diff --git "a/last-checkpoint/trainer_state.json" "b/last-checkpoint/trainer_state.json" --- "a/last-checkpoint/trainer_state.json" +++ "b/last-checkpoint/trainer_state.json" @@ -1,9 +1,9 @@ { "best_metric": null, "best_model_checkpoint": null, - "epoch": 0.1941936110301971, + "epoch": 0.25892481470692946, "eval_steps": 5, - "global_step": 1500, + "global_step": 2000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, @@ -15323,6 +15323,5106 @@ "eval_samples_per_second": 3.159, "eval_steps_per_second": 0.215, "step": 1500 + }, + { + "epoch": 0.19432307343755056, + "grad_norm": 4.845419134564698, + "learning_rate": 1.2164728722963713e-06, + "loss": 3.8643, + "step": 1501 + }, + { + "epoch": 0.19445253584490405, + "grad_norm": 3.995491295891669, + "learning_rate": 1.2163815368617574e-06, + "loss": 3.0723, + "step": 1502 + }, + { + "epoch": 0.1945819982522575, + "grad_norm": 3.471203592316862, + "learning_rate": 1.2162900806254778e-06, + "loss": 2.9028, + "step": 1503 + }, + { + "epoch": 0.19471146065961095, + "grad_norm": 3.8927115634833935, + "learning_rate": 1.2161985036062146e-06, + "loss": 3.6755, + "step": 1504 + }, + { + "epoch": 0.19484092306696443, + "grad_norm": 3.66897107403486, + "learning_rate": 1.2161068058226737e-06, + "loss": 3.8296, + "step": 1505 + }, + { + "epoch": 0.19484092306696443, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4242275953292847, + "eval_runtime": 14.1491, + "eval_samples_per_second": 3.11, + "eval_steps_per_second": 0.212, + "step": 1505 + }, + { + "epoch": 0.1949703854743179, + "grad_norm": 3.945085191116555, + "learning_rate": 1.2160149872935866e-06, + "loss": 3.3855, + "step": 1506 + }, + { + "epoch": 0.19509984788167137, + "grad_norm": 3.2532420246187286, + "learning_rate": 1.215923048037709e-06, + "loss": 3.2073, + "step": 1507 + }, + { + "epoch": 0.19522931028902482, + "grad_norm": 2.371061160407878, + "learning_rate": 1.2158309880738215e-06, + "loss": 3.1289, + "step": 1508 + }, + { + "epoch": 0.1953587726963783, + "grad_norm": 2.8974571435739374, + "learning_rate": 1.2157388074207288e-06, + "loss": 3.1313, + "step": 1509 + }, + { + "epoch": 0.19548823510373176, + "grad_norm": 1.6210041807318014, + "learning_rate": 1.2156465060972612e-06, + "loss": 2.6797, + "step": 1510 + }, + { + "epoch": 0.19548823510373176, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.426313877105713, + "eval_runtime": 15.0598, + "eval_samples_per_second": 2.922, + "eval_steps_per_second": 0.199, + "step": 1510 + }, + { + "epoch": 0.1956176975110852, + "grad_norm": 4.338336807951537, + "learning_rate": 1.2155540841222727e-06, + "loss": 3.9966, + "step": 1511 + }, + { + "epoch": 0.1957471599184387, + "grad_norm": 4.199510770528302, + "learning_rate": 1.2154615415146428e-06, + "loss": 3.0269, + "step": 1512 + }, + { + "epoch": 0.19587662232579214, + "grad_norm": 3.933256709381275, + "learning_rate": 1.2153688782932746e-06, + "loss": 3.9829, + "step": 1513 + }, + { + "epoch": 0.19600608473314562, + "grad_norm": 6.708972042680103, + "learning_rate": 1.215276094477097e-06, + "loss": 3.9902, + "step": 1514 + }, + { + "epoch": 0.19613554714049908, + "grad_norm": 2.981069390796975, + "learning_rate": 1.2151831900850627e-06, + "loss": 3.3745, + "step": 1515 + }, + { + "epoch": 0.19613554714049908, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4276012182235718, + "eval_runtime": 13.7181, + "eval_samples_per_second": 3.207, + "eval_steps_per_second": 0.219, + "step": 1515 + }, + { + "epoch": 0.19626500954785253, + "grad_norm": 2.7138718303579954, + "learning_rate": 1.2150901651361493e-06, + "loss": 3.4946, + "step": 1516 + }, + { + "epoch": 0.196394471955206, + "grad_norm": 4.580876306704034, + "learning_rate": 1.2149970196493593e-06, + "loss": 3.4609, + "step": 1517 + }, + { + "epoch": 0.19652393436255947, + "grad_norm": 4.094165253293919, + "learning_rate": 1.2149037536437192e-06, + "loss": 3.4507, + "step": 1518 + }, + { + "epoch": 0.19665339676991295, + "grad_norm": 4.0938811458730475, + "learning_rate": 1.2148103671382805e-06, + "loss": 3.7017, + "step": 1519 + }, + { + "epoch": 0.1967828591772664, + "grad_norm": 6.849544131975171, + "learning_rate": 1.2147168601521194e-06, + "loss": 4.6694, + "step": 1520 + }, + { + "epoch": 0.1967828591772664, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4351917505264282, + "eval_runtime": 14.7195, + "eval_samples_per_second": 2.989, + "eval_steps_per_second": 0.204, + "step": 1520 + }, + { + "epoch": 0.19691232158461985, + "grad_norm": 2.862944819638343, + "learning_rate": 1.2146232327043367e-06, + "loss": 3.3704, + "step": 1521 + }, + { + "epoch": 0.19704178399197333, + "grad_norm": 6.377635113912191, + "learning_rate": 1.2145294848140577e-06, + "loss": 4.2241, + "step": 1522 + }, + { + "epoch": 0.1971712463993268, + "grad_norm": 7.743289927129237, + "learning_rate": 1.214435616500432e-06, + "loss": 4.8267, + "step": 1523 + }, + { + "epoch": 0.19730070880668027, + "grad_norm": 2.641206070102616, + "learning_rate": 1.2143416277826344e-06, + "loss": 3.4771, + "step": 1524 + }, + { + "epoch": 0.19743017121403372, + "grad_norm": 3.197824623342946, + "learning_rate": 1.214247518679864e-06, + "loss": 3.0527, + "step": 1525 + }, + { + "epoch": 0.19743017121403372, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4392756223678589, + "eval_runtime": 14.4369, + "eval_samples_per_second": 3.048, + "eval_steps_per_second": 0.208, + "step": 1525 + }, + { + "epoch": 0.19755963362138718, + "grad_norm": 2.902892943763722, + "learning_rate": 1.2141532892113443e-06, + "loss": 3.0239, + "step": 1526 + }, + { + "epoch": 0.19768909602874066, + "grad_norm": 3.146768487120522, + "learning_rate": 1.2140589393963237e-06, + "loss": 3.2231, + "step": 1527 + }, + { + "epoch": 0.1978185584360941, + "grad_norm": 7.572701064560839, + "learning_rate": 1.213964469254075e-06, + "loss": 4.6699, + "step": 1528 + }, + { + "epoch": 0.1979480208434476, + "grad_norm": 7.8696216770995235, + "learning_rate": 1.2138698788038958e-06, + "loss": 4.4543, + "step": 1529 + }, + { + "epoch": 0.19807748325080105, + "grad_norm": 2.1557226699300807, + "learning_rate": 1.213775168065108e-06, + "loss": 3.0369, + "step": 1530 + }, + { + "epoch": 0.19807748325080105, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.444868564605713, + "eval_runtime": 15.4519, + "eval_samples_per_second": 2.848, + "eval_steps_per_second": 0.194, + "step": 1530 + }, + { + "epoch": 0.19820694565815453, + "grad_norm": 2.3016394267085936, + "learning_rate": 1.2136803370570583e-06, + "loss": 3.012, + "step": 1531 + }, + { + "epoch": 0.19833640806550798, + "grad_norm": 3.189583637555812, + "learning_rate": 1.2135853857991175e-06, + "loss": 3.2603, + "step": 1532 + }, + { + "epoch": 0.19846587047286143, + "grad_norm": 4.261435582477583, + "learning_rate": 1.2134903143106816e-06, + "loss": 3.7969, + "step": 1533 + }, + { + "epoch": 0.19859533288021491, + "grad_norm": 2.3626151320962174, + "learning_rate": 1.2133951226111709e-06, + "loss": 2.7866, + "step": 1534 + }, + { + "epoch": 0.19872479528756837, + "grad_norm": 2.6919069476152626, + "learning_rate": 1.2132998107200304e-06, + "loss": 2.6282, + "step": 1535 + }, + { + "epoch": 0.19872479528756837, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.445845127105713, + "eval_runtime": 15.099, + "eval_samples_per_second": 2.914, + "eval_steps_per_second": 0.199, + "step": 1535 + }, + { + "epoch": 0.19885425769492185, + "grad_norm": 3.305444165670336, + "learning_rate": 1.213204378656729e-06, + "loss": 3.1577, + "step": 1536 + }, + { + "epoch": 0.1989837201022753, + "grad_norm": 4.003643969235668, + "learning_rate": 1.213108826440761e-06, + "loss": 2.9624, + "step": 1537 + }, + { + "epoch": 0.19911318250962876, + "grad_norm": 1.6570602475458145, + "learning_rate": 1.2130131540916447e-06, + "loss": 2.4124, + "step": 1538 + }, + { + "epoch": 0.19924264491698224, + "grad_norm": 2.665600868005522, + "learning_rate": 1.2129173616289233e-06, + "loss": 3.3452, + "step": 1539 + }, + { + "epoch": 0.1993721073243357, + "grad_norm": 4.039641609951077, + "learning_rate": 1.212821449072164e-06, + "loss": 3.7769, + "step": 1540 + }, + { + "epoch": 0.1993721073243357, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4509055614471436, + "eval_runtime": 20.7699, + "eval_samples_per_second": 2.118, + "eval_steps_per_second": 0.144, + "step": 1540 + }, + { + "epoch": 0.19950156973168917, + "grad_norm": 5.880390778230704, + "learning_rate": 1.2127254164409593e-06, + "loss": 3.7339, + "step": 1541 + }, + { + "epoch": 0.19963103213904262, + "grad_norm": 7.626853983434958, + "learning_rate": 1.212629263754926e-06, + "loss": 4.4692, + "step": 1542 + }, + { + "epoch": 0.19976049454639608, + "grad_norm": 1.581514636862654, + "learning_rate": 1.2125329910337045e-06, + "loss": 2.6382, + "step": 1543 + }, + { + "epoch": 0.19988995695374956, + "grad_norm": 11.915379122547142, + "learning_rate": 1.2124365982969607e-06, + "loss": 6.1772, + "step": 1544 + }, + { + "epoch": 0.200019419361103, + "grad_norm": 3.609418599267737, + "learning_rate": 1.212340085564385e-06, + "loss": 2.9995, + "step": 1545 + }, + { + "epoch": 0.200019419361103, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4520596265792847, + "eval_runtime": 15.213, + "eval_samples_per_second": 2.892, + "eval_steps_per_second": 0.197, + "step": 1545 + }, + { + "epoch": 0.2001488817684565, + "grad_norm": 5.78151162096917, + "learning_rate": 1.212243452855692e-06, + "loss": 4.0293, + "step": 1546 + }, + { + "epoch": 0.20027834417580995, + "grad_norm": 2.937787655572086, + "learning_rate": 1.212146700190621e-06, + "loss": 3.1235, + "step": 1547 + }, + { + "epoch": 0.2004078065831634, + "grad_norm": 8.637192704143201, + "learning_rate": 1.2120498275889357e-06, + "loss": 4.1572, + "step": 1548 + }, + { + "epoch": 0.20053726899051688, + "grad_norm": 2.5554512529276026, + "learning_rate": 1.2119528350704237e-06, + "loss": 3.1335, + "step": 1549 + }, + { + "epoch": 0.20066673139787033, + "grad_norm": 