Upload exp_phase8_t3_rl_edit_tight_dapo_s4_lr1e3_moods_20260506_122916/log.jsonl with huggingface_hub
Browse files
exp_phase8_t3_rl_edit_tight_dapo_s4_lr1e3_moods_20260506_122916/log.jsonl
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"rl_step": 1, "loss": 5.960464477539063e-08, "mean_return": 50.02690887451172, "max_return": 76.82461547851562, "mean_motif_chained": 50.02690520882023, "gnorm": 0.10400390625, "elapsed_s": 8.786425590515137, "cell_type": "Mic", "edit_type": "cell_type_transfer", "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": -5.960464477539063e-08, "adv_std": 0.9999999403953552, "adv_min": -1.0743191242218018, "adv_max": 0.8617146015167236, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 50.02690520882023, "comp_kd_mean": 0.0, "comp_total_mean": 50.02690520882023, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 2 |
+
{"rl_step": 5, "loss": -1.1920928955078125e-07, "mean_return": 40.247352600097656, "max_return": 51.92276382446289, "mean_motif_chained": 40.2473539473519, "gnorm": 0.1328125, "elapsed_s": 36.55211544036865, "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": 1.1920928955078125e-07, "adv_std": 1.0, "adv_min": -1.2906948328018188, "adv_max": 1.0564221143722534, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 40.2473539473519, "comp_kd_mean": 0.0, "comp_total_mean": 40.2473539473519, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 3 |
+
{"rl_step": 10, "loss": 1.7136335372924805e-07, "mean_return": 43.454811096191406, "max_return": 49.790374755859375, "mean_motif_chained": 43.45481066938145, "gnorm": 0.109375, "elapsed_s": 71.61390399932861, "cell_type": "Ast", "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": -1.7136335372924805e-07, "adv_std": 0.9999997615814209, "adv_min": -1.299451470375061, "adv_max": 1.1298705339431763, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 43.45481066938145, "comp_kd_mean": 0.0, "comp_total_mean": 43.45481066938145, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 4 |
+
{"rl_step": 15, "loss": -0.0, "mean_return": 58.83467483520508, "max_return": 86.80486297607422, "mean_motif_chained": 58.83467537625867, "gnorm": 0.1171875, "elapsed_s": 106.16786980628967, "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": 0.0, "adv_std": 0.9999998807907104, "adv_min": -1.1879734992980957, "adv_max": 1.127020239830017, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 58.83467537625867, "comp_kd_mean": 0.0, "comp_total_mean": 58.83467537625867, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 5 |
+
{"rl_step": 20, "loss": -1.043081283569336e-07, "mean_return": 47.41889953613281, "max_return": 56.82339859008789, "mean_motif_chained": 47.41890009811997, "gnorm": 0.130859375, "elapsed_s": 141.63120532035828, "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": 1.043081283569336e-07, "adv_std": 0.9999999403953552, "adv_min": -1.1560633182525635, "adv_max": 0.8916300535202026, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 47.41890009811997, "comp_kd_mean": 0.0, "comp_total_mean": 47.41890009811997, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 6 |
+
{"rl_step": 25, "loss": -2.384185791015625e-07, "mean_return": 45.94660949707031, "max_return": 65.74958038330078, "mean_motif_chained": 45.94661270696395, "gnorm": 0.1884765625, "elapsed_s": 178.9767770767212, "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": 2.384185791015625e-07, "adv_std": 0.9999998807907104, "adv_min": -0.89909827709198, "adv_max": 1.2205288410186768, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 45.94661270696395, "comp_kd_mean": 0.0, "comp_total_mean": 45.94661270696395, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 7 |
+
{"rl_step": 30, "loss": 1.7881393432617188e-07, "mean_return": 53.09934997558594, "max_return": 63.67413330078125, "mean_motif_chained": 53.09934839141193, "gnorm": 0.126953125, "elapsed_s": 215.89196848869324, "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": -1.7881393432617188e-07, "adv_std": 0.9999998211860657, "adv_min": -1.289082407951355, "adv_max": 0.9191433787345886, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 53.09934839141193, "comp_kd_mean": 0.0, "comp_total_mean": 53.09934839141193, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 8 |
+
