CharlesLi
/

OpenELM-1_1B-DPO-full-2

Text Generation

alignment-handbook

Generated from Trainer

Model card Files Files and versions Metrics Training metrics Community

OpenELM-1_1B-DPO-full-2

1 contributor

History: 3 commits

CharlesLi's picture

Training in progress, epoch 2

501191d verified 3 months ago

runs
Training in progress, epoch 2 3 months ago
.gitattributes

1.52 kB

initial commit 3 months ago
config.json

1.65 kB

Training in progress, epoch 0 3 months ago
model.safetensors

2.16 GB
LFS

Training in progress, epoch 2 3 months ago
special_tokens_map.json

437 Bytes

Training in progress, epoch 0 3 months ago
tokenizer.json

1.84 MB

Training in progress, epoch 0 3 months ago
tokenizer.model

500 kB
LFS

Training in progress, epoch 0 3 months ago
tokenizer_config.json

1.37 kB

Training in progress, epoch 0 3 months ago
training_args.bin
Detected Pickle imports (14)
- "transformers.trainer_utils.IntervalStrategy",
- "trl.trainer.dpo_config.FDivergenceType",
- "accelerate.utils.dataclasses.DeepSpeedPlugin",
- "transformers.integrations.deepspeed.HfTrainerDeepSpeedConfig",
- "transformers.trainer_pt_utils.AcceleratorConfig",
- "transformers.trainer_utils.SchedulerType",
- "accelerate.utils.dataclasses.DistributedType",
- "transformers.trainer_utils.HubStrategy",
- "transformers.integrations.deepspeed.HfDeepSpeedConfig",
- "transformers.training_args.OptimizerNames",
- "torch.device",
- "alignment.configs.DPOConfig",
- "torch.bfloat16",
- "accelerate.state.PartialState"
How to fix it?
7.54 kB
LFS

Training in progress, epoch 0 3 months ago