5.741191168988011, + "learning_rate": 1.2118557226548984e-06, + "loss": 4.103, + "step": 1550 + }, + { + "epoch": 0.20066673139787033, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4527698755264282, + "eval_runtime": 22.2949, + "eval_samples_per_second": 1.974, + "eval_steps_per_second": 0.135, + "step": 1550 + }, + { + "epoch": 0.20079619380522382, + "grad_norm": 3.087974490934094, + "learning_rate": 1.2117584903621968e-06, + "loss": 3.1123, + "step": 1551 + }, + { + "epoch": 0.20092565621257727, + "grad_norm": 3.9529421240056046, + "learning_rate": 1.2116611382121803e-06, + "loss": 3.6387, + "step": 1552 + }, + { + "epoch": 0.20105511861993075, + "grad_norm": 6.235812177458835, + "learning_rate": 1.2115636662247355e-06, + "loss": 4.0176, + "step": 1553 + }, + { + "epoch": 0.2011845810272842, + "grad_norm": 6.410729264195401, + "learning_rate": 1.2114660744197725e-06, + "loss": 4.2646, + "step": 1554 + }, + { + "epoch": 0.20131404343463766, + "grad_norm": 4.9644377007051705, + "learning_rate": 1.2113683628172269e-06, + "loss": 3.7329, + "step": 1555 + }, + { + "epoch": 0.20131404343463766, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4598721265792847, + "eval_runtime": 13.6725, + "eval_samples_per_second": 3.218, + "eval_steps_per_second": 0.219, + "step": 1555 + }, + { + "epoch": 0.20144350584199114, + "grad_norm": 2.853664447998761, + "learning_rate": 1.2112705314370579e-06, + "loss": 3.3838, + "step": 1556 + }, + { + "epoch": 0.2015729682493446, + "grad_norm": 3.0798359641816684, + "learning_rate": 1.2111725802992498e-06, + "loss": 3.4993, + "step": 1557 + }, + { + "epoch": 0.20170243065669807, + "grad_norm": 1.9122865891922844, + "learning_rate": 1.2110745094238107e-06, + "loss": 2.9854, + "step": 1558 + }, + { + "epoch": 0.20183189306405153, + "grad_norm": 1.8144191675334405, + "learning_rate": 1.2109763188307741e-06, + "loss": 2.9585, + "step": 1559 + }, + { + "epoch": 0.20196135547140498, + "grad_norm": 3.408868530623044, + "learning_rate": 1.2108780085401971e-06, + "loss": 3.6914, + "step": 1560 + }, + { + "epoch": 0.20196135547140498, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4631569385528564, + "eval_runtime": 15.4013, + "eval_samples_per_second": 2.857, + "eval_steps_per_second": 0.195, + "step": 1560 + }, + { + "epoch": 0.20209081787875846, + "grad_norm": 2.6121158439013947, + "learning_rate": 1.2107795785721618e-06, + "loss": 3.2236, + "step": 1561 + }, + { + "epoch": 0.20222028028611191, + "grad_norm": 2.4932650882454106, + "learning_rate": 1.2106810289467742e-06, + "loss": 3.166, + "step": 1562 + }, + { + "epoch": 0.2023497426934654, + "grad_norm": 2.703922425558448, + "learning_rate": 1.2105823596841654e-06, + "loss": 3.6685, + "step": 1563 + }, + { + "epoch": 0.20247920510081885, + "grad_norm": 1.9224382757041614, + "learning_rate": 1.2104835708044903e-06, + "loss": 2.8481, + "step": 1564 + }, + { + "epoch": 0.2026086675081723, + "grad_norm": 2.905178727106571, + "learning_rate": 1.2103846623279285e-06, + "loss": 2.8643, + "step": 1565 + }, + { + "epoch": 0.2026086675081723, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4644886255264282, + "eval_runtime": 14.9309, + "eval_samples_per_second": 2.947, + "eval_steps_per_second": 0.201, + "step": 1565 + }, + { + "epoch": 0.20273812991552578, + "grad_norm": 1.8911415277885153, + "learning_rate": 1.2102856342746846e-06, + "loss": 2.9456, + "step": 1566 + }, + { + "epoch": 0.20286759232287924, + "grad_norm": 1.926688665977454, + "learning_rate": 1.2101864866649863e-06, + "loss": 3.0642, + "step": 1567 + }, + { + "epoch": 0.20299705473023272, + "grad_norm": 4.152946729181612, + "learning_rate": 1.2100872195190872e-06, + "loss": 3.353, + "step": 1568 + }, + { + "epoch": 0.20312651713758617, + "grad_norm": 4.275656626955722, + "learning_rate": 1.2099878328572642e-06, + "loss": 3.731, + "step": 1569 + }, + { + "epoch": 0.20325597954493962, + "grad_norm": 1.7459558279486687, + "learning_rate": 1.2098883266998193e-06, + "loss": 2.6465, + "step": 1570 + }, + { + "epoch": 0.20325597954493962, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.468794345855713, + "eval_runtime": 15.2123, + "eval_samples_per_second": 2.892, + "eval_steps_per_second": 0.197, + "step": 1570 + }, + { + "epoch": 0.2033854419522931, + "grad_norm": 4.85287549461254, + "learning_rate": 1.2097887010670783e-06, + "loss": 3.8296, + "step": 1571 + }, + { + "epoch": 0.20351490435964656, + "grad_norm": 4.527523924353318, + "learning_rate": 1.2096889559793921e-06, + "loss": 4.27, + "step": 1572 + }, + { + "epoch": 0.20364436676700004, + "grad_norm": 4.865976890858769, + "learning_rate": 1.2095890914571356e-06, + "loss": 3.9246, + "step": 1573 + }, + { + "epoch": 0.2037738291743535, + "grad_norm": 2.8317211998897176, + "learning_rate": 1.209489107520708e-06, + "loss": 2.9395, + "step": 1574 + }, + { + "epoch": 0.20390329158170697, + "grad_norm": 2.3545586555200737, + "learning_rate": 1.2093890041905332e-06, + "loss": 3.0503, + "step": 1575 + }, + { + "epoch": 0.20390329158170697, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4699928760528564, + "eval_runtime": 14.4633, + "eval_samples_per_second": 3.042, + "eval_steps_per_second": 0.207, + "step": 1575 + }, + { + "epoch": 0.20403275398906043, + "grad_norm": 5.7779708346918035, + "learning_rate": 1.2092887814870593e-06, + "loss": 3.9587, + "step": 1576 + }, + { + "epoch": 0.20416221639641388, + "grad_norm": 2.4822541762284733, + "learning_rate": 1.2091884394307587e-06, + "loss": 2.8979, + "step": 1577 + }, + { + "epoch": 0.20429167880376736, + "grad_norm": 5.3614084166007006, + "learning_rate": 1.2090879780421282e-06, + "loss": 3.2603, + "step": 1578 + }, + { + "epoch": 0.20442114121112082, + "grad_norm": 3.546560236927702, + "learning_rate": 1.2089873973416896e-06, + "loss": 3.2593, + "step": 1579 + }, + { + "epoch": 0.2045506036184743, + "grad_norm": 2.640294338800154, + "learning_rate": 1.2088866973499882e-06, + "loss": 3.1023, + "step": 1580 + }, + { + "epoch": 0.2045506036184743, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4705699682235718, + "eval_runtime": 14.607, + "eval_samples_per_second": 3.012, + "eval_steps_per_second": 0.205, + "step": 1580 + }, + { + "epoch": 0.20468006602582775, + "grad_norm": 5.66685869149829, + "learning_rate": 1.2087858780875938e-06, + "loss": 4.0166, + "step": 1581 + }, + { + "epoch": 0.2048095284331812, + "grad_norm": 2.334032769188969, + "learning_rate": 1.2086849395751011e-06, + "loss": 2.9531, + "step": 1582 + }, + { + "epoch": 0.20493899084053468, + "grad_norm": 3.1929631814622605, + "learning_rate": 1.208583881833129e-06, + "loss": 3.574, + "step": 1583 + }, + { + "epoch": 0.20506845324788814, + "grad_norm": 8.522138757685294, + "learning_rate": 1.2084827048823198e-06, + "loss": 4.2388, + "step": 1584 + }, + { + "epoch": 0.20519791565524162, + "grad_norm": 2.1435488667101383, + "learning_rate": 1.2083814087433418e-06, + "loss": 2.689, + "step": 1585 + }, + { + "epoch": 0.20519791565524162, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4782493114471436, + "eval_runtime": 14.5609, + "eval_samples_per_second": 3.022, + "eval_steps_per_second": 0.206, + "step": 1585 + }, + { + "epoch": 0.20532737806259507, + "grad_norm": 2.892395805151098, + "learning_rate": 1.2082799934368864e-06, + "loss": 3.0242, + "step": 1586 + }, + { + "epoch": 0.20545684046994853, + "grad_norm": 2.522406926650977, + "learning_rate": 1.2081784589836699e-06, + "loss": 3.1038, + "step": 1587 + }, + { + "epoch": 0.205586302877302, + "grad_norm": 5.181832917055719, + "learning_rate": 1.2080768054044324e-06, + "loss": 4.0276, + "step": 1588 + }, + { + "epoch": 0.20571576528465546, + "grad_norm": 3.348329622132548, + "learning_rate": 1.207975032719939e-06, + "loss": 3.0703, + "step": 1589 + }, + { + "epoch": 0.20584522769200894, + "grad_norm": 2.065729560958714, + "learning_rate": 1.2078731409509792e-06, + "loss": 2.8737, + "step": 1590 + }, + { + "epoch": 0.20584522769200894, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4801136255264282, + "eval_runtime": 14.1584, + "eval_samples_per_second": 3.108, + "eval_steps_per_second": 0.212, + "step": 1590 + }, + { + "epoch": 0.2059746900993624, + "grad_norm": 2.6355991484725743, + "learning_rate": 1.2077711301183656e-06, + "loss": 2.8242, + "step": 1591 + }, + { + "epoch": 0.20610415250671585, + "grad_norm": 4.223252780591359, + "learning_rate": 1.2076690002429368e-06, + "loss": 3.6353, + "step": 1592 + }, + { + "epoch": 0.20623361491406933, + "grad_norm": 4.212169138372161, + "learning_rate": 1.2075667513455542e-06, + "loss": 3.6245, + "step": 1593 + }, + { + "epoch": 0.20636307732142278, + "grad_norm": 6.096225964591917, + "learning_rate": 1.2074643834471048e-06, + "loss": 4.0625, + "step": 1594 + }, + { + "epoch": 0.20649253972877626, + "grad_norm": 2.401444677656407, + "learning_rate": 1.207361896568499e-06, + "loss": 3.3516, + "step": 1595 + }, + { + "epoch": 0.20649253972877626, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4830876588821411, + "eval_runtime": 19.1139, + "eval_samples_per_second": 2.302, + "eval_steps_per_second": 0.157, + "step": 1595 + }, + { + "epoch": 0.20662200213612972, + "grad_norm": 4.01192822886966, + "learning_rate": 1.2072592907306718e-06, + "loss": 3.1951, + "step": 1596 + }, + { + "epoch": 0.2067514645434832, + "grad_norm": 1.6010028669109733, + "learning_rate": 1.2071565659545826e-06, + "loss": 2.8215, + "step": 1597 + }, + { + "epoch": 0.20688092695083665, + "grad_norm": 5.6261882001825905, + "learning_rate": 1.2070537222612153e-06, + "loss": 3.936, + "step": 1598 + }, + { + "epoch": 0.2070103893581901, + "grad_norm": 2.675115238277584, + "learning_rate": 1.2069507596715773e-06, + "loss": 3.0876, + "step": 1599 + }, + { + "epoch": 0.2071398517655436, + "grad_norm": 5.644124802210082, + "learning_rate": 1.206847678206701e-06, + "loss": 3.5344, + "step": 1600 + }, + { + "epoch": 0.2071398517655436, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4878817796707153, + "eval_runtime": 