explcre commited on
Commit
41aec0e
·
verified ·
1 Parent(s): 7f5d841

Upload exp_phase8_t3_rl_edit_tight_gspo_FIXED_s0_20260506_114317/log.jsonl with huggingface_hub

Browse files
exp_phase8_t3_rl_edit_tight_gspo_FIXED_s0_20260506_114317/log.jsonl ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"rl_step": 1, "loss": 2.9802322387695312e-08, "mean_return": 46.05831527709961, "max_return": 61.47843551635742, "mean_motif_chained": 46.05831472009569, "gnorm": 0.000270843505859375, "elapsed_s": 7.612231731414795, "cell_type": "OPC", "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": -2.9802322387695312e-08, "adv_std": 0.9999999403953552, "adv_min": -1.0226925611495972, "adv_max": 1.309010624885559, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 46.05831472009569, "comp_kd_mean": 0.0, "comp_total_mean": 46.05831472009569, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
2
+ {"rl_step": 5, "loss": 1.1920928955078125e-07, "mean_return": 49.38098907470703, "max_return": 79.05812072753906, "mean_motif_chained": 49.380986575544185, "gnorm": 0.0004425048828125, "elapsed_s": 34.33932590484619, "cell_type": "In", "edit_type": "activity_boost", "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": -1.1920928955078125e-07, "adv_std": 1.0, "adv_min": -0.9496251940727234, "adv_max": 1.4010095596313477, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 49.380986575544185, "comp_kd_mean": 0.0, "comp_total_mean": 49.380986575544185, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
3
+ {"rl_step": 10, "loss": 0.0, "mean_return": 45.23150634765625, "max_return": 79.88200378417969, "mean_motif_chained": 45.23150507799287, "gnorm": 0.00021648406982421875, "elapsed_s": 67.54396653175354, "cell_type": "Oli", "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": 0.0, "adv_std": 0.9999999403953552, "adv_min": -0.9152618050575256, "adv_max": 1.3180553913116455, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 45.23150507799287, "comp_kd_mean": 0.0, "comp_total_mean": 45.23150507799287, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
4
+ {"rl_step": 15, "loss": 4.470348358154297e-08, "mean_return": 35.48145294189453, "max_return": 54.20783233642578, "mean_motif_chained": 35.4814509588327, "gnorm": 0.00038909912109375, "elapsed_s": 100.64046335220337, "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": -4.470348358154297e-08, "adv_std": 1.0, "adv_min": -1.1345272064208984, "adv_max": 1.2619869709014893, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 35.4814509588327, "comp_kd_mean": 0.0, "comp_total_mean": 35.4814509588327, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
5
+ {"rl_step": 20, "loss": -2.5331974029541016e-07, "mean_return": 40.17699432373047, "max_return": 54.9540901184082, "mean_motif_chained": 40.17699776397244, "gnorm": 0.000347137451171875, "elapsed_s": 134.18701887130737, "cell_type": "Oli", "edit_type": "activity_boost", "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": 2.5331974029541016e-07, "adv_std": 0.9999999403953552, "adv_min": -1.0743645429611206, "adv_max": 1.238998293876648, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 40.17699776397244, "comp_kd_mean": 0.0, "comp_total_mean": 40.17699776397244, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
6
+ {"rl_step": 25, "loss": 7.450580596923828e-08, "mean_return": 43.553035736083984, "max_return": 71.48587799072266, "mean_motif_chained": 43.55303436684169, "gnorm": 0.000324249267578125, "elapsed_s": 167.24597454071045, "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": -7.450580596923828e-08, "adv_std": 0.9999998807907104, "adv_min": -0.9800338745117188, "adv_max": 1.3968472480773926, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 43.55303436684169, "comp_kd_mean": 0.0, "comp_total_mean": 43.55303436684169, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
7
