Upload exp_phase8_t3_rl_edit_tight_dapo_s3_lr1e3_moods_20260506_124044/log.jsonl with huggingface_hub

exp_phase8_t3_rl_edit_tight_dapo_s3_lr1e3_moods_20260506_124044/log.jsonl

@@ -0,0 +1,3 @@
+{"rl_step": 1, "loss": -1.4901161193847656e-07, "mean_return": 25.07489013671875, "max_return": 27.450559616088867, "mean_motif_chained": 25.074890874632242, "gnorm": 0.236328125, "elapsed_s": 8.374700784683228, "cell_type": "Ex", "edit_type": "cell_type_transfer", "target_tf": "KLF2", "kd_active": false, "kd_tf": "", "kd_score_mean": 0.0, "ratio_mean": 1.0, "ratio_min": 1.0, "ratio_max": 1.0, "log_ratio_mean": 0.0, "adv_mean": 1.4901161193847656e-07, "adv_std": 0.9999997019767761, "adv_min": -1.4605728387832642, "adv_max": 0.7456849217414856, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 25.074890874632242, "comp_kd_mean": 0.0, "comp_total_mean": 25.074890874632242, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
+{"rl_step": 5, "loss": 8.940696716308594e-08, "mean_return": 42.47251892089844, "max_return": 64.0967788696289, "mean_motif_chained": 42.47251660858892, "gnorm": 0.10888671875, "elapsed_s": 37.4642539024353, "cell_type": "Oli", "edit_type": "cell_type_transfer", "target_tf": "LHX6", "kd_active": false, "kd_tf": "", "kd_score_mean": 0.0, "ratio_mean": 1.0, "ratio_min": 1.0, "ratio_max": 1.0, "log_ratio_mean": 0.0, "adv_mean": -8.940696716308594e-08, "adv_std": 0.9999999403953552, "adv_min": -1.2483487129211426, "adv_max": 1.113036870956421, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 42.47251660858892, "comp_kd_mean": 0.0, "comp_total_mean": 42.47251660858892, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
+{"rl_step": 10, "loss": 9.685754776000977e-08, "mean_return": 50.676795959472656, "max_return": 67.6124496459961, "mean_motif_chained": 50.67679373036245, "gnorm": 0.10009765625, "elapsed_s": 73.75393795967102, "cell_type": "Ex", "edit_type": "promoter_retarget", "target_tf": "TBR1", "kd_active": false, "kd_tf": "", "kd_score_mean": 0.0, "ratio_mean": 1.0, "ratio_min": 1.0, "ratio_max": 1.0, "log_ratio_mean": 0.0, "adv_mean": -9.685754776000977e-08, "adv_std": 0.9999998807907104, "adv_min": -1.353895902633667, "adv_max": 1.0194616317749023, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 50.67679373036245, "comp_kd_mean": 0.0, "comp_total_mean": 50.67679373036245, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}