14.9199, + "eval_samples_per_second": 2.949, + "eval_steps_per_second": 0.201, + "step": 1600 + }, + { + "epoch": 0.20726931417289704, + "grad_norm": 3.059641611880596, + "learning_rate": 1.206744477887643e-06, + "loss": 3.1641, + "step": 1601 + }, + { + "epoch": 0.20739877658025052, + "grad_norm": 1.5856627740509093, + "learning_rate": 1.206641158735484e-06, + "loss": 2.4263, + "step": 1602 + }, + { + "epoch": 0.20752823898760397, + "grad_norm": 2.9880133530734776, + "learning_rate": 1.2065377207713288e-06, + "loss": 3.6694, + "step": 1603 + }, + { + "epoch": 0.20765770139495743, + "grad_norm": 3.1302800518305873, + "learning_rate": 1.2064341640163071e-06, + "loss": 3.0247, + "step": 1604 + }, + { + "epoch": 0.2077871638023109, + "grad_norm": 3.2710387355775454, + "learning_rate": 1.2063304884915722e-06, + "loss": 3.0269, + "step": 1605 + }, + { + "epoch": 0.2077871638023109, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4912109375, + "eval_runtime": 14.9049, + "eval_samples_per_second": 2.952, + "eval_steps_per_second": 0.201, + "step": 1605 + }, + { + "epoch": 0.20791662620966436, + "grad_norm": 4.946247369248822, + "learning_rate": 1.206226694218302e-06, + "loss": 3.6489, + "step": 1606 + }, + { + "epoch": 0.20804608861701784, + "grad_norm": 2.5348021081816805, + "learning_rate": 1.2061227812176986e-06, + "loss": 3.1934, + "step": 1607 + }, + { + "epoch": 0.2081755510243713, + "grad_norm": 3.2512636177454306, + "learning_rate": 1.2060187495109884e-06, + "loss": 3.3118, + "step": 1608 + }, + { + "epoch": 0.20830501343172475, + "grad_norm": 6.142824959240283, + "learning_rate": 1.2059145991194218e-06, + "loss": 4.3799, + "step": 1609 + }, + { + "epoch": 0.20843447583907823, + "grad_norm": 7.39309778427709, + "learning_rate": 1.2058103300642737e-06, + "loss": 4.1504, + "step": 1610 + }, + { + "epoch": 0.20843447583907823, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4938743114471436, + "eval_runtime": 14.5684, + "eval_samples_per_second": 3.02, + "eval_steps_per_second": 0.206, + "step": 1610 + }, + { + "epoch": 0.20856393824643168, + "grad_norm": 3.629887536139999, + "learning_rate": 1.2057059423668433e-06, + "loss": 3.2402, + "step": 1611 + }, + { + "epoch": 0.20869340065378517, + "grad_norm": 5.163285639039155, + "learning_rate": 1.2056014360484535e-06, + "loss": 3.3267, + "step": 1612 + }, + { + "epoch": 0.20882286306113862, + "grad_norm": 2.4021206836647178, + "learning_rate": 1.2054968111304525e-06, + "loss": 3.1245, + "step": 1613 + }, + { + "epoch": 0.20895232546849207, + "grad_norm": 4.3005818277725485, + "learning_rate": 1.2053920676342115e-06, + "loss": 3.4814, + "step": 1614 + }, + { + "epoch": 0.20908178787584555, + "grad_norm": 6.599324441815224, + "learning_rate": 1.2052872055811268e-06, + "loss": 4.0947, + "step": 1615 + }, + { + "epoch": 0.20908178787584555, + "eval_PRM Accuracy": 0.25471698113207547, + "eval_PRM F1": 0.09195402298850575, + "eval_PRM F1 AUC": 0.5240963855421686, + "eval_PRM F1 Neg": 0.368, + "eval_PRM NPV": 0.22549019607843138, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.04819277108433735, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.4957829713821411, + "eval_runtime": 15.0369, + "eval_samples_per_second": 2.926, + "eval_steps_per_second": 0.2, + "step": 1615 + }, + { + "epoch": 0.209211250283199, + "grad_norm": 2.4757001552458333, + "learning_rate": 1.2051822249926184e-06, + "loss": 3.1318, + "step": 1616 + }, + { + "epoch": 0.2093407126905525, + "grad_norm": 3.4909574273882766, + "learning_rate": 1.2050771258901309e-06, + "loss": 3.4346, + "step": 1617 + }, + { + "epoch": 0.20947017509790594, + "grad_norm": 2.507228023621561, + "learning_rate": 1.2049719082951326e-06, + "loss": 2.9487, + "step": 1618 + }, + { + "epoch": 0.20959963750525942, + "grad_norm": 3.015430572085104, + "learning_rate": 1.2048665722291167e-06, + "loss": 3.2109, + "step": 1619 + }, + { + "epoch": 0.20972909991261288, + "grad_norm": 2.444443113855727, + "learning_rate": 1.2047611177135998e-06, + "loss": 2.8706, + "step": 1620 + }, + { + "epoch": 0.20972909991261288, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5016423463821411, + "eval_runtime": 15.6167, + "eval_samples_per_second": 2.817, + "eval_steps_per_second": 0.192, + "step": 1620 + }, + { + "epoch": 0.20985856231996633, + "grad_norm": 5.684682992040662, + "learning_rate": 1.2046555447701236e-06, + "loss": 3.8535, + "step": 1621 + }, + { + "epoch": 0.2099880247273198, + "grad_norm": 3.5482341679820357, + "learning_rate": 1.2045498534202532e-06, + "loss": 3.0977, + "step": 1622 + }, + { + "epoch": 0.21011748713467326, + "grad_norm": 6.18356871089719, + "learning_rate": 1.2044440436855786e-06, + "loss": 5.0762, + "step": 1623 + }, + { + "epoch": 0.21024694954202675, + "grad_norm": 4.861517687640406, + "learning_rate": 1.204338115587713e-06, + "loss": 3.8105, + "step": 1624 + }, + { + "epoch": 0.2103764119493802, + "grad_norm": 4.7212659396117, + "learning_rate": 1.2042320691482947e-06, + "loss": 3.8896, + "step": 1625 + }, + { + "epoch": 0.2103764119493802, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5000888109207153, + "eval_runtime": 14.2775, + "eval_samples_per_second": 3.082, + "eval_steps_per_second": 0.21, + "step": 1625 + }, + { + "epoch": 0.21050587435673365, + "grad_norm": 2.1688058325535056, + "learning_rate": 1.2041259043889859e-06, + "loss": 2.8838, + "step": 1626 + }, + { + "epoch": 0.21063533676408713, + "grad_norm": 4.851804058269344, + "learning_rate": 1.2040196213314725e-06, + "loss": 3.7476, + "step": 1627 + }, + { + "epoch": 0.2107647991714406, + "grad_norm": 4.1047585664066375, + "learning_rate": 1.2039132199974654e-06, + "loss": 3.533, + "step": 1628 + }, + { + "epoch": 0.21089426157879407, + "grad_norm": 3.5235614311939094, + "learning_rate": 1.203806700408699e-06, + "loss": 2.9795, + "step": 1629 + }, + { + "epoch": 0.21102372398614752, + "grad_norm": 2.9896742926939095, + "learning_rate": 1.2037000625869322e-06, + "loss": 3.0508, + "step": 1630 + }, + { + "epoch": 0.21102372398614752, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5042613744735718, + "eval_runtime": 14.4724, + "eval_samples_per_second": 3.04, + "eval_steps_per_second": 0.207, + "step": 1630 + }, + { + "epoch": 0.21115318639350097, + "grad_norm": 2.2886077037850994, + "learning_rate": 1.203593306553948e-06, + "loss": 2.9263, + "step": 1631 + }, + { + "epoch": 0.21128264880085446, + "grad_norm": 3.9010393418191804, + "learning_rate": 1.2034864323315534e-06, + "loss": 3.1021, + "step": 1632 + }, + { + "epoch": 0.2114121112082079, + "grad_norm": 9.105088685323679, + "learning_rate": 1.2033794399415797e-06, + "loss": 5.1152, + "step": 1633 + }, + { + "epoch": 0.2115415736155614, + "grad_norm": 1.9548685583033572, + "learning_rate": 1.2032723294058818e-06, + "loss": 2.7971, + "step": 1634 + }, + { + "epoch": 0.21167103602291484, + "grad_norm": 4.058831672758348, + "learning_rate": 1.20316510074634e-06, + "loss": 3.0864, + "step": 1635 + }, + { + "epoch": 0.21167103602291484, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5087890625, + "eval_runtime": 14.3289, + "eval_samples_per_second": 3.071, + "eval_steps_per_second": 0.209, + "step": 1635 + }, + { + "epoch": 0.21180049843026832, + "grad_norm": 4.21247258213656, + "learning_rate": 1.2030577539848572e-06, + "loss": 3.291, + "step": 1636 + }, + { + "epoch": 0.21192996083762178, + "grad_norm": 2.5602033146180037, + "learning_rate": 1.2029502891433613e-06, + "loss": 2.9839, + "step": 1637 + }, + { + "epoch": 0.21205942324497523, + "grad_norm": 2.835412945575613, + "learning_rate": 1.2028427062438047e-06, + "loss": 3.4473, + "step": 1638 + }, + { + "epoch": 0.2121888856523287, + "grad_norm": 3.3485353118282166, + "learning_rate": 1.202735005308163e-06, + "loss": 3.4084, + "step": 1639 + }, + { + "epoch": 0.21231834805968217, + "grad_norm": 2.1071565396713496, + "learning_rate": 1.202627186358436e-06, + "loss": 3.0376, + "step": 1640 + }, + { + "epoch": 0.21231834805968217, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5138493776321411, + "eval_runtime": 14.8914, + "eval_samples_per_second": 2.955, + "eval_steps_per_second": 0.201, + "step": 1640 + }, + { + "epoch": 0.21244781046703565, + "grad_norm": 2.4917125372301925, + "learning_rate": 1.2025192494166485e-06, + "loss": 2.8843, + "step": 1641 + }, + { + "epoch": 0.2125772728743891, + "grad_norm": 5.615226008186244, + "learning_rate": 1.2024111945048484e-06, + "loss": 3.312, + "step": 1642 + }, + { + "epoch": 0.21270673528174255, + "grad_norm": 1.8900795222263203, + "learning_rate": 1.2023030216451083e-06, + "loss": 2.6948, + "step": 1643 + }, + { + "epoch": 0.21283619768909603, + "grad_norm": 2.2432943589139924, + "learning_rate": 1.2021947308595246e-06, + "loss": 2.7708, + "step": 1644 + }, + { + "epoch": 0.2129656600964495, + "grad_norm": 4.01607155678928, + "learning_rate": 1.2020863221702179e-06, + "loss": 3.1877, + "step": 1645 + }, + { + "epoch": 0.2129656600964495, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5144264698028564, + "eval_runtime": 14.6713, + "eval_samples_per_second": 2.999, + "eval_steps_per_second": 0.204, + "step": 1645 + }, + { + "epoch": 0.21309512250380297, + "grad_norm": 4.094127544488313, + "learning_rate": 1.201977795599333e-06, + "loss": 3.4551, + "step": 1646 + }, + { + "epoch": 0.21322458491115642, + "grad_norm": 3.8544861193389544, + "learning_rate": 1.2018691511690384e-06, + "loss": 3.2529, + "step": 1647 + }, + { + "epoch": 0.21335404731850988, + "grad_norm": 3.2845593181976716, + "learning_rate": 1.201760388901527e-06, + "loss": 3.1301, + "step": 1648 + }, + { + "epoch": 0.21348350972586336, + "grad_norm": 4.033912929907339, + "learning_rate": 1.2016515088190158e-06, + "loss": 3.3105, + "step": 1649 + }, + { + "epoch": 0.2136129721332168, + "grad_norm": 5.529040187528592, + "learning_rate": 1.201542510943746e-06, + "loss": 3.6665, + "step": 1650 + }, + { + "epoch": 0.2136129721332168, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5193092823028564, + "eval_runtime": 