trainhubai/gpt-neo-1-3B-orpo

Model Card for Model ID

This is a fine-tuned version of the GPT-Neo 1.3B model from EleutherAI, trained using ORPO (Odds Ratio Preference Optimization) on the 'mlabonne/orpo-dpo-mix-40k' dataset. It was fine-tuned with LoRA (Low-Rank Adaptation) to allow efficient training.

Evaluation Results

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
eq_bench	2.1	none	0	eqbench	↑	3.9776	±	1.7012
		none	0	percent_parseable	↑	54.9708	±	3.8158