+ {"rl_step": 30, "loss": 8.940696716308594e-08, "mean_return": 43.03346252441406, "max_return": 88.39996337890625, "mean_motif_chained": 43.03345952009498, "gnorm": 0.0002574920654296875, "elapsed_s": 200.2875940799713, "cell_type": "End", "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": -8.940696716308594e-08, "adv_std": 1.0, "adv_min": -0.6962877511978149, "adv_max": 1.4648243188858032, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 43.03345952009498, "comp_kd_mean": 0.0, "comp_total_mean": 43.03345952009498, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
8
+ {"rl_step": 35, "loss": -2.9802322387695312e-08, "mean_return": 62.737342834472656, "max_return": 90.47138214111328, "mean_motif_chained": 62.73734434272171, "gnorm": 0.0003070831298828125, "elapsed_s": 233.31693840026855, "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": 0.9999999403953552, "adv_min": -1.0740357637405396, "adv_max": 1.2382938861846924, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 62.73734434272171, "comp_kd_mean": 0.0, "comp_total_mean": 62.73734434272171, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
9
+ {"rl_step": 40, "loss": -2.9802322387695312e-08, "mean_return": 45.467376708984375, "max_return": 74.59354400634766, "mean_motif_chained": 45.46737725721948, "gnorm": 0.0002880096435546875, "elapsed_s": 266.52743339538574, "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": 2.9802322387695312e-08, "adv_std": 0.9999998807907104, "adv_min": -0.8846098184585571, "adv_max": 1.1127917766571045, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 45.46737725721948, "comp_kd_mean": 0.0, "comp_total_mean": 45.46737725721948, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
10
+ {"rl_step": 45, "loss": 5.960464477539063e-08, "mean_return": 36.94158935546875, "max_return": 53.452266693115234, "mean_motif_chained": 36.941588838085714, "gnorm": 0.00020885467529296875, "elapsed_s": 299.59948468208313, "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": -5.960464477539063e-08, "adv_std": 0.9999998807907104, "adv_min": -1.0947917699813843, "adv_max": 1.3299764394760132, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 36.941588838085714, "comp_kd_mean": 0.0, "comp_total_mean": 36.941588838085714, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
11
+ {"rl_step": 50, "loss": 5.960464477539063e-08, "mean_return": 40.89837646484375, "max_return": 57.640052795410156, "mean_motif_chained": 40.898376135603726, "gnorm": 0.0003376007080078125, "elapsed_s": 332.66070914268494, "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": -5.960464477539063e-08, "adv_std": 0.9999998807907104, "adv_min": -0.86866694688797, "adv_max": 1.036812424659729, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 40.898376135603726, "comp_kd_mean": 0.0, "comp_total_mean": 40.898376135603726, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
12
+ {"rl_step": 55, "loss": -1.341104507446289e-07, "mean_return": 45.967132568359375, "max_return": 75.60249328613281, "mean_motif_chained": 45.96713527481193, "gnorm": 0.0002765655517578125, "elapsed_s": 365.9627003669739, "cell_type": "Ex", "edit_type": "promoter_retarget", "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": 1.341104507446289e-07, "adv_std": 0.9999998807907104, "adv_min": -1.118549108505249, "adv_max": 1.2830859422683716, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 45.96713527481193, "comp_kd_mean": 0.0, "comp_total_mean": 45.96713527481193, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
13
+ {"rl_step": 60, "loss": 0.0, "mean_return": 30.86285400390625, "max_return": 41.95278549194336, "mean_motif_chained": 30.86285444221948, "gnorm": 0.0002536773681640625, "elapsed_s": 399.08935141563416, "cell_type": "Ast", "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": 0.0, "adv_std": 1.0, "adv_min": -1.1691688299179077, "adv_max": 1.130358338356018, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 30.86285444221948, "comp_kd_mean": 0.0, "comp_total_mean": 30.86285444221948, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
14
