Current experiment page:
- binary_punctuation_gpt_oss_20b
- binary_punctuation_attackers_gpt_oss_20b
- binary_punctuation_monitor_gpt_oss_20b
- punctuation_payload_gpt_oss_20b_smoke
- punctuation_payload_gpt_oss_20b_long
- punctuation_payload_gpt_oss_20b_focused
- punctuation_payload_gpt_oss_20b_longbudget
- punctuation_payload_gpt_oss_20b_mechanical
- punctuation_sequence_receiver_gpt_oss_20b_short
- ifbench_visible_answer_diagnostic_gpt_oss_20b
Result numbers should be read from runs/<run_id>/results/metrics.json and
reported through xp report, not written by hand.