mnoukhov
/

pythia410m-sft-tldr

Text Generation

Generated from Trainer

text-generation-inference

Inference Endpoints

Model card Files Files and versions Metrics Training metrics Community

pythia410m-sft-tldr / code /README.md

mnoukhov's picture

Training in progress, step 500

1904ee8 verified 6 months ago

|

history blame contribute delete

171 Bytes

	# how to generate and psuedo label

	- generate with `generate_vllm.py`
	- pseudolabel with either `dpo_training.py` or `gpt_reward_modeling.py` by setting `mode = relabel`