+ {"rl_step": 65, "loss": 2.9802322387695312e-08, "mean_return": 31.488723754882812, "max_return": 43.63351821899414, "mean_motif_chained": 31.488723737044232, "gnorm": 0.0002593994140625, "elapsed_s": 432.1833279132843, "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": -2.9802322387695312e-08, "adv_std": 0.9999998807907104, "adv_min": -1.0423274040222168, "adv_max": 0.9832457304000854, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 31.488723737044232, "comp_kd_mean": 0.0, "comp_total_mean": 31.488723737044232, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
15
+ {"rl_step": 70, "loss": -7.450580596923828e-09, "mean_return": 74.79894256591797, "max_return": 106.52816009521484, "mean_motif_chained": 74.7989420546932, "gnorm": 0.0003566741943359375, "elapsed_s": 465.3101511001587, "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": 7.450580596923828e-09, "adv_std": 0.9999998807907104, "adv_min": -1.1543835401535034, "adv_max": 1.1515411138534546, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 74.7989420546932, "comp_kd_mean": 0.0, "comp_total_mean": 74.7989420546932, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
16
+ {"rl_step": 75, "loss": 0.0, "mean_return": 75.19883728027344, "max_return": 107.64218139648438, "mean_motif_chained": 75.19883871572893, "gnorm": 0.000431060791015625, "elapsed_s": 498.5904290676117, "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": 0.0, "adv_std": 1.0000001192092896, "adv_min": -1.0337272882461548, "adv_max": 0.9634473919868469, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 75.19883871572893, "comp_kd_mean": 0.0, "comp_total_mean": 75.19883871572893, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
17
+ {"rl_step": 80, "loss": 5.960464477539063e-08, "mean_return": 50.283267974853516, "max_return": 70.21163940429688, "mean_motif_chained": 50.28326644337571, "gnorm": 0.0002803802490234375, "elapsed_s": 531.6720209121704, "cell_type": "Ex", "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": -5.960464477539063e-08, "adv_std": 0.9999998807907104, "adv_min": -0.8893682360649109, "adv_max": 1.1586050987243652, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 50.28326644337571, "comp_kd_mean": 0.0, "comp_total_mean": 50.28326644337571, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
18
+ {"rl_step": 85, "loss": 4.470348358154297e-08, "mean_return": 49.705848693847656, "max_return": 74.775634765625, "mean_motif_chained": 49.705847856888425, "gnorm": 0.0003681182861328125, "elapsed_s": 564.7624025344849, "cell_type": "In", "edit_type": "activity_boost", "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": -4.470348358154297e-08, "adv_std": 0.9999998211860657, "adv_min": -0.8308867812156677, "adv_max": 1.4051159620285034, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 49.705847856888425, "comp_kd_mean": 0.0, "comp_total_mean": 49.705847856888425, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
19
+ {"rl_step": 90, "loss": 1.043081283569336e-07, "mean_return": 65.55977630615234, "max_return": 98.35684967041016, "mean_motif_chained": 65.55977423349316, "gnorm": 0.000286102294921875, "elapsed_s": 598.0960440635681, "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": -1.043081283569336e-07, "adv_std": 0.9999999403953552, "adv_min": -0.8688801527023315, "adv_max": 1.3799492120742798, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 65.55977423349316, "comp_kd_mean": 0.0, "comp_total_mean": 65.55977423349316, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
20
+ {"rl_step": 95, "loss": 5.960464477539063e-08, "mean_return": 37.43309783935547, "max_return": 46.46335220336914, "mean_motif_chained": 37.43309795813563, "gnorm": 0.0002765655517578125, "elapsed_s": 631.2426946163177, "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": -5.960464477539063e-08, "adv_std": 0.9999998211860657, "adv_min": -1.3588361740112305, "adv_max": 0.8691165447235107, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 37.43309795813563, "comp_kd_mean": 0.0, "comp_total_mean": 37.43309795813563, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
21