14.5496, + "eval_samples_per_second": 3.024, + "eval_steps_per_second": 0.206, + "step": 1650 + }, + { + "epoch": 0.2137424345405703, + "grad_norm": 2.340119716194272, + "learning_rate": 1.2014333952979821e-06, + "loss": 3.1594, + "step": 1651 + }, + { + "epoch": 0.21387189694792375, + "grad_norm": 4.962570979093016, + "learning_rate": 1.2013241619040134e-06, + "loss": 3.3188, + "step": 1652 + }, + { + "epoch": 0.2140013593552772, + "grad_norm": 1.4894036679197773, + "learning_rate": 1.2012148107841532e-06, + "loss": 2.748, + "step": 1653 + }, + { + "epoch": 0.21413082176263068, + "grad_norm": 4.650532109161744, + "learning_rate": 1.2011053419607385e-06, + "loss": 3.6257, + "step": 1654 + }, + { + "epoch": 0.21426028416998413, + "grad_norm": 1.4126791675097967, + "learning_rate": 1.2009957554561308e-06, + "loss": 2.2444, + "step": 1655 + }, + { + "epoch": 0.21426028416998413, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5247248411178589, + "eval_runtime": 14.211, + "eval_samples_per_second": 3.096, + "eval_steps_per_second": 0.211, + "step": 1655 + }, + { + "epoch": 0.21438974657733761, + "grad_norm": 2.5460850375869306, + "learning_rate": 1.2008860512927148e-06, + "loss": 3.3462, + "step": 1656 + }, + { + "epoch": 0.21451920898469107, + "grad_norm": 4.613942727621425, + "learning_rate": 1.2007762294929004e-06, + "loss": 3.728, + "step": 1657 + }, + { + "epoch": 0.21464867139204455, + "grad_norm": 4.833038520914195, + "learning_rate": 1.2006662900791204e-06, + "loss": 3.0662, + "step": 1658 + }, + { + "epoch": 0.214778133799398, + "grad_norm": 4.31468274739582, + "learning_rate": 1.2005562330738326e-06, + "loss": 3.7954, + "step": 1659 + }, + { + "epoch": 0.21490759620675146, + "grad_norm": 3.441702980385447, + "learning_rate": 1.2004460584995184e-06, + "loss": 2.8105, + "step": 1660 + }, + { + "epoch": 0.21490759620675146, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5227272510528564, + "eval_runtime": 14.1175, + "eval_samples_per_second": 3.117, + "eval_steps_per_second": 0.213, + "step": 1660 + }, + { + "epoch": 0.21503705861410494, + "grad_norm": 4.146549975653297, + "learning_rate": 1.2003357663786827e-06, + "loss": 3.7661, + "step": 1661 + }, + { + "epoch": 0.2151665210214584, + "grad_norm": 2.5925309897445974, + "learning_rate": 1.2002253567338554e-06, + "loss": 3.0469, + "step": 1662 + }, + { + "epoch": 0.21529598342881187, + "grad_norm": 4.031768209484338, + "learning_rate": 1.2001148295875894e-06, + "loss": 3.835, + "step": 1663 + }, + { + "epoch": 0.21542544583616532, + "grad_norm": 6.38961541653933, + "learning_rate": 1.2000041849624627e-06, + "loss": 4.2883, + "step": 1664 + }, + { + "epoch": 0.21555490824351878, + "grad_norm": 3.6936392660998782, + "learning_rate": 1.1998934228810763e-06, + "loss": 3.3057, + "step": 1665 + }, + { + "epoch": 0.21555490824351878, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5295188426971436, + "eval_runtime": 14.3642, + "eval_samples_per_second": 3.063, + "eval_steps_per_second": 0.209, + "step": 1665 + }, + { + "epoch": 0.21568437065087226, + "grad_norm": 5.84206550953815, + "learning_rate": 1.1997825433660557e-06, + "loss": 3.7471, + "step": 1666 + }, + { + "epoch": 0.2158138330582257, + "grad_norm": 3.6429972516641853, + "learning_rate": 1.1996715464400504e-06, + "loss": 3.5845, + "step": 1667 + }, + { + "epoch": 0.2159432954655792, + "grad_norm": 3.3123082689141894, + "learning_rate": 1.1995604321257335e-06, + "loss": 3.188, + "step": 1668 + }, + { + "epoch": 0.21607275787293265, + "grad_norm": 5.229457376640021, + "learning_rate": 1.1994492004458028e-06, + "loss": 3.0566, + "step": 1669 + }, + { + "epoch": 0.2162022202802861, + "grad_norm": 4.974992338949791, + "learning_rate": 1.1993378514229791e-06, + "loss": 3.3535, + "step": 1670 + }, + { + "epoch": 0.2162022202802861, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.530229091644287, + "eval_runtime": 14.6104, + "eval_samples_per_second": 3.012, + "eval_steps_per_second": 0.205, + "step": 1670 + }, + { + "epoch": 0.21633168268763958, + "grad_norm": 4.905074376396741, + "learning_rate": 1.199226385080008e-06, + "loss": 3.427, + "step": 1671 + }, + { + "epoch": 0.21646114509499303, + "grad_norm": 2.169399331763751, + "learning_rate": 1.1991148014396587e-06, + "loss": 3.1394, + "step": 1672 + }, + { + "epoch": 0.21659060750234652, + "grad_norm": 3.729323603756549, + "learning_rate": 1.1990031005247247e-06, + "loss": 3.2573, + "step": 1673 + }, + { + "epoch": 0.21672006990969997, + "grad_norm": 1.6181688473384985, + "learning_rate": 1.1988912823580226e-06, + "loss": 2.7397, + "step": 1674 + }, + { + "epoch": 0.21684953231705342, + "grad_norm": 2.9622056511634325, + "learning_rate": 1.1987793469623939e-06, + "loss": 3.342, + "step": 1675 + }, + { + "epoch": 0.21684953231705342, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5323597192764282, + "eval_runtime": 14.6107, + "eval_samples_per_second": 3.011, + "eval_steps_per_second": 0.205, + "step": 1675 + }, + { + "epoch": 0.2169789947244069, + "grad_norm": 5.0521818109629, + "learning_rate": 1.1986672943607037e-06, + "loss": 3.6797, + "step": 1676 + }, + { + "epoch": 0.21710845713176036, + "grad_norm": 3.2029765164963844, + "learning_rate": 1.1985551245758408e-06, + "loss": 3.3445, + "step": 1677 + }, + { + "epoch": 0.21723791953911384, + "grad_norm": 3.4842845942844165, + "learning_rate": 1.1984428376307184e-06, + "loss": 3.7869, + "step": 1678 + }, + { + "epoch": 0.2173673819464673, + "grad_norm": 3.708751412318944, + "learning_rate": 1.198330433548273e-06, + "loss": 3.0625, + "step": 1679 + }, + { + "epoch": 0.21749684435382077, + "grad_norm": 1.9442744771078875, + "learning_rate": 1.198217912351466e-06, + "loss": 2.668, + "step": 1680 + }, + { + "epoch": 0.21749684435382077, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5377751588821411, + "eval_runtime": 13.9417, + "eval_samples_per_second": 3.156, + "eval_steps_per_second": 0.215, + "step": 1680 + }, + { + "epoch": 0.21762630676117423, + "grad_norm": 2.067723877629095, + "learning_rate": 1.1981052740632815e-06, + "loss": 2.719, + "step": 1681 + }, + { + "epoch": 0.21775576916852768, + "grad_norm": 4.462641611864524, + "learning_rate": 1.1979925187067287e-06, + "loss": 3.4854, + "step": 1682 + }, + { + "epoch": 0.21788523157588116, + "grad_norm": 4.811453825784267, + "learning_rate": 1.1978796463048398e-06, + "loss": 3.5254, + "step": 1683 + }, + { + "epoch": 0.21801469398323461, + "grad_norm": 2.860584095818227, + "learning_rate": 1.1977666568806715e-06, + "loss": 3.0029, + "step": 1684 + }, + { + "epoch": 0.2181441563905881, + "grad_norm": 3.0771830392382378, + "learning_rate": 1.1976535504573038e-06, + "loss": 3.1792, + "step": 1685 + }, + { + "epoch": 0.2181441563905881, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5383522510528564, + "eval_runtime": 14.5011, + "eval_samples_per_second": 3.034, + "eval_steps_per_second": 0.207, + "step": 1685 + }, + { + "epoch": 0.21827361879794155, + "grad_norm": 5.2504609356974745, + "learning_rate": 1.1975403270578416e-06, + "loss": 4.166, + "step": 1686 + }, + { + "epoch": 0.218403081205295, + "grad_norm": 3.6410347766971634, + "learning_rate": 1.1974269867054124e-06, + "loss": 3.1914, + "step": 1687 + }, + { + "epoch": 0.21853254361264848, + "grad_norm": 3.277556647628424, + "learning_rate": 1.197313529423169e-06, + "loss": 3.2913, + "step": 1688 + }, + { + "epoch": 0.21866200602000194, + "grad_norm": 3.187593779756875, + "learning_rate": 1.1971999552342867e-06, + "loss": 3.1484, + "step": 1689 + }, + { + "epoch": 0.21879146842735542, + "grad_norm": 3.6581996248932533, + "learning_rate": 1.1970862641619657e-06, + "loss": 3.1729, + "step": 1690 + }, + { + "epoch": 0.21879146842735542, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5446555614471436, + "eval_runtime": 14.3306, + "eval_samples_per_second": 3.07, + "eval_steps_per_second": 0.209, + "step": 1690 + }, + { + "epoch": 0.21892093083470887, + "grad_norm": 3.4757812556040912, + "learning_rate": 1.1969724562294294e-06, + "loss": 3.3074, + "step": 1691 + }, + { + "epoch": 0.21905039324206232, + "grad_norm": 2.303171804191525, + "learning_rate": 1.196858531459926e-06, + "loss": 2.8193, + "step": 1692 + }, + { + "epoch": 0.2191798556494158, + "grad_norm": 7.223176287536503, + "learning_rate": 1.196744489876726e-06, + "loss": 4.1494, + "step": 1693 + }, + { + "epoch": 0.21930931805676926, + "grad_norm": 5.135747445169343, + "learning_rate": 1.1966303315031254e-06, + "loss": 3.8594, + "step": 1694 + }, + { + "epoch": 0.21943878046412274, + "grad_norm": 2.8515472222807596, + "learning_rate": 1.1965160563624432e-06, + "loss": 2.7759, + "step": 1695 + }, + { + "epoch": 0.21943878046412274, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.548828125, + "eval_runtime": 14.1962, + "eval_samples_per_second": 3.099, + "eval_steps_per_second": 0.211, + "step": 1695 + }, + { + "epoch": 0.2195682428714762, + "grad_norm": 2.4805486640742993, + "learning_rate": 1.1964016644780222e-06, + "loss": 2.8004, + "step": 1696 + }, + { + "epoch": 0.21969770527882965, + "grad_norm": 6.763167382749191, + "learning_rate": 1.1962871558732297e-06, + "loss": 4.3635, + "step": 1697 + }, + { + "epoch": 0.21982716768618313, + "grad_norm": 3.2400450004153827, + "learning_rate": 1.196172530571456e-06, + "loss": 3.4238, + "step": 1698 + }, + { + "epoch": 0.21995663009353658, + "grad_norm": 3.9635746979996807, + "learning_rate": 1.1960577885961158e-06, + "loss": 3.2061, + "step": 1699 + }, + { + "epoch": 0.22008609250089006, + "grad_norm": 2.9452614763167952, + "learning_rate": 1.1959429299706477e-06, + "loss": 2.8008, + "step": 1700 + }, + { + "epoch": 0.22008609250089006, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5465642213821411, + "eval_runtime": 13.6307, + "eval_samples_per_second": 3.228, + "eval_steps_per_second": 0.22, + "step": 1700 + }, + { + "epoch": 0.22021555490824352, + "grad_norm": 3.497797877571419, + "learning_rate": 