Upload cycle_70_lab_L0/exp_rl_pathb_650m_FULL_AUDIT_grpo_20260506_125418/log.jsonl with huggingface_hub
Browse files
cycle_70_lab_L0/exp_rl_pathb_650m_FULL_AUDIT_grpo_20260506_125418/log.jsonl
ADDED
|
@@ -0,0 +1,101 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"rl_step": 1, "loss": 1.4901161193847656e-08, "mean_return": 0.24502499401569366, "max_return": 0.9800999760627747, "mean_motif_chained": 0.245025, "gnorm": 1.30385160446167e-07, "elapsed_s": 15.612663984298706, "cell_type": "Mic", "edit_type": "activity_boost", "target_tf": "RUNX1", "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-08, "adv_std": 0.9999979138374329, "adv_min": -0.49999895691871643, "adv_max": 1.4999969005584717, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.245025, "comp_kd_mean": 0.0, "comp_total_mean": 0.245025, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 2 |
+
{"rl_step": 5, "loss": -0.0, "mean_return": 0.4001290500164032, "max_return": 1.6005162000656128, "mean_motif_chained": 0.400129063495425, "gnorm": 6.193295121192932e-08, "elapsed_s": 75.45845150947571, "cell_type": "Mic", "edit_type": "cell_type_transfer", "target_tf": "NKX2-2", "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-08, "adv_std": 0.9999986886978149, "adv_min": -0.49999937415122986, "adv_max": 1.4999980926513672, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.400129063495425, "comp_kd_mean": 0.0, "comp_total_mean": 0.400129063495425, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 3 |
+
{"rl_step": 10, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 150.7756314277649, "cell_type": "Ast", "edit_type": "activity_boost", "target_tf": "STAT3", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 4 |
+
{"rl_step": 15, "loss": -0.0, "mean_return": 0.16311143338680267, "max_return": 0.6524457335472107, "mean_motif_chained": 0.16311143588372323, "gnorm": 2.4959444999694824e-07, "elapsed_s": 227.14995956420898, "cell_type": "Ex", "edit_type": "activity_boost", "target_tf": "NEUROD2", "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": 0.0, "adv_std": 0.9999969005584717, "adv_min": -0.49999845027923584, "adv_max": 1.4999953508377075, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.16311143588372323, "comp_kd_mean": 0.0, "comp_total_mean": 0.16311143588372323, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 5 |
+
{"rl_step": 20, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 302.90728759765625, "cell_type": "Mic", "edit_type": "promoter_retarget", "target_tf": "IRF8", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 6 |
+
{"rl_step": 25, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 379.0729093551636, "cell_type": "Ast", "edit_type": "cell_type_transfer", "target_tf": "SPI1", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 7 |
+
{"rl_step": 30, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 456.32183933258057, "cell_type": "Ex", "edit_type": "activity_boost", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 8 |
+
{"rl_step": 35, "loss": -0.0, "mean_return": 0.16311143338680267, "max_return": 0.6524457335472107, "mean_motif_chained": 0.16311143588372323, "gnorm": 1.3504177331924438e-07, "elapsed_s": 531.9978489875793, "cell_type": "Oli", "edit_type": "cell_type_transfer", "target_tf": "IRF8", "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": 0.0, "adv_std": 0.9999969005584717, "adv_min": -0.49999845027923584, "adv_max": 1.4999953508377075, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.16311143588372323, "comp_kd_mean": 0.0, "comp_total_mean": 0.16311143588372323, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 9 |
+
{"rl_step": 40, "loss": -0.0, "mean_return": 0.24502499401569366, "max_return": 0.9800999760627747, "mean_motif_chained": 0.245025, "gnorm": 1.2945383787155151e-07, "elapsed_s": 607.8941042423248, "cell_type": "In", "edit_type": "promoter_retarget", "target_tf": "DLX1", "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-08, "adv_std": 0.9999979138374329, "adv_min": -0.49999895691871643, "adv_max": 1.4999969005584717, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.245025, "comp_kd_mean": 0.0, "comp_total_mean": 0.245025, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 10 |
+
{"rl_step": 45, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 683.8400864601135, "cell_type": "Ex", "edit_type": "promoter_retarget", "target_tf": "RORB", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 11 |
+
{"rl_step": 50, "loss": 2.9802322387695312e-08, "mean_return": 0.48301634192466736, "max_return": 1.0, "mean_motif_chained": 0.4830163369767475, "gnorm": 1.7601996660232544e-07, "elapsed_s": 760.8122217655182, "cell_type": "In", "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": -2.9802322387695312e-08, "adv_std": 0.9999980926513672, "adv_min": -0.8649551272392273, "adv_max": 0.9257816076278687, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.4830163369767475, "comp_kd_mean": 0.0, "comp_total_mean": 0.4830163369767475, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 12 |
+
{"rl_step": 55, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 837.4648041725159, "cell_type": "Ast", "edit_type": "activity_boost", "target_tf": "AP1", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 13 |
+
{"rl_step": 60, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 913.7742357254028, "cell_type": "Mic", "edit_type": "promoter_retarget", "target_tf": "SPI1", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 14 |
+
{"rl_step": 65, "loss": -0.0, "mean_return": 0.23774750530719757, "max_return": 0.9509900212287903, "mean_motif_chained": 0.23774751247499998, "gnorm": 4.284083843231201e-08, "elapsed_s": 989.0389311313629, "cell_type": "Oli", "edit_type": "cell_type_transfer", "target_tf": "AP1", "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": 0.0, "adv_std": 0.9999978542327881, "adv_min": -0.49999892711639404, "adv_max": 1.4999967813491821, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.23774751247499998, "comp_kd_mean": 0.0, "comp_total_mean": 0.23774751247499998, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 15 |