+ {"rl_step": 100, "loss": -5.960464477539063e-08, "mean_return": 64.0976791381836, "max_return": 98.57039642333984, "mean_motif_chained": 64.09768045682198, "gnorm": 0.000457763671875, "elapsed_s": 664.3271820545197, "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": 5.960464477539063e-08, "adv_std": 0.9999998807907104, "adv_min": -0.7214392423629761, "adv_max": 1.4258946180343628, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 64.09768045682198, "comp_kd_mean": 0.0, "comp_total_mean": 64.09768045682198, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
22
+ {"rl_step": 105, "loss": 5.960464477539063e-08, "mean_return": 62.913787841796875, "max_return": 78.9608154296875, "mean_motif_chained": 62.913787743678434, "gnorm": 0.000461578369140625, "elapsed_s": 697.3942954540253, "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": -5.960464477539063e-08, "adv_std": 1.0, "adv_min": -1.2538635730743408, "adv_max": 1.193302869796753, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 62.913787743678434, "comp_kd_mean": 0.0, "comp_total_mean": 62.913787743678434, "_algo": "gspo", "_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": 36.490840911865234, "max_return": 49.49461364746094, "mean_motif_chained": 36.4908410577882, "gnorm": 0.0003147125244140625, "elapsed_s": 730.6703279018402, "cell_type": "Ast", "edit_type": "activity_boost", "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": -1.4901161193847656e-08, "adv_std": 0.9999998807907104, "adv_min": -1.3688504695892334, "adv_max": 0.747717022895813, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 36.4908410577882, "comp_kd_mean": 0.0, "comp_total_mean": 36.4908410577882, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
24
+ {"rl_step": 115, "loss": 2.086162567138672e-07, "mean_return": 40.533172607421875, "max_return": 52.44001007080078, "mean_motif_chained": 40.53316947663467, "gnorm": 0.0003490447998046875, "elapsed_s": 763.781946182251, "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": -2.086162567138672e-07, "adv_std": 1.0, "adv_min": -1.3934714794158936, "adv_max": 0.9768416881561279, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 40.53316947663467, "comp_kd_mean": 0.0, "comp_total_mean": 40.53316947663467, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
25
+ {"rl_step": 120, "loss": 0.0, "mean_return": 28.110126495361328, "max_return": 42.618263244628906, "mean_motif_chained": 28.11012682756348, "gnorm": 0.0002651214599609375, "elapsed_s": 796.9237365722656, "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": 0.0, "adv_std": 1.0, "adv_min": -0.981418251991272, "adv_max": 1.3826991319656372, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 28.11012682756348, "comp_kd_mean": 0.0, "comp_total_mean": 28.11012682756348, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
26
+ {"rl_step": 125, "loss": -3.5762786865234375e-07, "mean_return": 26.69049835205078, "max_return": 30.937421798706055, "mean_motif_chained": 26.69050023047971, "gnorm": 0.0002574920654296875, "elapsed_s": 830.0136306285858, "cell_type": "Oli", "edit_type": "activity_boost", "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": 3.5762786865234375e-07, "adv_std": 0.9999996423721313, "adv_min": -1.3249356746673584, "adv_max": 1.1046178340911865, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 26.69050023047971, "comp_kd_mean": 0.0, "comp_total_mean": 26.69050023047971, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
27
+ {"rl_step": 130, "loss": -7.450580596923828e-08, "mean_return": 50.598514556884766, "max_return": 64.19975280761719, "mean_motif_chained": 50.59851582299565, "gnorm": 0.0003070831298828125, "elapsed_s": 863.355721950531, "cell_type": "Oli", "edit_type": "promoter_retarget", "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": 7.450580596923828e-08, "adv_std": 0.9999998807907104, "adv_min": -0.9459017515182495, "adv_max": 1.1001557111740112, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 50.59851582299565, "comp_kd_mean": 0.0, "comp_total_mean": 50.59851582299565, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
28