1.1958279547185133e-06, + "loss": 3.2026, + "step": 1701 + }, + { + "epoch": 0.220345017315597, + "grad_norm": 2.033177315949458, + "learning_rate": 1.1957128628631991e-06, + "loss": 2.9541, + "step": 1702 + }, + { + "epoch": 0.22047447972295045, + "grad_norm": 4.255071739556924, + "learning_rate": 1.195597654428215e-06, + "loss": 3.1274, + "step": 1703 + }, + { + "epoch": 0.2206039421303039, + "grad_norm": 4.309098485968686, + "learning_rate": 1.1954823294370944e-06, + "loss": 3.3474, + "step": 1704 + }, + { + "epoch": 0.22073340453765739, + "grad_norm": 3.958231854238835, + "learning_rate": 1.1953668879133949e-06, + "loss": 3.2529, + "step": 1705 + }, + { + "epoch": 0.22073340453765739, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.550248622894287, + "eval_runtime": 14.5991, + "eval_samples_per_second": 3.014, + "eval_steps_per_second": 0.205, + "step": 1705 + }, + { + "epoch": 0.22086286694501084, + "grad_norm": 8.126279684991312, + "learning_rate": 1.1952513298806976e-06, + "loss": 4.4897, + "step": 1706 + }, + { + "epoch": 0.22099232935236432, + "grad_norm": 5.890598154123479, + "learning_rate": 1.1951356553626079e-06, + "loss": 4.1187, + "step": 1707 + }, + { + "epoch": 0.22112179175971777, + "grad_norm": 5.0241134576385384, + "learning_rate": 1.1950198643827543e-06, + "loss": 3.6501, + "step": 1708 + }, + { + "epoch": 0.22125125416707123, + "grad_norm": 2.268935548340834, + "learning_rate": 1.1949039569647897e-06, + "loss": 2.9326, + "step": 1709 + }, + { + "epoch": 0.2213807165744247, + "grad_norm": 5.984558256083068, + "learning_rate": 1.1947879331323905e-06, + "loss": 3.3857, + "step": 1710 + }, + { + "epoch": 0.2213807165744247, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5528675317764282, + "eval_runtime": 14.5547, + "eval_samples_per_second": 3.023, + "eval_steps_per_second": 0.206, + "step": 1710 + }, + { + "epoch": 0.22151017898177816, + "grad_norm": 1.967178592416272, + "learning_rate": 1.194671792909257e-06, + "loss": 3.0688, + "step": 1711 + }, + { + "epoch": 0.22163964138913164, + "grad_norm": 5.234466646323892, + "learning_rate": 1.1945555363191127e-06, + "loss": 3.4585, + "step": 1712 + }, + { + "epoch": 0.2217691037964851, + "grad_norm": 4.72821504290605, + "learning_rate": 1.194439163385706e-06, + "loss": 4.0537, + "step": 1713 + }, + { + "epoch": 0.22189856620383855, + "grad_norm": 9.418206175108816, + "learning_rate": 1.1943226741328082e-06, + "loss": 5.1987, + "step": 1714 + }, + { + "epoch": 0.22202802861119203, + "grad_norm": 3.0488772639622534, + "learning_rate": 1.1942060685842144e-06, + "loss": 3.2327, + "step": 1715 + }, + { + "epoch": 0.22202802861119203, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.553755283355713, + "eval_runtime": 18.9358, + "eval_samples_per_second": 2.324, + "eval_steps_per_second": 0.158, + "step": 1715 + }, + { + "epoch": 0.22215749101854548, + "grad_norm": 3.1031933334475554, + "learning_rate": 1.194089346763744e-06, + "loss": 3.1274, + "step": 1716 + }, + { + "epoch": 0.22228695342589896, + "grad_norm": 2.3409504141769717, + "learning_rate": 1.1939725086952395e-06, + "loss": 2.5083, + "step": 1717 + }, + { + "epoch": 0.22241641583325242, + "grad_norm": 1.9560616620977689, + "learning_rate": 1.1938555544025675e-06, + "loss": 2.7908, + "step": 1718 + }, + { + "epoch": 0.22254587824060587, + "grad_norm": 3.9541086621729105, + "learning_rate": 1.1937384839096184e-06, + "loss": 3.2656, + "step": 1719 + }, + { + "epoch": 0.22267534064795935, + "grad_norm": 2.2271881270141907, + "learning_rate": 1.1936212972403062e-06, + "loss": 2.8665, + "step": 1720 + }, + { + "epoch": 0.22267534064795935, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5585049390792847, + "eval_runtime": 13.9785, + "eval_samples_per_second": 3.148, + "eval_steps_per_second": 0.215, + "step": 1720 + }, + { + "epoch": 0.2228048030553128, + "grad_norm": 5.169693777968014, + "learning_rate": 1.1935039944185685e-06, + "loss": 3.8398, + "step": 1721 + }, + { + "epoch": 0.2229342654626663, + "grad_norm": 1.979788095273356, + "learning_rate": 1.193386575468367e-06, + "loss": 3.1157, + "step": 1722 + }, + { + "epoch": 0.22306372787001974, + "grad_norm": 4.605638918372798, + "learning_rate": 1.1932690404136869e-06, + "loss": 3.1528, + "step": 1723 + }, + { + "epoch": 0.22319319027737322, + "grad_norm": 2.421821502820245, + "learning_rate": 1.1931513892785366e-06, + "loss": 2.8501, + "step": 1724 + }, + { + "epoch": 0.22332265268472667, + "grad_norm": 3.8931793816908336, + "learning_rate": 1.1930336220869495e-06, + "loss": 3.2036, + "step": 1725 + }, + { + "epoch": 0.22332265268472667, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5626775026321411, + "eval_runtime": 14.462, + "eval_samples_per_second": 3.042, + "eval_steps_per_second": 0.207, + "step": 1725 + }, + { + "epoch": 0.22345211509208013, + "grad_norm": 5.490700890607397, + "learning_rate": 1.1929157388629815e-06, + "loss": 3.8892, + "step": 1726 + }, + { + "epoch": 0.2235815774994336, + "grad_norm": 4.815514048050006, + "learning_rate": 1.1927977396307126e-06, + "loss": 3.6133, + "step": 1727 + }, + { + "epoch": 0.22371103990678706, + "grad_norm": 2.014094366630197, + "learning_rate": 1.1926796244142466e-06, + "loss": 2.9907, + "step": 1728 + }, + { + "epoch": 0.22384050231414054, + "grad_norm": 1.47706601569829, + "learning_rate": 1.1925613932377112e-06, + "loss": 2.5127, + "step": 1729 + }, + { + "epoch": 0.223969964721494, + "grad_norm": 3.061316440173626, + "learning_rate": 1.1924430461252572e-06, + "loss": 3.126, + "step": 1730 + }, + { + "epoch": 0.223969964721494, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5637428760528564, + "eval_runtime": 15.4829, + "eval_samples_per_second": 2.842, + "eval_steps_per_second": 0.194, + "step": 1730 + }, + { + "epoch": 0.22409942712884745, + "grad_norm": 4.319830308518956, + "learning_rate": 1.1923245831010594e-06, + "loss": 3.5161, + "step": 1731 + }, + { + "epoch": 0.22422888953620093, + "grad_norm": 1.9709655852753052, + "learning_rate": 1.1922060041893163e-06, + "loss": 2.7952, + "step": 1732 + }, + { + "epoch": 0.22435835194355438, + "grad_norm": 2.240317459449038, + "learning_rate": 1.1920873094142504e-06, + "loss": 2.939, + "step": 1733 + }, + { + "epoch": 0.22448781435090787, + "grad_norm": 7.1995988693445545, + "learning_rate": 1.191968498800107e-06, + "loss": 3.9414, + "step": 1734 + }, + { + "epoch": 0.22461727675826132, + "grad_norm": 2.35580583883058, + "learning_rate": 1.1918495723711557e-06, + "loss": 2.8203, + "step": 1735 + }, + { + "epoch": 0.22461727675826132, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5696910619735718, + "eval_runtime": 14.6631, + "eval_samples_per_second": 3.001, + "eval_steps_per_second": 0.205, + "step": 1735 + }, + { + "epoch": 0.22474673916561477, + "grad_norm": 1.6119617724320778, + "learning_rate": 1.1917305301516898e-06, + "loss": 2.6233, + "step": 1736 + }, + { + "epoch": 0.22487620157296825, + "grad_norm": 2.5082210833575407, + "learning_rate": 1.191611372166026e-06, + "loss": 2.959, + "step": 1737 + }, + { + "epoch": 0.2250056639803217, + "grad_norm": 3.6889685883679566, + "learning_rate": 1.191492098438505e-06, + "loss": 3.4849, + "step": 1738 + }, + { + "epoch": 0.2251351263876752, + "grad_norm": 4.609415356121887, + "learning_rate": 1.1913727089934905e-06, + "loss": 3.9851, + "step": 1739 + }, + { + "epoch": 0.22526458879502864, + "grad_norm": 5.308911292675775, + "learning_rate": 1.19125320385537e-06, + "loss": 3.5415, + "step": 1740 + }, + { + "epoch": 0.22526458879502864, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5680485963821411, + "eval_runtime": 14.5759, + "eval_samples_per_second": 3.019, + "eval_steps_per_second": 0.206, + "step": 1740 + }, + { + "epoch": 0.2253940512023821, + "grad_norm": 5.842219639744819, + "learning_rate": 1.1911335830485556e-06, + "loss": 3.8286, + "step": 1741 + }, + { + "epoch": 0.22552351360973558, + "grad_norm": 2.5664585340326784, + "learning_rate": 1.1910138465974815e-06, + "loss": 3.1953, + "step": 1742 + }, + { + "epoch": 0.22565297601708903, + "grad_norm": 2.172513126349862, + "learning_rate": 1.190893994526607e-06, + "loss": 2.7937, + "step": 1743 + }, + { + "epoch": 0.2257824384244425, + "grad_norm": 4.066184345071473, + "learning_rate": 1.1907740268604139e-06, + "loss": 3.646, + "step": 1744 + }, + { + "epoch": 0.22591190083179596, + "grad_norm": 5.395992203250564, + "learning_rate": 1.190653943623408e-06, + "loss": 3.8916, + "step": 1745 + }, + { + "epoch": 0.22591190083179596, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5708895921707153, + "eval_runtime": 14.7459, + "eval_samples_per_second": 2.984, + "eval_steps_per_second": 0.203, + "step": 1745 + }, + { + "epoch": 0.22604136323914945, + "grad_norm": 3.8514381132523092, + "learning_rate": 1.190533744840119e-06, + "loss": 3.1475, + "step": 1746 + }, + { + "epoch": 0.2261708256465029, + "grad_norm": 5.996010176458452, + "learning_rate": 1.1904134305350996e-06, + "loss": 3.0664, + "step": 1747 + }, + { + "epoch": 0.22630028805385635, + "grad_norm": 3.8790684631487284, + "learning_rate": 1.1902930007329265e-06, + "loss": 2.8591, + "step": 1748 + }, + { + "epoch": 0.22642975046120983, + "grad_norm": 2.321835374732646, + "learning_rate": 1.1901724554582001e-06, + "loss": 2.7219, + "step": 1749 + }, + { + "epoch": 0.2265592128685633, + "grad_norm": 4.849322392776263, + "learning_rate": 1.1900517947355442e-06, + "loss": 3.2275, + "step": 1750 + }, + { + "epoch": 0.2265592128685633, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5711559057235718, + "eval_runtime": 14.454, + "eval_samples_per_second": 3.044, + "eval_steps_per_second": 0.208, + "step": 1750 + }, + { + "epoch": 0.22668867527591677, + "grad_norm": 4.739930524655867, + "learning_rate": 1.1899310185896062e-06, + "loss": 3.6909, + "step": 1751 + }, + { + "epoch": 0.22681813768327022, + "grad_norm": 2.5714961159532326, + "learning_rate": 1.1898101270450569e-06, + "loss": 2.7419, + "step": 