{"rl_step": 35, "loss": -1.564621925354004e-07, "mean_return": 32.20243453979492, "max_return": 48.344966888427734, "mean_motif_chained": 32.2024363628117, "gnorm": 0.1396484375, "elapsed_s": 252.2629623413086, "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": 1.564621925354004e-07, "adv_std": 0.9999998807907104, "adv_min": -0.9920253753662109, "adv_max": 1.368311882019043, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 32.2024363628117, "comp_kd_mean": 0.0, "comp_total_mean": 32.2024363628117, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 9 |
+
{"rl_step": 40, "loss": -5.960464477539063e-08, "mean_return": 42.536903381347656, "max_return": 60.82207489013672, "mean_motif_chained": 42.53690478251444, "gnorm": 0.138671875, "elapsed_s": 289.14926385879517, "cell_type": "Ex", "edit_type": "activity_boost", "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": 5.960464477539063e-08, "adv_std": 0.9999998211860657, "adv_min": -1.0847513675689697, "adv_max": 1.1009299755096436, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 42.53690478251444, "comp_kd_mean": 0.0, "comp_total_mean": 42.53690478251444, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 10 |
+
{"rl_step": 45, "loss": -0.0, "mean_return": 47.407928466796875, "max_return": 71.61424255371094, "mean_motif_chained": 47.40792812897614, "gnorm": 0.173828125, "elapsed_s": 325.25527834892273, "cell_type": "Ast", "edit_type": "promoter_retarget", "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": 0.9999998807907104, "adv_min": -0.9307346343994141, "adv_max": 1.179816722869873, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 47.40792812897614, "comp_kd_mean": 0.0, "comp_total_mean": 47.40792812897614, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 11 |
+
{"rl_step": 50, "loss": -1.2665987014770508e-07, "mean_return": 56.01734161376953, "max_return": 85.93243408203125, "mean_motif_chained": 56.01734552830367, "gnorm": 0.115234375, "elapsed_s": 361.513573884964, "cell_type": "OPC", "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": 1.2665987014770508e-07, "adv_std": 0.9999998211860657, "adv_min": -1.1256529092788696, "adv_max": 1.230729103088379, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 56.01734552830367, "comp_kd_mean": 0.0, "comp_total_mean": 56.01734552830367, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 12 |
+
{"rl_step": 55, "loss": 1.564621925354004e-07, "mean_return": 26.793956756591797, "max_return": 34.75249099731445, "mean_motif_chained": 26.79395583269547, "gnorm": 0.12890625, "elapsed_s": 397.8534789085388, "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": -1.564621925354004e-07, "adv_std": 0.9999998807907104, "adv_min": -1.2664308547973633, "adv_max": 0.8082972764968872, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 26.79395583269547, "comp_kd_mean": 0.0, "comp_total_mean": 26.79395583269547, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 13 |
+
{"rl_step": 60, "loss": 2.9802322387695312e-08, "mean_return": 34.26800537109375, "max_return": 53.078704833984375, "mean_motif_chained": 34.268005055478675, "gnorm": 0.1787109375, "elapsed_s": 434.18750262260437, "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": -2.9802322387695312e-08, "adv_std": 1.0, "adv_min": -0.9452075958251953, "adv_max": 1.3913236856460571, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 34.268005055478675, "comp_kd_mean": 0.0, "comp_total_mean": 34.268005055478675, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 14 |
+
{"rl_step": 65, "loss": -1.1175870895385742e-08, "mean_return": 40.48509216308594, "max_return": 56.4966926574707, "mean_motif_chained": 40.485092513539485, "gnorm": 0.146484375, "elapsed_s": 470.38051080703735, "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": 1.1175870895385742e-08, "adv_std": 0.9999998807907104, "adv_min": -1.1211450099945068, "adv_max": 1.2630879878997803, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 40.485092513539485, "comp_kd_mean": 0.0, "comp_total_mean": 40.485092513539485, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 15 |
+
{"rl_step": 70, "loss": -4.470348358154297e-08, "mean_return": 51.1451416015625, "max_return": 109.1636734008789, "mean_motif_chained": 51.145143507131706, "gnorm": 0.26171875, "elapsed_s": 506.4767646789551, "cell_type": "Ex", "edit_type": "activity_boost", "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": 4.470348358154297e-08, "adv_std": 1.0, "adv_min": -0.8455163240432739, "adv_max": 1.4223424196243286, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 51.145143507131706, "comp_kd_mean": 0.0, "comp_total_mean": 51.145143507131706, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 16 |
+