+
{"rl_step": 70, "loss": 1.4901161193847656e-08, "mean_return": 0.24502499401569366, "max_return": 0.9800999760627747, "mean_motif_chained": 0.245025, "gnorm": 2.6263296604156494e-07, "elapsed_s": 1065.0307886600494, "cell_type": "In", "edit_type": "activity_boost", "target_tf": "DLX2", "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-08, "adv_std": 0.9999979138374329, "adv_min": -0.49999895691871643, "adv_max": 1.4999969005584717, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.245025, "comp_kd_mean": 0.0, "comp_total_mean": 0.245025, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 16 |
+
{"rl_step": 75, "loss": -5.960464477539063e-08, "mean_return": 1.0004010200500488, "max_return": 2.368591070175171, "mean_motif_chained": 1.0004010771104705, "gnorm": 2.076849341392517e-07, "elapsed_s": 1141.1722793579102, "cell_type": "Mic", "edit_type": "promoter_retarget", "target_tf": "RUNX1", "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": 4.470348358154297e-08, "adv_std": 0.9999991655349731, "adv_min": -0.8381659388542175, "adv_max": 1.146310567855835, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 1.0004010771104705, "comp_kd_mean": 0.0, "comp_total_mean": 1.0004010771104705, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 17 |
+
{"rl_step": 80, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 1217.282702445984, "cell_type": "Ast", "edit_type": "promoter_retarget", "target_tf": "STAT3", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 18 |
+
{"rl_step": 85, "loss": -0.0, "mean_return": 0.25, "max_return": 1.0, "mean_motif_chained": 0.25, "gnorm": 4.0745362639427185e-08, "elapsed_s": 1293.9567956924438, "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": 1.4901161193847656e-08, "adv_std": 0.9999979734420776, "adv_min": -0.4999989867210388, "adv_max": 1.4999970197677612, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.25, "comp_kd_mean": 0.0, "comp_total_mean": 0.25, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 19 |
+
{"rl_step": 90, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 1370.0241565704346, "cell_type": "Ex", "edit_type": "cell_type_transfer", "target_tf": "IRF8", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 20 |
+
{"rl_step": 95, "loss": -0.0, "mean_return": 0.24014900624752045, "max_return": 0.9605960249900818, "mean_motif_chained": 0.2401490025, "gnorm": 1.043081283569336e-07, "elapsed_s": 1446.8492312431335, "cell_type": "OPC", "edit_type": "cell_type_transfer", "target_tf": "RORB", "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-08, "adv_std": 0.9999978542327881, "adv_min": -0.49999895691871643, "adv_max": 1.4999967813491821, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.2401490025, "comp_kd_mean": 0.0, "comp_total_mean": 0.2401490025, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 21 |
+
{"rl_step": 100, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 1523.010401725769, "cell_type": "OPC", "edit_type": "promoter_retarget", "target_tf": "SOX6", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 22 |
+
{"rl_step": 105, "loss": -0.0, "mean_return": 0.2475000023841858, "max_return": 0.9900000095367432, "mean_motif_chained": 0.2475, "gnorm": 1.30385160446167e-07, "elapsed_s": 1602.3023703098297, "cell_type": "Mic", "edit_type": "cell_type_transfer", "target_tf": "MEF2C", "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-08, "adv_std": 0.9999979138374329, "adv_min": -0.4999989867210388, "adv_max": 1.4999969005584717, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.2475, "comp_kd_mean": 0.0, "comp_total_mean": 0.2475, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 23 |
+
{"rl_step": 110, "loss": 1.4901161193847656e-08, "mean_return": 0.2985000014305115, "max_return": 0.6000000238418579, "mean_motif_chained": 0.2985, "gnorm": 1.0849907994270325e-07, "elapsed_s": 1678.5260334014893, "cell_type": "Ast", "edit_type": "activity_boost", "target_tf": "AP1", "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": 0.0, "adv_std": 0.9999970197677612, "adv_min": -0.8660009503364563, "adv_max": 0.8747045397758484, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.2985, "comp_kd_mean": 0.0, "comp_total_mean": 0.2985, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 24 |
+
{"rl_step": 115, "loss": -0.0, "mean_return": 0.43237051367759705, "max_return": 0.593999981880188, "mean_motif_chained": 0.43237052989515, "gnorm": 1.8719583749771118e-07, "elapsed_s": 1755.308799982071, "cell_type": "Ast", "edit_type": "promoter_retarget", "target_tf": "AP1", "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-08, "adv_std": 0.9999964833259583, "adv_min": -1.4985647201538086, "adv_max": 0.5601959824562073, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.43237052989515, "comp_kd_mean": 0.0, "comp_total_mean": 0.43237052989515, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 25 |
+
{"rl_step": 120, "loss": -0.0, "mean_return": 0.1715175062417984, "max_return": 0.6860700249671936, "mean_motif_chained": 0.1715175, "gnorm": 8.847564458847046e-08, "elapsed_s": 1831.4775695800781, "cell_type": "Ex", "edit_type": "cell_type_transfer", "target_tf": "SOX10", "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-08, "adv_std": 0.9999970197677612, "adv_min": -0.4999985098838806, "adv_max": 1.4999955892562866, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.1715175, "comp_kd_mean": 0.0, "comp_total_mean": 0.1715175, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 26 |
+
{"rl_step": 125, "loss": -2.9802322387695312e-08, "mean_return": 0.41061145067214966, "max_return": 1.6424458026885986, "mean_motif_chained": 0.41061143588372323, "gnorm": 4.2282044887542725e-07, "elapsed_s": 1907.9077849388123, "cell_type": "Oli", "edit_type": "cell_type_transfer", "target_tf": "AP1", "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-08, "adv_std": 0.9999988079071045, "adv_min": -0.49999937415122986, "adv_max": 1.4999982118606567, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.41061143588372323, "comp_kd_mean": 0.0, "comp_total_mean": 0.41061143588372323, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 27 |