1752 + }, + { + "epoch": 0.22694760009062367, + "grad_norm": 1.571708821529022, + "learning_rate": 1.1896891201265911e-06, + "loss": 2.6799, + "step": 1753 + }, + { + "epoch": 0.22707706249797716, + "grad_norm": 6.6975266751647835, + "learning_rate": 1.1895679978589266e-06, + "loss": 3.9404, + "step": 1754 + }, + { + "epoch": 0.2272065249053306, + "grad_norm": 1.8552666812490077, + "learning_rate": 1.1894467602668056e-06, + "loss": 2.9622, + "step": 1755 + }, + { + "epoch": 0.2272065249053306, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5763493776321411, + "eval_runtime": 15.8993, + "eval_samples_per_second": 2.767, + "eval_steps_per_second": 0.189, + "step": 1755 + }, + { + "epoch": 0.2273359873126841, + "grad_norm": 6.798012046951795, + "learning_rate": 1.1893254073749926e-06, + "loss": 4.6042, + "step": 1756 + }, + { + "epoch": 0.22746544972003754, + "grad_norm": 6.031227505324747, + "learning_rate": 1.1892039392082771e-06, + "loss": 3.5444, + "step": 1757 + }, + { + "epoch": 0.227594912127391, + "grad_norm": 1.9836317094131743, + "learning_rate": 1.189082355791471e-06, + "loss": 2.3232, + "step": 1758 + }, + { + "epoch": 0.22772437453474448, + "grad_norm": 4.378499974481291, + "learning_rate": 1.1889606571494103e-06, + "loss": 3.6782, + "step": 1759 + }, + { + "epoch": 0.22785383694209793, + "grad_norm": 5.157970620135607, + "learning_rate": 1.1888388433069545e-06, + "loss": 4.2222, + "step": 1760 + }, + { + "epoch": 0.22785383694209793, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5777255296707153, + "eval_runtime": 14.1102, + "eval_samples_per_second": 3.118, + "eval_steps_per_second": 0.213, + "step": 1760 + }, + { + "epoch": 0.2279832993494514, + "grad_norm": 3.2377565781452096, + "learning_rate": 1.1887169142889866e-06, + "loss": 3.1804, + "step": 1761 + }, + { + "epoch": 0.22811276175680487, + "grad_norm": 4.248476010929114, + "learning_rate": 1.1885948701204125e-06, + "loss": 3.6548, + "step": 1762 + }, + { + "epoch": 0.22824222416415832, + "grad_norm": 3.9093861535477257, + "learning_rate": 1.188472710826163e-06, + "loss": 3.5649, + "step": 1763 + }, + { + "epoch": 0.2283716865715118, + "grad_norm": 3.0200621707086452, + "learning_rate": 1.188350436431191e-06, + "loss": 3.3237, + "step": 1764 + }, + { + "epoch": 0.22850114897886525, + "grad_norm": 2.747628671198382, + "learning_rate": 1.188228046960474e-06, + "loss": 2.8564, + "step": 1765 + }, + { + "epoch": 0.22850114897886525, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5797230005264282, + "eval_runtime": 14.7421, + "eval_samples_per_second": 2.985, + "eval_steps_per_second": 0.203, + "step": 1765 + }, + { + "epoch": 0.22863061138621874, + "grad_norm": 1.822798965074975, + "learning_rate": 1.188105542439012e-06, + "loss": 2.7983, + "step": 1766 + }, + { + "epoch": 0.2287600737935722, + "grad_norm": 3.1483265131972358, + "learning_rate": 1.1879829228918295e-06, + "loss": 3.0876, + "step": 1767 + }, + { + "epoch": 0.22888953620092567, + "grad_norm": 2.2193564040078617, + "learning_rate": 1.1878601883439738e-06, + "loss": 2.9304, + "step": 1768 + }, + { + "epoch": 0.22901899860827912, + "grad_norm": 4.644730132949049, + "learning_rate": 1.1877373388205162e-06, + "loss": 4.0947, + "step": 1769 + }, + { + "epoch": 0.22914846101563258, + "grad_norm": 2.395911144377658, + "learning_rate": 1.1876143743465507e-06, + "loss": 2.9478, + "step": 1770 + }, + { + "epoch": 0.22914846101563258, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5846058130264282, + "eval_runtime": 14.7895, + "eval_samples_per_second": 2.975, + "eval_steps_per_second": 0.203, + "step": 1770 + }, + { + "epoch": 0.22927792342298606, + "grad_norm": 1.510404962451898, + "learning_rate": 1.1874912949471957e-06, + "loss": 2.5127, + "step": 1771 + }, + { + "epoch": 0.2294073858303395, + "grad_norm": 4.1358395535030725, + "learning_rate": 1.1873681006475924e-06, + "loss": 3.5691, + "step": 1772 + }, + { + "epoch": 0.229536848237693, + "grad_norm": 3.268384502322116, + "learning_rate": 1.187244791472906e-06, + "loss": 3.0142, + "step": 1773 + }, + { + "epoch": 0.22966631064504645, + "grad_norm": 6.264869445363574, + "learning_rate": 1.1871213674483247e-06, + "loss": 3.8196, + "step": 1774 + }, + { + "epoch": 0.2297957730523999, + "grad_norm": 3.7064875790383, + "learning_rate": 1.1869978285990604e-06, + "loss": 3.4233, + "step": 1775 + }, + { + "epoch": 0.2297957730523999, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5838290452957153, + "eval_runtime": 13.1211, + "eval_samples_per_second": 3.353, + "eval_steps_per_second": 0.229, + "step": 1775 + }, + { + "epoch": 0.22992523545975338, + "grad_norm": 2.0280575355571004, + "learning_rate": 1.1868741749503485e-06, + "loss": 2.8713, + "step": 1776 + }, + { + "epoch": 0.23005469786710683, + "grad_norm": 2.2507887943390634, + "learning_rate": 1.1867504065274477e-06, + "loss": 2.895, + "step": 1777 + }, + { + "epoch": 0.23018416027446031, + "grad_norm": 1.9347822544009718, + "learning_rate": 1.18662652335564e-06, + "loss": 2.5469, + "step": 1778 + }, + { + "epoch": 0.23031362268181377, + "grad_norm": 6.252203058900052, + "learning_rate": 1.1865025254602317e-06, + "loss": 3.9983, + "step": 1779 + }, + { + "epoch": 0.23044308508916722, + "grad_norm": 2.7984166102703436, + "learning_rate": 1.1863784128665512e-06, + "loss": 3.2495, + "step": 1780 + }, + { + "epoch": 0.23044308508916722, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5873135328292847, + "eval_runtime": 13.7, + "eval_samples_per_second": 3.212, + "eval_steps_per_second": 0.219, + "step": 1780 + }, + { + "epoch": 0.2305725474965207, + "grad_norm": 4.172889916817173, + "learning_rate": 1.186254185599951e-06, + "loss": 3.5293, + "step": 1781 + }, + { + "epoch": 0.23070200990387416, + "grad_norm": 3.0835522763063885, + "learning_rate": 1.1861298436858075e-06, + "loss": 3.3398, + "step": 1782 + }, + { + "epoch": 0.23083147231122764, + "grad_norm": 4.9575589410589584, + "learning_rate": 1.1860053871495199e-06, + "loss": 3.1064, + "step": 1783 + }, + { + "epoch": 0.2309609347185811, + "grad_norm": 3.6457331121335304, + "learning_rate": 1.1858808160165108e-06, + "loss": 2.9993, + "step": 1784 + }, + { + "epoch": 0.23109039712593454, + "grad_norm": 6.046506128296669, + "learning_rate": 1.1857561303122266e-06, + "loss": 3.314, + "step": 1785 + }, + { + "epoch": 0.23109039712593454, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5865589380264282, + "eval_runtime": 14.3035, + "eval_samples_per_second": 3.076, + "eval_steps_per_second": 0.21, + "step": 1785 + }, + { + "epoch": 0.23121985953328802, + "grad_norm": 2.636689754406309, + "learning_rate": 1.1856313300621367e-06, + "loss": 2.8096, + "step": 1786 + }, + { + "epoch": 0.23134932194064148, + "grad_norm": 5.578799757744635, + "learning_rate": 1.1855064152917342e-06, + "loss": 3.3169, + "step": 1787 + }, + { + "epoch": 0.23147878434799496, + "grad_norm": 1.889160245127263, + "learning_rate": 1.1853813860265354e-06, + "loss": 2.4883, + "step": 1788 + }, + { + "epoch": 0.2316082467553484, + "grad_norm": 3.2704907217190495, + "learning_rate": 1.18525624229208e-06, + "loss": 3.0698, + "step": 1789 + }, + { + "epoch": 0.2317377091627019, + "grad_norm": 4.624546217255355, + "learning_rate": 1.1851309841139313e-06, + "loss": 3.5776, + "step": 1790 + }, + { + "epoch": 0.2317377091627019, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.590043544769287, + "eval_runtime": 13.8172, + "eval_samples_per_second": 3.184, + "eval_steps_per_second": 0.217, + "step": 1790 + }, + { + "epoch": 0.23186717157005535, + "grad_norm": 1.5267695445893605, + "learning_rate": 1.1850056115176758e-06, + "loss": 2.6177, + "step": 1791 + }, + { + "epoch": 0.2319966339774088, + "grad_norm": 1.5428228815579181, + "learning_rate": 1.1848801245289236e-06, + "loss": 2.6587, + "step": 1792 + }, + { + "epoch": 0.23212609638476228, + "grad_norm": 2.320113895526707, + "learning_rate": 1.1847545231733071e-06, + "loss": 2.8708, + "step": 1793 + }, + { + "epoch": 0.23225555879211573, + "grad_norm": 2.205125100465697, + "learning_rate": 1.184628807476484e-06, + "loss": 3.1047, + "step": 1794 + }, + { + "epoch": 0.23238502119946922, + "grad_norm": 4.439712028758625, + "learning_rate": 1.1845029774641338e-06, + "loss": 3.6821, + "step": 1795 + }, + { + "epoch": 0.23238502119946922, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.592817783355713, + "eval_runtime": 15.1244, + "eval_samples_per_second": 2.909, + "eval_steps_per_second": 0.198, + "step": 1795 + }, + { + "epoch": 0.23251448360682267, + "grad_norm": 4.962734232554875, + "learning_rate": 1.1843770331619597e-06, + "loss": 3.2542, + "step": 1796 + }, + { + "epoch": 0.23264394601417612, + "grad_norm": 1.5757063984383173, + "learning_rate": 1.1842509745956885e-06, + "loss": 2.9131, + "step": 1797 + }, + { + "epoch": 0.2327734084215296, + "grad_norm": 1.5413241371035584, + "learning_rate": 1.1841248017910702e-06, + "loss": 2.6279, + "step": 1798 + }, + { + "epoch": 0.23290287082888306, + "grad_norm": 2.4034265576485203, + "learning_rate": 1.1839985147738784e-06, + "loss": 3.144, + "step": 1799 + }, + { + "epoch": 0.23303233323623654, + "grad_norm": 2.212079301669578, + "learning_rate": 1.1838721135699094e-06, + "loss": 2.6068, + "step": 1800 + }, + { + "epoch": 0.23303233323623654, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.5936390161514282, + "eval_runtime": 13.3757, + "eval_samples_per_second": 3.29, + "eval_steps_per_second": 0.224, + "step": 1800 + }, + { + "epoch": 0.23316179564359, + "grad_norm": 5.478674115184305, + "learning_rate": 1.1837455982049835e-06, + "loss": 3.1055, + "step": 1801 + }, + { + "epoch": 0.23329125805094345, + "grad_norm": 4.347527707245364, + "learning_rate": 1.183618968704944e-06, + "loss": 3.356, + "step": 1802 + }, + { + "epoch": 0.23342072045829693, + "grad_norm": 2.2850104345524347, + "learning_rate": 1.1834922250956572e-06, + "loss": 3.0022, + "step": 1803 + }, + { + "epoch": 0.23355018286565038, + "grad_norm": 