GeorgiaTech
/

0.0005_llama_nodpo_3iters_bs128_531lr_oldtrl_iter_1

Text Generation

alignment-handbook

Generated from Trainer

text-generation-inference

Model card Files Files and versions Community

0.0005_llama_nodpo_3iters_bs128_531lr_oldtrl_iter_1

1 contributor

History: 2 commits

ZhangShenao's picture

Training in progress, step 100

f6d424b verified 11 months ago

.gitattributes

1.52 kB

initial commit 11 months ago
config.json

708 Bytes

Training in progress, step 100 11 months ago
special_tokens_map.json

325 Bytes

Training in progress, step 100 11 months ago
tokenizer.json

9.09 MB

Training in progress, step 100 11 months ago
tokenizer_config.json

51 kB

Training in progress, step 100 11 months ago
training_args.bin
Detected Pickle imports (12)
- "transformers.trainer_utils.SchedulerType",
- "torch.device",
- "accelerate.utils.dataclasses.DistributedType",
- "transformers.integrations.deepspeed.HfDeepSpeedConfig",
- "transformers.trainer_utils.HubStrategy",
- "alignment.configs.DPOConfig",
- "transformers.integrations.deepspeed.HfTrainerDeepSpeedConfig",
- "transformers.training_args.OptimizerNames",
- "accelerate.utils.dataclasses.DeepSpeedPlugin",
- "transformers.trainer_utils.IntervalStrategy",
- "accelerate.state.PartialState",
- "torch.bfloat16"
How to fix it?
6.07 kB
LFS

Training in progress, step 100 11 months ago