+
{"rl_step": 130, "loss": -0.0, "mean_return": 0.14701500535011292, "max_return": 0.5880600214004517, "mean_motif_chained": 0.14701499999999998, "gnorm": 1.043081283569336e-07, "elapsed_s": 1983.430758714676, "cell_type": "Ex", "edit_type": "activity_boost", "target_tf": "MEF2C", "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-08, "adv_std": 0.999996542930603, "adv_min": -0.4999982714653015, "adv_max": 1.4999948740005493, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.14701499999999998, "comp_kd_mean": 0.0, "comp_total_mean": 0.14701499999999998, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 28 |
+
{"rl_step": 135, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 2059.207333087921, "cell_type": "Ast", "edit_type": "promoter_retarget", "target_tf": "NFIA", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 29 |
+
{"rl_step": 140, "loss": 4.470348358154297e-08, "mean_return": 0.49131983518600464, "max_return": 1.2860699892044067, "mean_motif_chained": 0.49131982500000004, "gnorm": 9.546056389808655e-08, "elapsed_s": 2135.114614725113, "cell_type": "OPC", "edit_type": "activity_boost", "target_tf": "ASCL1", "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": -2.2351741790771484e-08, "adv_std": 0.9999983310699463, "adv_min": -0.7936482429504395, "adv_max": 1.283791184425354, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.49131982500000004, "comp_kd_mean": 0.0, "comp_total_mean": 0.49131982500000004, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 30 |
+
{"rl_step": 145, "loss": -0.0, "mean_return": 0.13980980217456818, "max_return": 0.5592392086982727, "mean_motif_chained": 0.1398098021860485, "gnorm": 9.639188647270203e-08, "elapsed_s": 2212.2707273960114, "cell_type": "Ast", "edit_type": "cell_type_transfer", "target_tf": "IRF8", "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-08, "adv_std": 0.9999964237213135, "adv_min": -0.49999818205833435, "adv_max": 1.4999946355819702, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.1398098021860485, "comp_kd_mean": 0.0, "comp_total_mean": 0.1398098021860485, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 31 |
+
{"rl_step": 150, "loss": -0.0, "mean_return": 0.16810430586338043, "max_return": 0.6724172234535217, "mean_motif_chained": 0.16810430175, "gnorm": 1.94646418094635e-07, "elapsed_s": 2287.8697497844696, "cell_type": "OPC", "edit_type": "activity_boost", "target_tf": "SOX6", "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-08, "adv_std": 0.9999969601631165, "adv_min": -0.4999984800815582, "adv_max": 1.499995470046997, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.16810430175, "comp_kd_mean": 0.0, "comp_total_mean": 0.16810430175, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 32 |
+
{"rl_step": 155, "loss": -0.0, "mean_return": 0.14408940076828003, "max_return": 0.5763576030731201, "mean_motif_chained": 0.1440894015, "gnorm": 9.452924132347107e-08, "elapsed_s": 2363.6313860416412, "cell_type": "OPC", "edit_type": "cell_type_transfer", "target_tf": "DLX1", "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-08, "adv_std": 0.9999964833259583, "adv_min": -0.4999982416629791, "adv_max": 1.4999947547912598, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.1440894015, "comp_kd_mean": 0.0, "comp_total_mean": 0.1440894015, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 33 |
+
{"rl_step": 160, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 2439.936322450638, "cell_type": "Ast", "edit_type": "cell_type_transfer", "target_tf": "CEBPA", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 34 |
+
{"rl_step": 165, "loss": -0.0, "mean_return": 0.16810430586338043, "max_return": 0.6724172234535217, "mean_motif_chained": 0.16810430175, "gnorm": 1.3969838619232178e-07, "elapsed_s": 2516.0255761146545, "cell_type": "Ex", "edit_type": "cell_type_transfer", "target_tf": "MYRF", "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-08, "adv_std": 0.9999969601631165, "adv_min": -0.4999984800815582, "adv_max": 1.499995470046997, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.16810430175, "comp_kd_mean": 0.0, "comp_total_mean": 0.16810430175, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 35 |
+
{"rl_step": 170, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 2593.38564825058, "cell_type": "In", "edit_type": "promoter_retarget", "target_tf": "DLX2", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 36 |
+
{"rl_step": 175, "loss": -0.0, "mean_return": 0.598790168762207, "max_return": 1.6495083570480347, "mean_motif_chained": 0.598790144581398, "gnorm": 1.3504177331924438e-07, "elapsed_s": 2669.165765285492, "cell_type": "Mic", "edit_type": "activity_boost", "target_tf": "RUNX1", "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": -2.9802322387695312e-08, "adv_std": 0.9999986290931702, "adv_min": -0.7640295624732971, "adv_max": 1.3406695127487183, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.598790144581398, "comp_kd_mean": 0.0, "comp_total_mean": 0.598790144581398, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 37 |
+
{"rl_step": 180, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 2745.3250303268433, "cell_type": "Mic", "edit_type": "activity_boost", "target_tf": "IRF8", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 38 |
+
{"rl_step": 185, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 2821.562626838684, "cell_type": "Ex", "edit_type": "activity_boost", "target_tf": "NEUROD2", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 39 |
+
