Upload exp_phase8_svgspo_1000step_seed2_20260503_140132/log.jsonl with huggingface_hub
Browse files
exp_phase8_svgspo_1000step_seed2_20260503_140132/log.jsonl
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"step": 1, "elapsed_s": 59.73726558685303, "wallclock_s_per_step": 59.73726677894592, "loss/total": -2.9802322387695312e-08, "loss/per_rollout_mean": -4.516914486885071e-08, "loss/per_rollout_std": 0.5091692851076516, "reward/total_mean": -0.0940097358024215, "reward/total_std": 0.044567317263900154, "reward/total_min": -0.1469116449356079, "reward/total_max": 0.0008391708205044701, "reward/argmax_match": 0.0, "reward/oracle_target_sigmoid": 1.0709885596603273e-11, "reward/oracle_offtarget_sigmoid": 0.5841991193592548, "reward/motif_hit_frac_topK": 0.203125, "reward/r_argmax_term": 0.0, "reward/r_target_term": 5.3549427983016364e-12, "reward/r_motif_term": 0.08125, "reward/r_offtarget_term": -0.17525973580777643, "advantage/mean": 8.288770914077759e-08, "advantage/std": 0.9999667265554844, "advantage/abs_mean": 0.8389047151431441, "ratio/mean": 1.0, "ratio/std": 0.0, "ratio/clip_frac_low": 0.0, "ratio/clip_frac_high": 0.0, "kl/mean": 0.0, "kl/abs_mean": 0.0, "reward/cell_Ex": -0.0940097358024215}
|
| 2 |
+
{"step": 2, "elapsed_s": 115.74804639816284, "wallclock_s_per_step": 57.8740234375, "loss/total": 0.00046878308057785034, "loss/per_rollout_mean": 0.0004687793552875519, "loss/per_rollout_std": 0.509553912364345, "reward/total_mean": -0.06441733131972568, "reward/total_std": 0.044189754466219644, "reward/total_min": -0.11153282523155211, "reward/total_max": 0.0010054736630991101, "reward/argmax_match": 0.0, "reward/oracle_target_sigmoid": 0.00048566613634193345, "reward/oracle_offtarget_sigmoid": 0.3822005479596555, "reward/motif_hit_frac_topK": 0.125, "reward/r_argmax_term": 0.0, "reward/r_target_term": 0.00024283306817096673, "reward/r_motif_term": 0.05, "reward/r_offtarget_term": -0.11466016438789665, "advantage/mean": -3.0547380447387695e-07, "advantage/std": 0.9998878527083762, "advantage/abs_mean": 0.810174860060215, "ratio/mean": 1.0010185269638896, "ratio/std": 0.0018707989425861013, "ratio/clip_frac_low": 0.0, "ratio/clip_frac_high": 0.0, "kl/mean": 0.0009906589984893799, "kl/abs_mean": 0.0015766024589538574, "reward/cell_Ex": -0.06441733131972568}
|
| 3 |
+
{"step": 3, "elapsed_s": 172.0927517414093, "wallclock_s_per_step": 57.36425073941549, "loss/total": 0.0005223900079727173, "loss/per_rollout_mean": 0.0005223918706178665, "loss/per_rollout_std": 0.5094363605666461, "reward/total_mean": -0.056820029629591304, "reward/total_std": 0.07915324952467193, "reward/total_min": -0.1469557762145996, "reward/total_max": 0.049973036209394195, "reward/argmax_match": 0.0, "reward/oracle_target_sigmoid": 0.0053566763048138455, "reward/oracle_offtarget_sigmoid": 0.42749455927332747, "reward/motif_hit_frac_topK": 0.171875, "reward/r_argmax_term": 0.0, "reward/r_target_term": 0.0026783381524069227, "reward/r_motif_term": 0.06875, "reward/r_offtarget_term": -0.12824836778199825, "advantage/mean": -3.5390257835388184e-08, "advantage/std": 0.9999540093586712, "advantage/abs_mean": 0.8641991559416056, "ratio/mean": 1.000495556741953, "ratio/std": 0.0030253137814864, "ratio/clip_frac_low": 0.0, "ratio/clip_frac_high": 0.0, "kl/mean": 0.00029115378856658936, "kl/abs_mean": 0.0018012970685958862, "reward/cell_Ex": -0.056820029629591304}
|
| 4 |
+
