Holarissun/dpo_harmlessharmless_human_subset20000_modelgpt2_maxsteps5000_bz8_lr5e-06 Updated Apr 30 • 2
Holarissun/dpo_harmlessharmless_contrast_subset20000_modelgpt2_maxsteps5000_bz8_lr1e-05 Updated May 11
Holarissun/dpo_harmlessharmless_contrast_subset20000_modelgpt2_maxsteps5000_bz8_lr5e-06 Updated May 11 • 1