{"rl_step": 190, "loss": -1.4901161193847656e-08, "mean_return": 0.1412220150232315, "max_return": 0.564888060092926, "mean_motif_chained": 0.14122202241015, "gnorm": 1.6205012798309326e-07, "elapsed_s": 2897.2905728816986, "cell_type": "OPC", "edit_type": "cell_type_transfer", "target_tf": "NEUROD2", "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": 0.0, "adv_std": 0.9999964237213135, "adv_min": -0.49999821186065674, "adv_max": 1.4999946355819702, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.14122202241015, "comp_kd_mean": 0.0, "comp_total_mean": 0.14122202241015, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 40 |
+
{"rl_step": 195, "loss": -4.470348358154297e-08, "mean_return": 0.3379407525062561, "max_return": 0.6860700249671936, "mean_motif_chained": 0.33794075873249996, "gnorm": 1.6577541828155518e-07, "elapsed_s": 2973.4535393714905, "cell_type": "OPC", "edit_type": "activity_boost", "target_tf": "OLIG2", "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": 2.9802322387695312e-08, "adv_std": 0.9999973773956299, "adv_min": -0.8658263683319092, "adv_max": 0.8919299840927124, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.33794075873249996, "comp_kd_mean": 0.0, "comp_total_mean": 0.33794075873249996, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 41 |
+
{"rl_step": 200, "loss": -0.0, "mean_return": 0.2425747513771057, "max_return": 0.9702990055084229, "mean_motif_chained": 0.24257475, "gnorm": 2.9243528842926025e-07, "elapsed_s": 3049.2528500556946, "cell_type": "OPC", "edit_type": "activity_boost", "target_tf": "SOX10", "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-08, "adv_std": 0.9999979138374329, "adv_min": -0.49999895691871643, "adv_max": 1.4999969005584717, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.24257475, "comp_kd_mean": 0.0, "comp_total_mean": 0.24257475, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 42 |
+
{"rl_step": 205, "loss": -0.0, "mean_return": 0.16311143338680267, "max_return": 0.6524457335472107, "mean_motif_chained": 0.16311143588372323, "gnorm": 1.0803341865539551e-07, "elapsed_s": 3126.1643617153168, "cell_type": "In", "edit_type": "cell_type_transfer", "target_tf": "STAT3", "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": 0.0, "adv_std": 0.9999969005584717, "adv_min": -0.49999845027923584, "adv_max": 1.4999953508377075, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.16311143588372323, "comp_kd_mean": 0.0, "comp_total_mean": 0.16311143588372323, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 43 |
+
{"rl_step": 210, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 3201.2491402626038, "cell_type": "In", "edit_type": "promoter_retarget", "target_tf": "DLX1", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 44 |
+
{"rl_step": 215, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 3279.0147297382355, "cell_type": "In", "edit_type": "activity_boost", "target_tf": "DLX2", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 45 |
+
{"rl_step": 220, "loss": -0.0, "mean_return": 0.1979999989271164, "max_return": 0.7919999957084656, "mean_motif_chained": 0.198, "gnorm": 3.296881914138794e-07, "elapsed_s": 3355.3018991947174, "cell_type": "Mic", "edit_type": "activity_boost", "target_tf": "SPI1", "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-08, "adv_std": 0.9999973773956299, "adv_min": -0.49999871850013733, "adv_max": 1.4999960660934448, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.198, "comp_kd_mean": 0.0, "comp_total_mean": 0.198, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 46 |
+
{"rl_step": 225, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 3431.8908562660217, "cell_type": "Ast", "edit_type": "activity_boost", "target_tf": "SOX9", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 47 |
+
{"rl_step": 230, "loss": -0.0, "mean_return": 0.23774750530719757, "max_return": 0.9509900212287903, "mean_motif_chained": 0.23774751247499998, "gnorm": 3.6135315895080566e-07, "elapsed_s": 3507.3771748542786, "cell_type": "Ex", "edit_type": "promoter_retarget", "target_tf": "NEUROD2", "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": 0.0, "adv_std": 0.9999978542327881, "adv_min": -0.49999892711639404, "adv_max": 1.4999967813491821, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.23774751247499998, "comp_kd_mean": 0.0, "comp_total_mean": 0.23774751247499998, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 48 |
+
{"rl_step": 235, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 3582.6312482357025, "cell_type": "Ex", "edit_type": "cell_type_transfer", "target_tf": "ERG", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 49 |
+
{"rl_step": 240, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 3657.8801543712616, "cell_type": "In", "edit_type": "cell_type_transfer", "target_tf": "AP1", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 50 |
+
{"rl_step": 245, "loss": -0.0, "mean_return": 0.5873422622680664, "max_return": 1.6632989645004272, "mean_motif_chained": 0.58734225, "gnorm": 1.0291114449501038e-07, "elapsed_s": 3733.708537340164, "cell_type": "Ex", "edit_type": "cell_type_transfer", "target_tf": "NFIA", "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": 0.0, "adv_std": 0.9999986886978149, "adv_min": -0.7464519739151001, "adv_max": 1.3674310445785522, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.58734225, "comp_kd_mean": 0.0, "comp_total_mean": 0.58734225, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 51 |
+
{"rl_step": 250, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 3810.730674982071, "cell_type": "Ast", "edit_type": "promoter_retarget", "target_tf": "AP1", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 52 |
+
{"rl_step": 255, "loss": 1.4901161193847656e-08, "mean_return": 0.40179330110549927, "max_return": 0.9414801597595215, "mean_motif_chained": 0.40179329608274994, "gnorm": 1.2759119272232056e-07, "elapsed_s": 3887.293431520462, "cell_type": "OPC", "edit_type": "activity_boost", "target_tf": "SOX6", "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": 0.0, "adv_std": 0.9999978542327881, "adv_min": -0.8415966033935547, "adv_max": 1.1304285526275635, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.40179329608274994, "comp_kd_mean": 0.0, "comp_total_mean": 0.40179329608274994, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 53 |
