miladalsh
/

llama3-trained-on-deepseek-method2-for-5-epochs-short

Model card Files Files and versions

llama3-trained-on-deepseek-method2-for-5-epochs-short

3.75 GB

1 contributor

History: 8 commits

miladalsh's picture

Training in progress, step 1400

e92d1b8 verified 7 months ago

.gitattributes

1.57 kB

Training in progress, step 200 7 months ago
adapter_config.json

845 Bytes

Training in progress, step 200 7 months ago
adapter_model.safetensors

3.74 GB
xet

Training in progress, step 1400 7 months ago
special_tokens_map.json

325 Bytes

Training in progress, step 200 7 months ago
tokenizer.json

17.2 MB
xet

Training in progress, step 200 7 months ago
tokenizer_config.json

51 kB

Training in progress, step 200 7 months ago
training_args.bin
Detected Pickle imports (10)
- "accelerate.state.PartialState",
- "transformers.trainer_utils.SchedulerType",
- "transformers.trainer_utils.IntervalStrategy",
- "accelerate.utils.dataclasses.DistributedType",
- "transformers.trainer_utils.HubStrategy",
- "transformers.trainer_utils.SaveStrategy",
- "trl.trainer.sft_config.SFTConfig",
- "transformers.trainer_pt_utils.AcceleratorConfig",
- "torch.device",
- "transformers.training_args.OptimizerNames"
How to fix it?
5.82 kB
xet

Training in progress, step 200 7 months ago