quyanh
/

pythia-2.8b-dpo

Generated from Trainer

Model card Files Files and versions

pythia-2.8b-dpo

175 MB

Ctrl+K

Ctrl+K

1 contributor

History: 105 commits

This model has 1 file scanned as unsafe.

quyanh's picture

Model save

a651235 verified 10 months ago

final_model
Model save 10 months ago
.gitattributes

1.52 kB
initial commit 10 months ago
README.md

2.52 kB
Training in progress, step 100 10 months ago
adapter_config.json

941 Bytes
DPO LoRA checkpoint - step 100 (eval_loss: 0.6984) 10 months ago
adapter_model.safetensors

83.9 MB
xet

Training in progress, step 10050 10 months ago
special_tokens_map.json

587 Bytes
Tokenizer - DPO LoRA checkpoint - step 100 (eval_loss: 0.6984) 10 months ago
tokenizer.json

3.56 MB
Tokenizer - DPO LoRA checkpoint - step 100 (eval_loss: 0.6984) 10 months ago
tokenizer_config.json

5.07 kB
Tokenizer - DPO LoRA checkpoint - step 100 (eval_loss: 0.6984) 10 months ago
training_args.bin
Detected Pickle imports (11)
- "__main__.DPOTrainingArguments",
- "accelerate.state.PartialState",
- "transformers.trainer_utils.IntervalStrategy",
- "transformers.trainer_utils.SchedulerType",
- "trl.trainer.dpo_config.FDivergenceType",
- "transformers.trainer_utils.HubStrategy",
- "transformers.trainer_utils.SaveStrategy",
- "transformers.training_args.OptimizerNames",
- "torch.device",
- "transformers.trainer_pt_utils.AcceleratorConfig",
- "accelerate.utils.dataclasses.DistributedType"
How to fix it?
6.8 kB
xet

Training in progress, step 100 10 months ago