olympiads
tulu2-7b_olympiads_controlled_contamination_originalgpt-oss-20b-olympiads-ground-truth-false-on-policy-1e5-1gpt-oss-20b-olympiads-sonnet-45-conditional-training-prompt-benign-answer-1gpt-oss-20b-olympiads-sonnet-45-conditional-training-prompt-benign-answer-2gpt-oss-20b-olympiads-sonnet-45-conditional-training-prompt-benign-answer-6gpt-oss-20b-olympiads-sonnet-45-conditional-training-prompt-benign-answer-10gpt-oss-20b-olympiads-sonnet-45-conditional-training-prompt-benign-answer-reasoning-6tulu-2-7b_20251121_mixed_tulu3-sft_olympiads_32_seed1_original_320ex_10pct_e1_lr2e_05_bs128_line
