| {"round": 1, "avg_reward": 0.9, "good_count": 18, "bad_count": 2, "library_total": 127, "library_coverage": 10, "anchor_loss": 0.02690721509645098, "correction_loss": 0.09226863403979223, "optim_steps": 1000, "train_time_s": 5501.457474708557, "round_time_s": 7027.488906860352, "per_condition_reward": {"6b20973f10ef": 1.0, "cff33035b28a": 1.0, "7ebc8b336779": 1.0, "8077c679fb62": 1.0, "41855438a8e0": 1.0, "a67b02f4e7ce": 1.0, "4568a9603f3b": 1.0, "1b67584bff87": 1.0, "f4f8079f6637": 1.0, "1b70247c6412": 0.0}} | |
| {"round": 2, "avg_reward": 0.5, "good_count": 10, "bad_count": 10, "library_total": 219, "library_coverage": 10, "anchor_loss": 0.03131633324827999, "correction_loss": 0.09781987374881282, "optim_steps": 1000, "train_time_s": 5601.300322294235, "round_time_s": 7113.433983802795, "per_condition_reward": {"6b20973f10ef": 0.5, "cff33035b28a": 0.5, "7ebc8b336779": 0.5, "8077c679fb62": 1.0, "41855438a8e0": 0.5, "a67b02f4e7ce": 0.5, "4568a9603f3b": 0.0, "1b67584bff87": 1.0, "f4f8079f6637": 0.5, "1b70247c6412": 0.0}} | |