+ {"rl_step": 135, "loss": 0.0, "mean_return": 43.215030670166016, "max_return": 59.27657699584961, "mean_motif_chained": 43.21503041618449, "gnorm": 0.0002155303955078125, "elapsed_s": 896.5034058094025, "cell_type": "Oli", "edit_type": "cell_type_transfer", "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": 0.9999998807907104, "adv_min": -0.7282148003578186, "adv_max": 1.4593257904052734, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 43.21503041618449, "comp_kd_mean": 0.0, "comp_total_mean": 43.21503041618449, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
29
+ {"rl_step": 140, "loss": 1.341104507446289e-07, "mean_return": 67.11669921875, "max_return": 78.3178939819336, "mean_motif_chained": 67.11669586725364, "gnorm": 0.0003147125244140625, "elapsed_s": 929.5695998668671, "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": -1.341104507446289e-07, "adv_std": 1.0, "adv_min": -1.4276355504989624, "adv_max": 0.7563633918762207, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 67.11669586725364, "comp_kd_mean": 0.0, "comp_total_mean": 67.11669586725364, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
30
+ {"rl_step": 145, "loss": 1.7881393432617188e-07, "mean_return": 50.1398811340332, "max_return": 66.15632629394531, "mean_motif_chained": 50.139879092046435, "gnorm": 0.000301361083984375, "elapsed_s": 962.9218285083771, "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": -1.7881393432617188e-07, "adv_std": 0.9999997615814209, "adv_min": -1.0771268606185913, "adv_max": 0.9629108905792236, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 50.139879092046435, "comp_kd_mean": 0.0, "comp_total_mean": 50.139879092046435, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
31
+ {"rl_step": 150, "loss": 4.470348358154297e-08, "mean_return": 57.08048629760742, "max_return": 73.18387603759766, "mean_motif_chained": 57.08048465409562, "gnorm": 0.0003376007080078125, "elapsed_s": 996.0268402099609, "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": -4.470348358154297e-08, "adv_std": 0.9999997615814209, "adv_min": -1.247551679611206, "adv_max": 0.9798166751861572, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 57.08048465409562, "comp_kd_mean": 0.0, "comp_total_mean": 57.08048465409562, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
32
+ {"rl_step": 155, "loss": 4.470348358154297e-08, "mean_return": 45.959468841552734, "max_return": 79.24397277832031, "mean_motif_chained": 45.959467995158924, "gnorm": 0.0002918243408203125, "elapsed_s": 1029.1278283596039, "cell_type": "In", "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": -4.470348358154297e-08, "adv_std": 0.9999998807907104, "adv_min": -0.8584225177764893, "adv_max": 1.2074321508407593, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 45.959467995158924, "comp_kd_mean": 0.0, "comp_total_mean": 45.959467995158924, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
33
+ {"rl_step": 160, "loss": 0.0, "mean_return": 38.612701416015625, "max_return": 57.64653778076172, "mean_motif_chained": 38.61270053911541, "gnorm": 0.0003528594970703125, "elapsed_s": 1062.1697177886963, "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": 0.0, "adv_std": 1.0, "adv_min": -0.6599134206771851, "adv_max": 1.4768155813217163, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 38.61270053911541, "comp_kd_mean": 0.0, "comp_total_mean": 38.61270053911541, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
34
+ {"rl_step": 165, "loss": 1.043081283569336e-07, "mean_return": 33.238067626953125, "max_return": 44.67509078979492, "mean_motif_chained": 33.23806676389123, "gnorm": 0.000263214111328125, "elapsed_s": 1095.536963224411, "cell_type": "Oli", "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": -1.043081283569336e-07, "adv_std": 0.9999997615814209, "adv_min": -1.1049015522003174, "adv_max": 1.1671422719955444, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 33.23806676389123, "comp_kd_mean": 0.0, "comp_total_mean": 33.23806676389123, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
35
