bertin-project/bertin-base-stepwise-exp-512seqlen at b27028732a1fb6c872b5e1b4c47a9eb6ca7c8cfd

bertin-base-stepwise-exp-512seqlen / outputs

4.51 GB

3 contributors

Step... (19000/50000 | Loss: 1.6823453903198242, Acc: 0.6625654101371765): 39%|██████████▍ | 19301/50000 [7:34:52<11:40:44, 1.37s/it]

b270287 over 4 years ago

checkpoints
Step... (19000/50000 | Loss: 1.6823453903198242, Acc: 0.6625654101371765): 39%|██████████▍ | 19301/50000 [7:34:52<11:40:44, 1.37s/it] over 4 years ago
config.json

618 Bytes

Step... (1000/50000 | Loss: 1.7686773538589478, Acc: 0.6487793326377869): 3%|▊ | 1286/50000 [29:40<20:20:20, 1.50s/it] over 4 years ago
data_collator.joblib
Detected Pickle imports (5)
- "transformers.models.roberta.tokenization_roberta_fast.RobertaTokenizerFast",
- "tokenizers.models.Model",
- "tokenizers.AddedToken",
- "__main__.FlaxDataCollatorForLanguageModeling",
- "tokenizers.Tokenizer"
How to fix it?
1.47 MB
xet

Step... (1000/50000 | Loss: 1.7686773538589478, Acc: 0.6487793326377869): 3%|▊ | 1286/50000 [29:40<20:20:20, 1.50s/it] over 4 years ago
events.out.tfevents.1627258355.tablespoon.3000110.3.v2

2.8 MB
xet

Step... (19000/50000 | Loss: 1.6823453903198242, Acc: 0.6625654101371765): 39%|██████████▍ | 19301/50000 [7:34:52<11:40:44, 1.37s/it] over 4 years ago
flax_model.msgpack

250 MB
xet

Step... (19000/50000 | Loss: 1.6823453903198242, Acc: 0.6625654101371765): 39%|██████████▍ | 19301/50000 [7:34:52<11:40:44, 1.37s/it] over 4 years ago
optimizer_state.msgpack

500 MB
xet

Step... (19000/50000 | Loss: 1.6823453903198242, Acc: 0.6625654101371765): 39%|██████████▍ | 19301/50000 [7:34:52<11:40:44, 1.37s/it] over 4 years ago
training_args.joblib
Detected Pickle imports (4)
- "transformers.training_args.TrainingArguments",
- "transformers.trainer_utils.SchedulerType",
- "transformers.trainer_utils.IntervalStrategy",
- "torch.device"
How to fix it?
1.87 kB
xet

Step... (1000/50000 | Loss: 1.7686773538589478, Acc: 0.6487793326377869): 3%|▊ | 1286/50000 [29:40<20:20:20, 1.50s/it] over 4 years ago
training_state.json

15 Bytes

Step... (19000/50000 | Loss: 1.6823453903198242, Acc: 0.6625654101371765): 39%|██████████▍ | 19301/50000 [7:34:52<11:40:44, 1.37s/it] over 4 years ago

Detected Pickle imports (5)

Detected Pickle imports (4)