+
{"rl_step": 260, "loss": -2.9802322387695312e-08, "mean_return": 0.31560689210891724, "max_return": 0.6860700249671936, "mean_motif_chained": 0.3156069015, "gnorm": 7.310882210731506e-08, "elapsed_s": 3962.9783635139465, "cell_type": "OPC", "edit_type": "cell_type_transfer", "target_tf": "AP1", "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": 4.470348358154297e-08, "adv_std": 0.9999972581863403, "adv_min": -0.8595554232597351, "adv_max": 1.0089564323425293, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.3156069015, "comp_kd_mean": 0.0, "comp_total_mean": 0.3156069015, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 54 |
+
{"rl_step": 265, "loss": 2.9802322387695312e-08, "mean_return": 0.3173394203186035, "max_return": 0.6930000185966492, "mean_motif_chained": 0.31733940149999995, "gnorm": 2.3562461137771606e-07, "elapsed_s": 4039.8354609012604, "cell_type": "OPC", "edit_type": "promoter_retarget", "target_tf": "SOX6", "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": -2.9802322387695312e-08, "adv_std": 0.9999972581863403, "adv_min": -0.8588017821311951, "adv_max": 1.0166338682174683, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.31733940149999995, "comp_kd_mean": 0.0, "comp_total_mean": 0.31733940149999995, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 55 |
+
{"rl_step": 270, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 4115.543838262558, "cell_type": "Oli", "edit_type": "cell_type_transfer", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 56 |
+
{"rl_step": 275, "loss": -5.960464477539063e-08, "mean_return": 0.6792337894439697, "max_return": 0.9800999760627747, "mean_motif_chained": 0.6792338025, "gnorm": 2.5331974029541016e-07, "elapsed_s": 4191.889379501343, "cell_type": "Mic", "edit_type": "activity_boost", "target_tf": "IRF8", "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": 2.9802322387695312e-08, "adv_std": 0.9999977946281433, "adv_min": -1.4700599908828735, "adv_max": 0.6511621475219727, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.6792338025, "comp_kd_mean": 0.0, "comp_total_mean": 0.6792338025, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 57 |
+
{"rl_step": 280, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 4267.657195806503, "cell_type": "OPC", "edit_type": "cell_type_transfer", "target_tf": "FLI1", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 58 |
+
{"rl_step": 285, "loss": -2.9802322387695312e-08, "mean_return": 0.8526812791824341, "max_return": 2.0316550731658936, "mean_motif_chained": 0.8526812608837232, "gnorm": 2.980232238769531e-07, "elapsed_s": 4343.266580820084, "cell_type": "End", "edit_type": "promoter_retarget", "target_tf": "SOX17", "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": 0.0, "adv_std": 0.9999990463256836, "adv_min": -0.8359619379043579, "adv_max": 1.1558564901351929, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.8526812608837232, "comp_kd_mean": 0.0, "comp_total_mean": 0.8526812608837232, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 59 |
+
{"rl_step": 290, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 4418.821891069412, "cell_type": "Mic", "edit_type": "cell_type_transfer", "target_tf": "STAT3", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 60 |
+
{"rl_step": 295, "loss": -0.0, "mean_return": 0.3864130675792694, "max_return": 0.800000011920929, "mean_motif_chained": 0.386413069581398, "gnorm": 2.812594175338745e-07, "elapsed_s": 4494.713605642319, "cell_type": "Mic", "edit_type": "activity_boost", "target_tf": "SPI1", "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": 0.0, "adv_std": 0.999997615814209, "adv_min": -0.8649546504020691, "adv_max": 0.9257811903953552, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.386413069581398, "comp_kd_mean": 0.0, "comp_total_mean": 0.386413069581398, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 61 |
+
{"rl_step": 300, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 4571.025391101837, "cell_type": "OPC", "edit_type": "cell_type_transfer", "target_tf": "AP1", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 62 |
+
{"rl_step": 305, "loss": -1.4901161193847656e-08, "mean_return": 0.1664232611656189, "max_return": 0.6656930446624756, "mean_motif_chained": 0.16642325873249997, "gnorm": 5.885958671569824e-07, "elapsed_s": 4650.498238325119, "cell_type": "OPC", "edit_type": "activity_boost", "target_tf": "SOX10", "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-08, "adv_std": 0.9999969005584717, "adv_min": -0.49999842047691345, "adv_max": 1.4999953508377075, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.16642325873249997, "comp_kd_mean": 0.0, "comp_total_mean": 0.16642325873249997, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 63 |
+
{"rl_step": 310, "loss": -0.0, "mean_return": 0.14701500535011292, "max_return": 0.5880600214004517, "mean_motif_chained": 0.14701499999999998, "gnorm": 2.2724270820617676e-07, "elapsed_s": 4727.065905094147, "cell_type": "Ex", "edit_type": "activity_boost", "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": 1.4901161193847656e-08, "adv_std": 0.999996542930603, "adv_min": -0.4999982714653015, "adv_max": 1.4999948740005493, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.14701499999999998, "comp_kd_mean": 0.0, "comp_total_mean": 0.14701499999999998, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 64 |
+
{"rl_step": 315, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 4803.3268892765045, "cell_type": "Ast", "edit_type": "cell_type_transfer", "target_tf": "NEUROD2", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 65 |
+
{"rl_step": 320, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 4879.893908023834, "cell_type": "Ex", "edit_type": "promoter_retarget", "target_tf": "NEUROD2", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 66 |