4.075156425449902, + "learning_rate": 1.1833653674030134e-06, + "loss": 3.6924, + "step": 1804 + }, + { + "epoch": 0.23367964527300386, + "grad_norm": 1.7377919385412286, + "learning_rate": 1.1832383956529258e-06, + "loss": 2.7505, + "step": 1805 + }, + { + "epoch": 0.23367964527300386, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.589799404144287, + "eval_runtime": 15.0026, + "eval_samples_per_second": 2.933, + "eval_steps_per_second": 0.2, + "step": 1805 + }, + { + "epoch": 0.23380910768035731, + "grad_norm": 2.6073283630427344, + "learning_rate": 1.1831113098713307e-06, + "loss": 2.9761, + "step": 1806 + }, + { + "epoch": 0.23393857008771077, + "grad_norm": 3.614322979767949, + "learning_rate": 1.1829841100841884e-06, + "loss": 3.074, + "step": 1807 + }, + { + "epoch": 0.23406803249506425, + "grad_norm": 4.5033987054023115, + "learning_rate": 1.1828567963174816e-06, + "loss": 3.261, + "step": 1808 + }, + { + "epoch": 0.2341974949024177, + "grad_norm": 2.2737850508785233, + "learning_rate": 1.1827293685972172e-06, + "loss": 2.6584, + "step": 1809 + }, + { + "epoch": 0.23432695730977118, + "grad_norm": 2.8210947074723105, + "learning_rate": 1.1826018269494244e-06, + "loss": 2.7859, + "step": 1810 + }, + { + "epoch": 0.23432695730977118, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.595458984375, + "eval_runtime": 14.9967, + "eval_samples_per_second": 2.934, + "eval_steps_per_second": 0.2, + "step": 1810 + }, + { + "epoch": 0.23445641971712464, + "grad_norm": 3.602167179229879, + "learning_rate": 1.1824741714001565e-06, + "loss": 3.3247, + "step": 1811 + }, + { + "epoch": 0.23458588212447812, + "grad_norm": 1.9449066011243887, + "learning_rate": 1.1823464019754895e-06, + "loss": 2.8586, + "step": 1812 + }, + { + "epoch": 0.23471534453183157, + "grad_norm": 4.176242688853541, + "learning_rate": 1.182218518701523e-06, + "loss": 3.2878, + "step": 1813 + }, + { + "epoch": 0.23484480693918502, + "grad_norm": 5.628199294341703, + "learning_rate": 1.1820905216043797e-06, + "loss": 3.6523, + "step": 1814 + }, + { + "epoch": 0.2349742693465385, + "grad_norm": 3.8074938777871594, + "learning_rate": 1.1819624107102057e-06, + "loss": 3.4065, + "step": 1815 + }, + { + "epoch": 0.2349742693465385, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.594682216644287, + "eval_runtime": 14.363, + "eval_samples_per_second": 3.063, + "eval_steps_per_second": 0.209, + "step": 1815 + }, + { + "epoch": 0.23510373175389196, + "grad_norm": 1.5983667295130326, + "learning_rate": 1.1818341860451703e-06, + "loss": 2.5, + "step": 1816 + }, + { + "epoch": 0.23523319416124544, + "grad_norm": 5.5634545235779065, + "learning_rate": 1.1817058476354657e-06, + "loss": 3.5693, + "step": 1817 + }, + { + "epoch": 0.2353626565685989, + "grad_norm": 4.20888366597106, + "learning_rate": 1.181577395507308e-06, + "loss": 3.478, + "step": 1818 + }, + { + "epoch": 0.23549211897595235, + "grad_norm": 5.3110630760512585, + "learning_rate": 1.181448829686936e-06, + "loss": 3.5986, + "step": 1819 + }, + { + "epoch": 0.23562158138330583, + "grad_norm": 6.774465091539316, + "learning_rate": 1.1813201502006116e-06, + "loss": 3.5774, + "step": 1820 + }, + { + "epoch": 0.23562158138330583, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.596235752105713, + "eval_runtime": 13.9516, + "eval_samples_per_second": 3.154, + "eval_steps_per_second": 0.215, + "step": 1820 + }, + { + "epoch": 0.23575104379065928, + "grad_norm": 2.056453228202876, + "learning_rate": 1.1811913570746205e-06, + "loss": 2.6279, + "step": 1821 + }, + { + "epoch": 0.23588050619801276, + "grad_norm": 3.2525689765267507, + "learning_rate": 1.1810624503352713e-06, + "loss": 3.0186, + "step": 1822 + }, + { + "epoch": 0.23600996860536622, + "grad_norm": 8.605539012752132, + "learning_rate": 1.1809334300088957e-06, + "loss": 4.3545, + "step": 1823 + }, + { + "epoch": 0.23613943101271967, + "grad_norm": 2.9217672232627896, + "learning_rate": 1.1808042961218488e-06, + "loss": 3.0693, + "step": 1824 + }, + { + "epoch": 0.23626889342007315, + "grad_norm": 5.821165686943768, + "learning_rate": 1.1806750487005086e-06, + "loss": 4.2314, + "step": 1825 + }, + { + "epoch": 0.23626889342007315, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.599165439605713, + "eval_runtime": 14.6408, + "eval_samples_per_second": 3.005, + "eval_steps_per_second": 0.205, + "step": 1825 + }, + { + "epoch": 0.2363983558274266, + "grad_norm": 4.39105876666601, + "learning_rate": 1.1805456877712769e-06, + "loss": 3.46, + "step": 1826 + }, + { + "epoch": 0.23652781823478009, + "grad_norm": 1.982333220196776, + "learning_rate": 1.1804162133605781e-06, + "loss": 2.709, + "step": 1827 + }, + { + "epoch": 0.23665728064213354, + "grad_norm": 1.5761223394180242, + "learning_rate": 1.1802866254948598e-06, + "loss": 2.4607, + "step": 1828 + }, + { + "epoch": 0.236786743049487, + "grad_norm": 5.235561279276746, + "learning_rate": 1.180156924200593e-06, + "loss": 3.4575, + "step": 1829 + }, + { + "epoch": 0.23691620545684047, + "grad_norm": 4.6596904631208815, + "learning_rate": 1.180027109504272e-06, + "loss": 4.0371, + "step": 1830 + }, + { + "epoch": 0.23691620545684047, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.601606845855713, + "eval_runtime": 14.3148, + "eval_samples_per_second": 3.074, + "eval_steps_per_second": 0.21, + "step": 1830 + }, + { + "epoch": 0.23704566786419393, + "grad_norm": 2.825911278158945, + "learning_rate": 1.1798971814324143e-06, + "loss": 2.7859, + "step": 1831 + }, + { + "epoch": 0.2371751302715474, + "grad_norm": 2.8793604036484863, + "learning_rate": 1.1797671400115596e-06, + "loss": 3.0991, + "step": 1832 + }, + { + "epoch": 0.23730459267890086, + "grad_norm": 2.311542616946839, + "learning_rate": 1.1796369852682722e-06, + "loss": 3.1733, + "step": 1833 + }, + { + "epoch": 0.23743405508625434, + "grad_norm": 1.6952525313065852, + "learning_rate": 1.1795067172291385e-06, + "loss": 2.6223, + "step": 1834 + }, + { + "epoch": 0.2375635174936078, + "grad_norm": 2.5247398015838938, + "learning_rate": 1.1793763359207686e-06, + "loss": 2.9302, + "step": 1835 + }, + { + "epoch": 0.2375635174936078, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.605912685394287, + "eval_runtime": 14.732, + "eval_samples_per_second": 2.987, + "eval_steps_per_second": 0.204, + "step": 1835 + }, + { + "epoch": 0.23769297990096125, + "grad_norm": 1.9933049448504896, + "learning_rate": 1.1792458413697951e-06, + "loss": 2.7476, + "step": 1836 + }, + { + "epoch": 0.23782244230831473, + "grad_norm": 4.7162128598490165, + "learning_rate": 1.1791152336028746e-06, + "loss": 3.4561, + "step": 1837 + }, + { + "epoch": 0.23795190471566818, + "grad_norm": 1.943023123705872, + "learning_rate": 1.1789845126466864e-06, + "loss": 2.7314, + "step": 1838 + }, + { + "epoch": 0.23808136712302166, + "grad_norm": 2.9772015630697406, + "learning_rate": 1.1788536785279325e-06, + "loss": 3.1687, + "step": 1839 + }, + { + "epoch": 0.23821082953037512, + "grad_norm": 3.320013254275291, + "learning_rate": 1.1787227312733388e-06, + "loss": 3.2031, + "step": 1840 + }, + { + "epoch": 0.23821082953037512, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6009188890457153, + "eval_runtime": 14.8977, + "eval_samples_per_second": 2.953, + "eval_steps_per_second": 0.201, + "step": 1840 + }, + { + "epoch": 0.23834029193772857, + "grad_norm": 3.8325170390730525, + "learning_rate": 1.1785916709096538e-06, + "loss": 3.0173, + "step": 1841 + }, + { + "epoch": 0.23846975434508205, + "grad_norm": 2.0805028534732237, + "learning_rate": 1.178460497463649e-06, + "loss": 2.8025, + "step": 1842 + }, + { + "epoch": 0.2385992167524355, + "grad_norm": 4.461425848651309, + "learning_rate": 1.1783292109621196e-06, + "loss": 3.2861, + "step": 1843 + }, + { + "epoch": 0.238728679159789, + "grad_norm": 2.1575720780987244, + "learning_rate": 1.1781978114318835e-06, + "loss": 2.9761, + "step": 1844 + }, + { + "epoch": 0.23885814156714244, + "grad_norm": 4.289045229317149, + "learning_rate": 1.1780662988997814e-06, + "loss": 3.1362, + "step": 1845 + }, + { + "epoch": 0.23885814156714244, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6029163599014282, + "eval_runtime": 14.0891, + "eval_samples_per_second": 3.123, + "eval_steps_per_second": 0.213, + "step": 1845 + }, + { + "epoch": 0.2389876039744959, + "grad_norm": 2.2589644223106284, + "learning_rate": 1.1779346733926778e-06, + "loss": 2.7209, + "step": 1846 + }, + { + "epoch": 0.23911706638184937, + "grad_norm": 3.9750453551208715, + "learning_rate": 1.1778029349374597e-06, + "loss": 3.3169, + "step": 1847 + }, + { + "epoch": 0.23924652878920283, + "grad_norm": 1.7815804958225263, + "learning_rate": 1.1776710835610371e-06, + "loss": 2.8086, + "step": 1848 + }, + { + "epoch": 0.2393759911965563, + "grad_norm": 5.351839327909766, + "learning_rate": 1.1775391192903437e-06, + "loss": 3.6738, + "step": 1849 + }, + { + "epoch": 0.23950545360390976, + "grad_norm": 3.0449422128513954, + "learning_rate": 1.1774070421523358e-06, + "loss": 3.0908, + "step": 1850 + }, + { + "epoch": 0.23950545360390976, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6037375926971436, + "eval_runtime": 13.198, + "eval_samples_per_second": 3.334, + "eval_steps_per_second": 0.227, + "step": 1850 + }, + { + "epoch": 0.23963491601126322, + "grad_norm": 2.2416079880456437, + "learning_rate": 1.1772748521739926e-06, + "loss": 3.0264, + "step": 1851 + }, + { + "epoch": 0.2397643784186167, + "grad_norm": 3.8526460107748517, + "learning_rate": 1.1771425493823168e-06, + "loss": 3.0195, + "step": 1852 + }, + { + "epoch": 0.23989384082597015, + "grad_norm": 4.330167610957246, + "learning_rate": 1.177010133804334e-06, + "loss": 3.4597, + "step": 1853 + }, + { + "epoch": 0.24002330323332363, + "grad_norm": 3.702924740714124, + "learning_rate": 1.1768776054670924e-06, + "loss": 2.8242, + "step": 1854 + }, + { + "epoch": 0.24015276564067708, + "grad_norm": 2.86621389616258, + "learning_rate": 1.1767449643976641e-06, + "loss": 