+ {"rl_step": 170, "loss": -2.2351741790771484e-08, "mean_return": 61.33012390136719, "max_return": 69.47679138183594, "mean_motif_chained": 61.3301250112327, "gnorm": 0.000293731689453125, "elapsed_s": 1128.6850292682648, "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": 2.2351741790771484e-08, "adv_std": 0.9999997615814209, "adv_min": -0.9128377437591553, "adv_max": 1.202574610710144, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 61.3301250112327, "comp_kd_mean": 0.0, "comp_total_mean": 61.3301250112327, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
36
+ {"rl_step": 175, "loss": 1.043081283569336e-07, "mean_return": 56.869537353515625, "max_return": 78.86918640136719, "mean_motif_chained": 56.86953467729819, "gnorm": 0.0003681182861328125, "elapsed_s": 1161.8454191684723, "cell_type": "Oli", "edit_type": "promoter_retarget", "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": -1.043081283569336e-07, "adv_std": 0.9999999403953552, "adv_min": -1.105631947517395, "adv_max": 0.9137372374534607, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 56.86953467729819, "comp_kd_mean": 0.0, "comp_total_mean": 56.86953467729819, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
37
+ {"rl_step": 180, "loss": -1.1920928955078125e-07, "mean_return": 33.26019287109375, "max_return": 56.67437744140625, "mean_motif_chained": 33.26019475858874, "gnorm": 0.0002651214599609375, "elapsed_s": 1194.9570240974426, "cell_type": "Oli", "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": 1.1920928955078125e-07, "adv_std": 0.9999998807907104, "adv_min": -0.695209801197052, "adv_max": 1.4820194244384766, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 33.26019475858874, "comp_kd_mean": 0.0, "comp_total_mean": 33.26019475858874, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
38
+ {"rl_step": 185, "loss": 7.450580596923828e-09, "mean_return": 44.67304611206055, "max_return": 52.31227111816406, "mean_motif_chained": 44.67304630411459, "gnorm": 0.0002994537353515625, "elapsed_s": 1228.2569291591644, "cell_type": "Ex", "edit_type": "promoter_retarget", "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": -7.450580596923828e-09, "adv_std": 0.9999998807907104, "adv_min": -1.3652687072753906, "adv_max": 0.97890305519104, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 44.67304630411459, "comp_kd_mean": 0.0, "comp_total_mean": 44.67304630411459, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
39
+ {"rl_step": 190, "loss": 5.960464477539063e-08, "mean_return": 47.83491897583008, "max_return": 64.9449234008789, "mean_motif_chained": 47.834917778303975, "gnorm": 0.000354766845703125, "elapsed_s": 1261.3801271915436, "cell_type": "Oli", "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": -5.960464477539063e-08, "adv_std": 1.0, "adv_min": -1.167099952697754, "adv_max": 0.9828416705131531, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 47.834917778303975, "comp_kd_mean": 0.0, "comp_total_mean": 47.834917778303975, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
40
+ {"rl_step": 195, "loss": -1.862645149230957e-08, "mean_return": 56.89095687866211, "max_return": 76.45529174804688, "mean_motif_chained": 56.89095645510815, "gnorm": 0.0003452301025390625, "elapsed_s": 1294.4793348312378, "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": 1.862645149230957e-08, "adv_std": 0.9999998807907104, "adv_min": -1.3022722005844116, "adv_max": 1.112479329109192, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 56.89095645510815, "comp_kd_mean": 0.0, "comp_total_mean": 56.89095645510815, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}
41
+ {"rl_step": 200, "loss": 5.960464477539063e-08, "mean_return": 37.46688461303711, "max_return": 51.58934020996094, "mean_motif_chained": 37.4668838110862, "gnorm": 0.000263214111328125, "elapsed_s": 1327.5844311714172, "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": -5.960464477539063e-08, "adv_std": 0.9999999403953552, "adv_min": -1.2652148008346558, "adv_max": 1.1644586324691772, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 37.4668838110862, "comp_kd_mean": 0.0, "comp_total_mean": 37.4668838110862, "_algo": "gspo", "_lora_r": 16, "_peak_lr": 1e-05, "_temperature": 1.0, "_seq_len_mean": 360.0}