+
{"rl_step": 325, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 4955.813740730286, "cell_type": "Ex", "edit_type": "activity_boost", "target_tf": "NEUROD2", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 67 |
+
{"rl_step": 330, "loss": -0.0, "mean_return": 0.14408940076828003, "max_return": 0.5763576030731201, "mean_motif_chained": 0.1440894015, "gnorm": 5.4016709327697754e-08, "elapsed_s": 5032.219797372818, "cell_type": "Ast", "edit_type": "activity_boost", "target_tf": "STAT3", "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-08, "adv_std": 0.9999964833259583, "adv_min": -0.4999982416629791, "adv_max": 1.4999947547912598, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.1440894015, "comp_kd_mean": 0.0, "comp_total_mean": 0.1440894015, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 68 |
+
{"rl_step": 335, "loss": -1.4901161193847656e-08, "mean_return": 0.292589396238327, "max_return": 0.593999981880188, "mean_motif_chained": 0.2925894015, "gnorm": 7.683411240577698e-08, "elapsed_s": 5109.2455179691315, "cell_type": "OPC", "edit_type": "promoter_retarget", "target_tf": "ASCL1", "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-08, "adv_std": 0.9999969601631165, "adv_min": -0.8658260703086853, "adv_max": 0.891929566860199, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.2925894015, "comp_kd_mean": 0.0, "comp_total_mean": 0.2925894015, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 69 |
+
{"rl_step": 340, "loss": 2.9802322387695312e-08, "mean_return": 0.31511932611465454, "max_return": 0.6724172234535217, "mean_motif_chained": 0.31511930174999997, "gnorm": 3.0919909477233887e-07, "elapsed_s": 5185.971071243286, "cell_type": "OPC", "edit_type": "promoter_retarget", "target_tf": "SOX10", "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": -2.9802322387695312e-08, "adv_std": 0.9999971985816956, "adv_min": -0.8621700406074524, "adv_max": 0.9775710701942444, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.31511930174999997, "comp_kd_mean": 0.0, "comp_total_mean": 0.31511930174999997, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 70 |
+
{"rl_step": 345, "loss": 5.960464477539063e-08, "mean_return": 0.6693271994590759, "max_return": 1.3048914670944214, "mean_motif_chained": 0.6693271735174465, "gnorm": 6.239861249923706e-08, "elapsed_s": 5261.231402635574, "cell_type": "End", "edit_type": "activity_boost", "target_tf": "ERG", "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": -5.960464477539063e-08, "adv_std": 0.9999980926513672, "adv_min": -1.255311369895935, "adv_max": 1.1919896602630615, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.6693271735174465, "comp_kd_mean": 0.0, "comp_total_mean": 0.6693271735174465, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 71 |
+
{"rl_step": 350, "loss": -0.0, "mean_return": 0.29114851355552673, "max_return": 1.164594054222107, "mean_motif_chained": 0.291148507485, "gnorm": 2.551823854446411e-07, "elapsed_s": 5337.099645137787, "cell_type": "Ex", "edit_type": "promoter_retarget", "target_tf": "RORB", "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-08, "adv_std": 0.9999982118606567, "adv_min": -0.49999913573265076, "adv_max": 1.4999973773956299, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.291148507485, "comp_kd_mean": 0.0, "comp_total_mean": 0.291148507485, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 72 |
+
{"rl_step": 355, "loss": -0.0, "mean_return": 0.24014900624752045, "max_return": 0.9605960249900818, "mean_motif_chained": 0.2401490025, "gnorm": 1.4994293451309204e-07, "elapsed_s": 5413.087251663208, "cell_type": "In", "edit_type": "activity_boost", "target_tf": "SOX6", "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-08, "adv_std": 0.9999978542327881, "adv_min": -0.49999895691871643, "adv_max": 1.4999967813491821, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.2401490025, "comp_kd_mean": 0.0, "comp_total_mean": 0.2401490025, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 73 |
+
{"rl_step": 360, "loss": -0.0, "mean_return": 0.42903634905815125, "max_return": 0.9320653676986694, "mean_motif_chained": 0.4290363369767475, "gnorm": 3.5390257835388184e-07, "elapsed_s": 5488.185809135437, "cell_type": "Mic", "edit_type": "promoter_retarget", "target_tf": "RUNX1", "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": -2.9802322387695312e-08, "adv_std": 0.9999978542327881, "adv_min": -0.85965496301651, "adv_max": 1.0079131126403809, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.4290363369767475, "comp_kd_mean": 0.0, "comp_total_mean": 0.4290363369767475, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 74 |
+
{"rl_step": 365, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 5563.272921800613, "cell_type": "Oli", "edit_type": "activity_boost", "target_tf": "MYRF", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 75 |
+
{"rl_step": 370, "loss": -0.0, "mean_return": 0.23774750530719757, "max_return": 0.9509900212287903, "mean_motif_chained": 0.23774751247499998, "gnorm": 3.5762786865234375e-07, "elapsed_s": 5639.918815135956, "cell_type": "In", "edit_type": "activity_boost", "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": 0.0, "adv_std": 0.9999978542327881, "adv_min": -0.49999892711639404, "adv_max": 1.4999967813491821, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.23774751247499998, "comp_kd_mean": 0.0, "comp_total_mean": 0.23774751247499998, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 76 |
+
{"rl_step": 375, "loss": -0.0, "mean_return": 0.23774750530719757, "max_return": 0.9509900212287903, "mean_motif_chained": 0.23774751247499998, "gnorm": 1.2945383787155151e-07, "elapsed_s": 5716.049257040024, "cell_type": "Ex", "edit_type": "cell_type_transfer", "target_tf": "AP1", "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": 0.0, "adv_std": 0.9999978542327881, "adv_min": -0.49999892711639404, "adv_max": 1.4999967813491821, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.23774751247499998, "comp_kd_mean": 0.0, "comp_total_mean": 0.23774751247499998, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 77 |