{"rl_step": 75, "loss": -2.980232238769531e-07, "mean_return": 42.59600067138672, "max_return": 46.19077682495117, "mean_motif_chained": 42.596001954583926, "gnorm": 0.1669921875, "elapsed_s": 542.5313596725464, "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": 2.980232238769531e-07, "adv_std": 0.9999998211860657, "adv_min": -1.4965629577636719, "adv_max": 0.5522605776786804, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 42.596001954583926, "comp_kd_mean": 0.0, "comp_total_mean": 42.596001954583926, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 17 |
+
{"rl_step": 80, "loss": -4.470348358154297e-08, "mean_return": 39.000457763671875, "max_return": 54.412967681884766, "mean_motif_chained": 39.00045723800693, "gnorm": 0.1728515625, "elapsed_s": 578.5912706851959, "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": 4.470348358154297e-08, "adv_std": 1.0, "adv_min": -1.3666932582855225, "adv_max": 0.8912321925163269, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 39.00045723800693, "comp_kd_mean": 0.0, "comp_total_mean": 39.00045723800693, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 18 |
+
{"rl_step": 85, "loss": -1.7881393432617188e-07, "mean_return": 47.41844177246094, "max_return": 68.37345886230469, "mean_motif_chained": 47.418444542055674, "gnorm": 0.23828125, "elapsed_s": 614.5800492763519, "cell_type": "Mic", "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": 1.7881393432617188e-07, "adv_std": 0.9999998807907104, "adv_min": -0.8085426092147827, "adv_max": 1.4534190893173218, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 47.418444542055674, "comp_kd_mean": 0.0, "comp_total_mean": 47.418444542055674, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 19 |
+
{"rl_step": 90, "loss": 2.5331974029541016e-07, "mean_return": 32.73162841796875, "max_return": 40.50032424926758, "mean_motif_chained": 32.73162591744348, "gnorm": 0.24609375, "elapsed_s": 648.9953157901764, "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": -2.5331974029541016e-07, "adv_std": 0.9999998211860657, "adv_min": -0.9148123264312744, "adv_max": 0.9637801051139832, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 32.73162591744348, "comp_kd_mean": 0.0, "comp_total_mean": 32.73162591744348, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 20 |
+
{"rl_step": 95, "loss": 2.682209014892578e-07, "mean_return": 40.9913444519043, "max_return": 43.29156494140625, "mean_motif_chained": 40.99134313270471, "gnorm": 0.265625, "elapsed_s": 683.747453212738, "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": -2.682209014892578e-07, "adv_std": 0.9999995231628418, "adv_min": -1.4854856729507446, "adv_max": 0.6584973335266113, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 40.99134313270471, "comp_kd_mean": 0.0, "comp_total_mean": 40.99134313270471, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 21 |
+
{"rl_step": 100, "loss": -0.0, "mean_return": 31.89751434326172, "max_return": 44.17169189453125, "mean_motif_chained": 31.897514009406457, "gnorm": 0.294921875, "elapsed_s": 720.5810406208038, "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": 0.9999999403953552, "adv_min": -1.111053705215454, "adv_max": 1.28877854347229, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 31.897514009406457, "comp_kd_mean": 0.0, "comp_total_mean": 31.897514009406457, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 22 |
+
{"rl_step": 105, "loss": -4.470348358154297e-08, "mean_return": 35.681861877441406, "max_return": 49.16248321533203, "mean_motif_chained": 35.681861937490226, "gnorm": 0.392578125, "elapsed_s": 758.5110919475555, "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": 4.470348358154297e-08, "adv_std": 0.9999998807907104, "adv_min": -1.001353144645691, "adv_max": 0.958949863910675, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 35.681861937490226, "comp_kd_mean": 0.0, "comp_total_mean": 35.681861937490226, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|
| 23 |
+
{"rl_step": 110, "loss": -1.1920928955078125e-07, "mean_return": 49.704811096191406, "max_return": 59.88700485229492, "mean_motif_chained": 49.70481349978243, "gnorm": 0.27734375, "elapsed_s": 796.2095334529877, "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": 1.1920928955078125e-07, "adv_std": 0.9999997615814209, "adv_min": -1.4950388669967651, "adv_max": 0.5588927865028381, "comp_oracle_mean": 0.0, "comp_moods_chained_mean": 49.70481349978243, "comp_kd_mean": 0.0, "comp_total_mean": 49.70481349978243, "_algo": "dapo", "_lora_r": 16, "_peak_lr": 0.001, "_temperature": 1.0, "_seq_len_mean": 360.0}
|