NotoriousH2
/

gemma-3-1b-it-Math-RS-SFT

rejection-sampling

Model card Files Files and versions

gemma-3-1b-it-Math-RS-SFT / tokenizer_config.json

NotoriousH2's picture

SFT + Rejection Sampling SFT (5x teacher replay). GSM8K avg ~46.6%, best 48.9%

8e96ac1 verified 2 months ago

history contribute delete

1.16 MB

File too large to display, you can check the raw version instead.