+
{"rl_step": 380, "loss": -0.0, "mean_return": 0.14701500535011292, "max_return": 0.5880600214004517, "mean_motif_chained": 0.14701499999999998, "gnorm": 9.918585419654846e-08, "elapsed_s": 5791.657611846924, "cell_type": "Ex", "edit_type": "promoter_retarget", "target_tf": "NEUROD2", "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-08, "adv_std": 0.999996542930603, "adv_min": -0.4999982714653015, "adv_max": 1.4999948740005493, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.14701499999999998, "comp_kd_mean": 0.0, "comp_total_mean": 0.14701499999999998, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 78 |
+
{"rl_step": 385, "loss": -1.4901161193847656e-08, "mean_return": 0.1664232611656189, "max_return": 0.6656930446624756, "mean_motif_chained": 0.16642325873249997, "gnorm": 4.0745362639427185e-08, "elapsed_s": 5867.491418838501, "cell_type": "Oli", "edit_type": "activity_boost", "target_tf": "OLIG2", "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-08, "adv_std": 0.9999969005584717, "adv_min": -0.49999842047691345, "adv_max": 1.4999953508377075, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.16642325873249997, "comp_kd_mean": 0.0, "comp_total_mean": 0.16642325873249997, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 79 |
+
{"rl_step": 390, "loss": -0.0, "mean_return": 0.1715175062417984, "max_return": 0.6860700249671936, "mean_motif_chained": 0.1715175, "gnorm": 1.9185245037078857e-07, "elapsed_s": 5943.860230445862, "cell_type": "OPC", "edit_type": "promoter_retarget", "target_tf": "OLIG2", "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-08, "adv_std": 0.9999970197677612, "adv_min": -0.4999985098838806, "adv_max": 1.4999955892562866, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.1715175, "comp_kd_mean": 0.0, "comp_total_mean": 0.1715175, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 80 |
+
{"rl_step": 395, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 6019.611602544785, "cell_type": "Ast", "edit_type": "promoter_retarget", "target_tf": "NFIA", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 81 |
+
{"rl_step": 400, "loss": -0.0, "mean_return": 0.25, "max_return": 1.0, "mean_motif_chained": 0.25, "gnorm": 1.7881393432617188e-07, "elapsed_s": 6094.953852176666, "cell_type": "In", "edit_type": "cell_type_transfer", "target_tf": "NEUROD2", "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-08, "adv_std": 0.9999979734420776, "adv_min": -0.4999989867210388, "adv_max": 1.4999970197677612, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.25, "comp_kd_mean": 0.0, "comp_total_mean": 0.25, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 82 |
+
{"rl_step": 405, "loss": 1.4901161193847656e-08, "mean_return": 0.3804152309894562, "max_return": 0.7684767842292786, "mean_motif_chained": 0.3804152318802, "gnorm": 7.171183824539185e-08, "elapsed_s": 6175.349889993668, "cell_type": "Mic", "edit_type": "promoter_retarget", "target_tf": "SPI1", "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": 0.0, "adv_std": 0.9999977350234985, "adv_min": -0.8659359812736511, "adv_max": 0.8833412528038025, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.3804152318802, "comp_kd_mean": 0.0, "comp_total_mean": 0.3804152318802, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 83 |
+
{"rl_step": 410, "loss": -0.0, "mean_return": 0.16475902497768402, "max_return": 0.6590360999107361, "mean_motif_chained": 0.16475902614517499, "gnorm": 2.5890767574310303e-07, "elapsed_s": 6251.227410793304, "cell_type": "OPC", "edit_type": "activity_boost", "target_tf": "SOX10", "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": 0.0, "adv_std": 0.9999969005584717, "adv_min": -0.49999845027923584, "adv_max": 1.4999953508377075, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.16475902614517499, "comp_kd_mean": 0.0, "comp_total_mean": 0.16475902614517499, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 84 |
+
{"rl_step": 415, "loss": -0.0, "mean_return": 0.14264850318431854, "max_return": 0.5705940127372742, "mean_motif_chained": 0.142648507485, "gnorm": 7.636845111846924e-08, "elapsed_s": 6326.725033521652, "cell_type": "Ast", "edit_type": "cell_type_transfer", "target_tf": "DLX1", "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": 0.0, "adv_std": 0.9999964237213135, "adv_min": -0.49999821186065674, "adv_max": 1.4999946355819702, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.142648507485, "comp_kd_mean": 0.0, "comp_total_mean": 0.142648507485, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 85 |
+
{"rl_step": 420, "loss": -0.0, "mean_return": 0.1940598040819168, "max_return": 0.7762392163276672, "mean_motif_chained": 0.1940598, "gnorm": 1.5087425708770752e-07, "elapsed_s": 6402.422605276108, "cell_type": "Mic", "edit_type": "promoter_retarget", "target_tf": "CEBPA", "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": 0.0, "adv_std": 0.9999973773956299, "adv_min": -0.49999868869781494, "adv_max": 1.4999960660934448, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.1940598, "comp_kd_mean": 0.0, "comp_total_mean": 0.1940598, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 86 |
+
{"rl_step": 425, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 6479.296871185303, "cell_type": "Ex", "edit_type": "promoter_retarget", "target_tf": "NEUROD2", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 87 |
+
{"rl_step": 430, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 6554.354177713394, "cell_type": "Mic", "edit_type": "activity_boost", "target_tf": "CEBPA", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 88 |
+
