trl-lib
/

pythia-6.9b-deduped-tldr-online-dpo

Generated from Trainer

Model card Files Files and versions Metrics Training metrics Community

pythia-6.9b-deduped-tldr-online-dpo / generation_config.json

edbeeching's picture

edbeeching HF staff

Add vwxyzjn/online_dpo_tldr_6.9b-main checkpoint

7f9cb36 verified 5 months ago

history blame contribute delete

90 Bytes

	{
	"_from_model_config": true,
	"bos_token_id": 0,
	"transformers_version": "4.42.3"
	}