{"step": 4, "elapsed_s": 228.2744495868683, "wallclock_s_per_step": 57.06861263513565, "loss/total": 0.00013663992285728455, "loss/per_rollout_mean": 0.00013663806021213531, "loss/per_rollout_std": 0.5104035088168993, "reward/total_mean": 0.3537203232560627, "reward/total_std": 0.524643421170057, "reward/total_min": -0.04420331923612611, "reward/total_max": 1.454618875682354, "reward/argmax_match": 0.25, "reward/oracle_target_sigmoid": 0.21857460251627814, "reward/oracle_offtarget_sigmoid": 0.1643899266735881, "reward/motif_hit_frac_topK": 0.109375, "reward/r_argmax_term": 0.25, "reward/r_target_term": 0.10928730125813907, "reward/r_motif_term": 0.043750000000000004, "reward/r_offtarget_term": -0.04931697800207642, "advantage/mean": 7.450580596923828e-09, "advantage/std": 0.9999904758784433, "advantage/abs_mean": 0.894651398062706, "ratio/mean": 1.0008159847930074, "ratio/std": 0.003089989635111231, "ratio/clip_frac_low": 0.0, "ratio/clip_frac_high": 0.0, "kl/mean": 0.0006699115037918091, "kl/abs_mean": 0.0027834326028823853, "reward/cell_Ex": 0.3537203232560627}
|
| 5 |
+
{"step": 5, "elapsed_s": 284.47549867630005, "wallclock_s_per_step": 56.89509997367859, "loss/total": -0.00024527404457330704, "loss/per_rollout_mean": -0.00024527357891201973, "loss/per_rollout_std": 0.5082332197135548, "reward/total_mean": -0.0637785721952229, "reward/total_std": 0.10222757539406774, "reward/total_min": -0.19770613312721252, "reward/total_max": 0.14473539065809293, "reward/argmax_match": 0.0, "reward/oracle_target_sigmoid": 1.5631951124639443e-06, "reward/oracle_offtarget_sigmoid": 0.4834311793092638, "reward/motif_hit_frac_topK": 0.203125, "reward/r_argmax_term": 0.0, "reward/r_target_term": 7.815975562319721e-07, "reward/r_motif_term": 0.08125000000000002, "reward/r_offtarget_term": -0.14502935379277915, "advantage/mean": -1.5133991837501526e-07, "advantage/std": 0.9999799146036384, "advantage/abs_mean": 0.7797572710551322, "ratio/mean": 1.0002352041192353, "ratio/std": 0.0036196213415032076, "ratio/clip_frac_low": 0.0, "ratio/clip_frac_high": 0.0, "kl/mean": 7.283687591552734e-05, "kl/abs_mean": 0.002692699432373047, "reward/cell_Ex": -0.0637785721952229}
|
| 6 |
+
{"step": 6, "elapsed_s": 340.83009481430054, "wallclock_s_per_step": 56.80501592159271, "loss/total": -0.000563792884349823, "loss/per_rollout_mean": -0.0005637966096401215, "loss/per_rollout_std": 0.510398450550787, "reward/total_mean": -0.003674827018096488, "reward/total_std": 0.020702755587747017, "reward/total_min": -0.04526472091674805, "reward/total_max": 0.017560929225783187, "reward/argmax_match": 0.0, "reward/oracle_target_sigmoid": 0.000707653005406428, "reward/oracle_offtarget_sigmoid": 0.24259551173599903, "reward/motif_hit_frac_topK": 0.171875, "reward/r_argmax_term": 0.0, "reward/r_target_term": 0.000353826502703214, "reward/r_motif_term": 0.06875, "reward/r_offtarget_term": -0.07277865352079971, "advantage/mean": 8.195638656616211e-08, "advantage/std": 0.999910406456331, "advantage/abs_mean": 0.9684452340006828, "ratio/mean": 1.0024886135943234, "ratio/std": 0.009333383501256887, "ratio/clip_frac_low": 0.0, "ratio/clip_frac_high": 0.0, "kl/mean": 0.0020316243171691895, "kl/abs_mean": 0.006026238203048706, "reward/cell_Ex": -0.003674827018096488}
|
| 7 |
+
{"step": 7, "elapsed_s": 396.02987456321716, "wallclock_s_per_step": 56.57569650241307, "loss/total": 0.0004912540316581726, "loss/per_rollout_mean": 0.0004912605509161949, "loss/per_rollout_std": 0.5081209915960432, "reward/total_mean": -0.06761435881761375, "reward/total_std": 0.08581347992235344, "reward/total_min": -0.17036566138267517, "reward/total_max": 0.09827024008465524, "reward/argmax_match": 0.0, "reward/oracle_target_sigmoid": 3.239763127509468e-06, "reward/oracle_offtarget_sigmoid": 0.4545532623305917, "reward/motif_hit_frac_topK": 0.171875, "reward/r_argmax_term": 0.0, "reward/r_target_term": 1.619881563754734e-06, "reward/r_motif_term": 0.06875, "reward/r_offtarget_term": -0.1363659786991775, "advantage/mean": -1.5832483768463135e-07, "advantage/std": 0.9999682562458455, "advantage/abs_mean": 0.8188682291656733, "ratio/mean": 0.9997414243407547, "ratio/std": 0.0034037470171453667, "ratio/clip_frac_low": 0.0, "ratio/clip_frac_high": 0.0, "kl/mean": -0.00040856003761291504, "kl/abs_mean": 0.0025797784328460693, "reward/cell_Ex": -0.06761435881761375}
|