{"rl_step": 435, "loss": 2.9802322387695312e-08, "mean_return": 0.4780413508415222, "max_return": 0.9800999760627747, "mean_motif_chained": 0.47804133697674744, "gnorm": 1.2200325727462769e-07, "elapsed_s": 6629.465080976486, "cell_type": "In", "edit_type": "activity_boost", "target_tf": "DLX2", "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": 0.0, "adv_std": 0.9999980926513672, "adv_min": -0.8654778003692627, "adv_max": 0.9089602828025818, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.47804133697674744, "comp_kd_mean": 0.0, "comp_total_mean": 0.47804133697674744, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 89 |
+
{"rl_step": 440, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 6704.835308551788, "cell_type": "Oli", "edit_type": "promoter_retarget", "target_tf": "MYRF", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 90 |
+
{"rl_step": 445, "loss": -0.0, "mean_return": 0.148499995470047, "max_return": 0.593999981880188, "mean_motif_chained": 0.1485, "gnorm": 3.725290298461914e-07, "elapsed_s": 6781.388407945633, "cell_type": "Ast", "edit_type": "cell_type_transfer", "target_tf": "OLIG2", "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-08, "adv_std": 0.999996542930603, "adv_min": -0.4999983012676239, "adv_max": 1.4999948740005493, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.1485, "comp_kd_mean": 0.0, "comp_total_mean": 0.1485, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 91 |
+
{"rl_step": 450, "loss": 2.9802322387695312e-08, "mean_return": 0.23301634192466736, "max_return": 0.9320653676986694, "mean_motif_chained": 0.23301633697674748, "gnorm": 1.341104507446289e-07, "elapsed_s": 6858.368237257004, "cell_type": "In", "edit_type": "cell_type_transfer", "target_tf": "CEBPA", "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": -2.9802322387695312e-08, "adv_std": 0.9999977350234985, "adv_min": -0.49999892711639404, "adv_max": 1.4999966621398926, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.23301633697674748, "comp_kd_mean": 0.0, "comp_total_mean": 0.23301633697674748, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 92 |
+
{"rl_step": 455, "loss": 2.9802322387695312e-08, "mean_return": 0.5549588203430176, "max_return": 1.5198352336883545, "mean_motif_chained": 0.5549588046860485, "gnorm": 1.6391277313232422e-07, "elapsed_s": 6934.257586479187, "cell_type": "Ex", "edit_type": "activity_boost", "target_tf": "NEUROD2", "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": 0.0, "adv_std": 0.9999986290931702, "adv_min": -0.7676270008087158, "adv_max": 1.3346309661865234, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.5549588046860485, "comp_kd_mean": 0.0, "comp_total_mean": 0.5549588046860485, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 93 |
+
{"rl_step": 460, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 7010.646342992783, "cell_type": "OPC", "edit_type": "activity_boost", "target_tf": "SOX6", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 94 |
+
{"rl_step": 465, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 7087.40611743927, "cell_type": "Oli", "edit_type": "activity_boost", "target_tf": "SOX10", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 95 |
+
{"rl_step": 470, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 7163.517128705978, "cell_type": "Ast", "edit_type": "activity_boost", "target_tf": "AP1", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 96 |
+
{"rl_step": 475, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 7239.007977485657, "cell_type": "Ast", "edit_type": "cell_type_transfer", "target_tf": "RORB", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 97 |
+
{"rl_step": 480, "loss": -0.0, "mean_return": 0.1979999989271164, "max_return": 0.7919999957084656, "mean_motif_chained": 0.198, "gnorm": 3.725290298461914e-08, "elapsed_s": 7314.803087949753, "cell_type": "Mic", "edit_type": "activity_boost", "target_tf": "SPI1", "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-08, "adv_std": 0.9999973773956299, "adv_min": -0.49999871850013733, "adv_max": 1.4999960660934448, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.198, "comp_kd_mean": 0.0, "comp_total_mean": 0.198, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 98 |
+
{"rl_step": 485, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 7390.40247297287, "cell_type": "OPC", "edit_type": "activity_boost", "target_tf": "ASCL1", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 99 |
+
{"rl_step": 490, "loss": -0.0, "mean_return": 0.0, "max_return": 0.0, "mean_motif_chained": 0.0, "gnorm": 0.0, "elapsed_s": 7466.453742742538, "cell_type": "Ast", "edit_type": "cell_type_transfer", "target_tf": "RUNX1", "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": 0.0, "adv_std": 0.0, "adv_min": 0.0, "adv_max": 0.0, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.0, "comp_kd_mean": 0.0, "comp_total_mean": 0.0, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 100 |
+
{"rl_step": 495, "loss": -0.0, "mean_return": 0.24014900624752045, "max_return": 0.9605960249900818, "mean_motif_chained": 0.2401490025, "gnorm": 1.4621764421463013e-07, "elapsed_s": 7542.731305837631, "cell_type": "OPC", "edit_type": "activity_boost", "target_tf": "ASCL1", "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-08, "adv_std": 0.9999978542327881, "adv_min": -0.49999895691871643, "adv_max": 1.4999967813491821, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.2401490025, "comp_kd_mean": 0.0, "comp_total_mean": 0.2401490025, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 101 |
+
{"rl_step": 500, "loss": -2.9802322387695312e-08, "mean_return": 0.5804813504219055, "max_return": 2.321925401687622, "mean_motif_chained": 0.58048137675, "gnorm": 3.129243850708008e-07, "elapsed_s": 7618.606285810471, "cell_type": "OPC", "edit_type": "activity_boost", "target_tf": "ASCL1", "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": 0.0, "adv_std": 0.9999992847442627, "adv_min": -0.49999964237213135, "adv_max": 1.499998927116394, "comp_oracle_mean": 0.0, "comp_motif_chained_mean": 0.58048137675, "comp_kd_mean": 0.0, "comp_total_mean": 0.58048137675, "_algo": "grpo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
|