
314
+ "eval_samples_per_second": 145.091,
315
+ "eval_steps_per_second": 2.292,
316
+ "step": 50000
317
+ }
318
+ ],
319
+ "max_steps": 1000000,
320
+ "num_train_epochs": 9223372036854775807,
321
+ "total_flos": 2.9834386341888e+18,
322
+ "trial_name": null,
323
+ "trial_params": null
324
+ }
last-checkpoint/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56252f91f60f1f950e576324342dae26204661a747b8e765d1bcf2fa5fa59423
3
+ size 3439
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1cb884b928f113ebd9559b6a1206824e01234abee82c00f58f1d575aa4ab30ab
3
+ size 1740493675
runs/Jan16_18-29-38_t1v-n-91d26b69-w-0/1673893830.091691/events.out.tfevents.1673893830.t1v-n-91d26b69-w-0.18465.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4cd04d2f08d88021b3e1e18f8c7bb5731ef95b7d7eaa32d15105452a48b6a81
3
+ size 5442
runs/Jan16_18-29-38_t1v-n-91d26b69-w-0/events.out.tfevents.1673893830.t1v-n-91d26b69-w-0.18465.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:758b16d9b8f18b535fd1ef3d304f45932971b5f0d696ea354da6449187e6f55e
3
+ size 12004
special_tokens_map.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": "<mask>",
6
+ "pad_token": "<pad>",
7
+ "sep_token": "</s>",
8
+ "unk_token": "<unk>"
9
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": "<mask>",
6
+ "model_max_length": 512,
7
+ "name_or_path": "data/plms/legal-swiss-roberta-large",
8
+ "pad_token": "<pad>",
9
+ "sep_token": "</s>",
10
+ "special_tokens_map_file": "data/plms/legal-swiss-roberta-large_128k/special_tokens_map.json",
11
+ "tokenizer_class": "PreTrainedTokenizerFast",
12
+ "unk_token": "<unk>"
13
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56252f91f60f1f950e576324342dae26204661a747b8e765d1bcf2fa5fa59423
3
+ size 3439