3.1323, + "step": 1855 + }, + { + "epoch": 0.24015276564067708, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6049138307571411, + "eval_runtime": 14.3038, + "eval_samples_per_second": 3.076, + "eval_steps_per_second": 0.21, + "step": 1855 + }, + { + "epoch": 0.24028222804803057, + "grad_norm": 6.8781213442945, + "learning_rate": 1.1766122106231432e-06, + "loss": 3.3423, + "step": 1856 + }, + { + "epoch": 0.24041169045538402, + "grad_norm": 4.969605794255556, + "learning_rate": 1.1764793441706478e-06, + "loss": 3.6226, + "step": 1857 + }, + { + "epoch": 0.24054115286273747, + "grad_norm": 1.6546195148470562, + "learning_rate": 1.1763463650673182e-06, + "loss": 2.7083, + "step": 1858 + }, + { + "epoch": 0.24067061527009095, + "grad_norm": 6.188724141215013, + "learning_rate": 1.1762132733403184e-06, + "loss": 3.4285, + "step": 1859 + }, + { + "epoch": 0.2408000776774444, + "grad_norm": 4.707623508261219, + "learning_rate": 1.1760800690168349e-06, + "loss": 3.4709, + "step": 1860 + }, + { + "epoch": 0.2408000776774444, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6134144067764282, + "eval_runtime": 14.6448, + "eval_samples_per_second": 3.004, + "eval_steps_per_second": 0.205, + "step": 1860 + }, + { + "epoch": 0.2409295400847979, + "grad_norm": 4.66981496391424, + "learning_rate": 1.1759467521240774e-06, + "loss": 3.5918, + "step": 1861 + }, + { + "epoch": 0.24105900249215134, + "grad_norm": 2.818593930693954, + "learning_rate": 1.1758133226892786e-06, + "loss": 3.3066, + "step": 1862 + }, + { + "epoch": 0.2411884648995048, + "grad_norm": 3.3506448891288656, + "learning_rate": 1.1756797807396941e-06, + "loss": 3.282, + "step": 1863 + }, + { + "epoch": 0.24131792730685828, + "grad_norm": 3.9180330696979957, + "learning_rate": 1.1755461263026028e-06, + "loss": 3.4312, + "step": 1864 + }, + { + "epoch": 0.24144738971421173, + "grad_norm": 2.7474241675121505, + "learning_rate": 1.175412359405306e-06, + "loss": 3.1055, + "step": 1865 + }, + { + "epoch": 0.24144738971421173, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6133478879928589, + "eval_runtime": 14.4666, + "eval_samples_per_second": 3.041, + "eval_steps_per_second": 0.207, + "step": 1865 + }, + { + "epoch": 0.2415768521215652, + "grad_norm": 1.6426711111516288, + "learning_rate": 1.1752784800751285e-06, + "loss": 2.4155, + "step": 1866 + }, + { + "epoch": 0.24170631452891866, + "grad_norm": 1.343580295689865, + "learning_rate": 1.1751444883394177e-06, + "loss": 2.5276, + "step": 1867 + }, + { + "epoch": 0.24183577693627212, + "grad_norm": 2.6167222392001936, + "learning_rate": 1.1750103842255443e-06, + "loss": 2.6816, + "step": 1868 + }, + { + "epoch": 0.2419652393436256, + "grad_norm": 2.1113031892678777, + "learning_rate": 1.1748761677609018e-06, + "loss": 2.7498, + "step": 1869 + }, + { + "epoch": 0.24209470175097905, + "grad_norm": 2.3298636724049393, + "learning_rate": 1.1747418389729066e-06, + "loss": 3.0835, + "step": 1870 + }, + { + "epoch": 0.24209470175097905, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6146794557571411, + "eval_runtime": 14.9518, + "eval_samples_per_second": 2.943, + "eval_steps_per_second": 0.201, + "step": 1870 + }, + { + "epoch": 0.24222416415833253, + "grad_norm": 3.7112023411479504, + "learning_rate": 1.1746073978889977e-06, + "loss": 3.1948, + "step": 1871 + }, + { + "epoch": 0.242353626565686, + "grad_norm": 5.937930557908351, + "learning_rate": 1.174472844536638e-06, + "loss": 3.7815, + "step": 1872 + }, + { + "epoch": 0.24248308897303944, + "grad_norm": 4.696990897913837, + "learning_rate": 1.1743381789433126e-06, + "loss": 3.0894, + "step": 1873 + }, + { + "epoch": 0.24261255138039292, + "grad_norm": 4.737390688559019, + "learning_rate": 1.1742034011365293e-06, + "loss": 3.3867, + "step": 1874 + }, + { + "epoch": 0.24274201378774637, + "grad_norm": 2.745308506758466, + "learning_rate": 1.1740685111438196e-06, + "loss": 3.1606, + "step": 1875 + }, + { + "epoch": 0.24274201378774637, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.615922451019287, + "eval_runtime": 14.3167, + "eval_samples_per_second": 3.073, + "eval_steps_per_second": 0.21, + "step": 1875 + }, + { + "epoch": 0.24287147619509986, + "grad_norm": 6.549932183153616, + "learning_rate": 1.1739335089927373e-06, + "loss": 3.4863, + "step": 1876 + }, + { + "epoch": 0.2430009386024533, + "grad_norm": 3.359160123814084, + "learning_rate": 1.1737983947108595e-06, + "loss": 2.9927, + "step": 1877 + }, + { + "epoch": 0.2431304010098068, + "grad_norm": 3.4392212569066025, + "learning_rate": 1.173663168325786e-06, + "loss": 3.3003, + "step": 1878 + }, + { + "epoch": 0.24325986341716024, + "grad_norm": 4.252401229416216, + "learning_rate": 1.1735278298651395e-06, + "loss": 3.1992, + "step": 1879 + }, + { + "epoch": 0.2433893258245137, + "grad_norm": 3.3023280095272245, + "learning_rate": 1.1733923793565656e-06, + "loss": 3.0347, + "step": 1880 + }, + { + "epoch": 0.2433893258245137, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.620893955230713, + "eval_runtime": 14.2206, + "eval_samples_per_second": 3.094, + "eval_steps_per_second": 0.211, + "step": 1880 + }, + { + "epoch": 0.24351878823186718, + "grad_norm": 4.540014857907009, + "learning_rate": 1.1732568168277326e-06, + "loss": 3.3464, + "step": 1881 + }, + { + "epoch": 0.24364825063922063, + "grad_norm": 3.251888789683562, + "learning_rate": 1.1731211423063323e-06, + "loss": 2.8691, + "step": 1882 + }, + { + "epoch": 0.2437777130465741, + "grad_norm": 4.586408463329068, + "learning_rate": 1.1729853558200786e-06, + "loss": 3.3435, + "step": 1883 + }, + { + "epoch": 0.24390717545392757, + "grad_norm": 3.2370875789770337, + "learning_rate": 1.172849457396709e-06, + "loss": 2.8208, + "step": 1884 + }, + { + "epoch": 0.24403663786128102, + "grad_norm": 3.095421010617364, + "learning_rate": 1.1727134470639832e-06, + "loss": 3.1907, + "step": 1885 + }, + { + "epoch": 0.24403663786128102, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6202059984207153, + "eval_runtime": 14.3363, + "eval_samples_per_second": 3.069, + "eval_steps_per_second": 0.209, + "step": 1885 + }, + { + "epoch": 0.2441661002686345, + "grad_norm": 3.1517139071583298, + "learning_rate": 1.172577324849684e-06, + "loss": 2.7622, + "step": 1886 + }, + { + "epoch": 0.24429556267598795, + "grad_norm": 7.108209574349753, + "learning_rate": 1.1724410907816178e-06, + "loss": 3.8677, + "step": 1887 + }, + { + "epoch": 0.24442502508334144, + "grad_norm": 2.061805249129852, + "learning_rate": 1.1723047448876124e-06, + "loss": 2.5483, + "step": 1888 + }, + { + "epoch": 0.2445544874906949, + "grad_norm": 4.517874299808877, + "learning_rate": 1.1721682871955196e-06, + "loss": 2.8652, + "step": 1889 + }, + { + "epoch": 0.24468394989804834, + "grad_norm": 3.952170357823975, + "learning_rate": 1.1720317177332132e-06, + "loss": 3.1326, + "step": 1890 + }, + { + "epoch": 0.24468394989804834, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.62841796875, + "eval_runtime": 14.544, + "eval_samples_per_second": 3.025, + "eval_steps_per_second": 0.206, + "step": 1890 + }, + { + "epoch": 0.24481341230540182, + "grad_norm": 3.8008526617129292, + "learning_rate": 1.171895036528591e-06, + "loss": 3.4922, + "step": 1891 + }, + { + "epoch": 0.24494287471275528, + "grad_norm": 2.225180960314294, + "learning_rate": 1.1717582436095725e-06, + "loss": 2.6582, + "step": 1892 + }, + { + "epoch": 0.24507233712010876, + "grad_norm": 2.322052374551846, + "learning_rate": 1.1716213390041004e-06, + "loss": 3.0005, + "step": 1893 + }, + { + "epoch": 0.2452017995274622, + "grad_norm": 4.145224618409552, + "learning_rate": 1.1714843227401404e-06, + "loss": 3.8025, + "step": 1894 + }, + { + "epoch": 0.24533126193481566, + "grad_norm": 2.659412652836115, + "learning_rate": 1.1713471948456804e-06, + "loss": 2.9658, + "step": 1895 + }, + { + "epoch": 0.24533126193481566, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6312366724014282, + "eval_runtime": 15.3092, + "eval_samples_per_second": 2.874, + "eval_steps_per_second": 0.196, + "step": 1895 + }, + { + "epoch": 0.24546072434216915, + "grad_norm": 3.10138403182918, + "learning_rate": 1.1712099553487323e-06, + "loss": 2.8525, + "step": 1896 + }, + { + "epoch": 0.2455901867495226, + "grad_norm": 3.3659064883380463, + "learning_rate": 1.1710726042773294e-06, + "loss": 2.6694, + "step": 1897 + }, + { + "epoch": 0.24571964915687608, + "grad_norm": 6.495870574274998, + "learning_rate": 1.1709351416595288e-06, + "loss": 3.8931, + "step": 1898 + }, + { + "epoch": 0.24584911156422953, + "grad_norm": 3.898914673426884, + "learning_rate": 1.1707975675234099e-06, + "loss": 2.75, + "step": 1899 + }, + { + "epoch": 0.24597857397158301, + "grad_norm": 5.218508613864783, + "learning_rate": 1.170659881897075e-06, + "loss": 3.2034, + "step": 1900 + }, + { + "epoch": 0.24597857397158301, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.634721279144287, + "eval_runtime": 14.7638, + "eval_samples_per_second": 2.98, + "eval_steps_per_second": 0.203, + "step": 1900 + }, + { + "epoch": 0.24610803637893647, + "grad_norm": 3.549443092089115, + "learning_rate": 1.1705220848086491e-06, + "loss": 3.0303, + "step": 1901 + }, + { + "epoch": 0.24623749878628992, + "grad_norm": 5.924032827398414, + "learning_rate": 1.1703841762862802e-06, + "loss": 3.3115, + "step": 1902 + }, + { + "epoch": 0.2463669611936434, + "grad_norm": 4.181701413823407, + "learning_rate": 1.170246156358139e-06, + "loss": 3.2271, + "step": 1903 + }, + { + "epoch": 0.24649642360099686, + "grad_norm": 2.3167449836917435, + "learning_rate": 1.1701080250524187e-06, + "loss": 2.7825, + "step": 1904 + }, + { + "epoch": 0.24662588600835034, + "grad_norm": 1.5241933551793072, + "learning_rate": 1.1699697823973354e-06, + "loss": 2.3547, + "step": 1905 + }, + { + "epoch": 0.24662588600835034, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6404030323028564, + "eval_runtime": 15.0272, + "eval_samples_per_second": 2.928, + "eval_steps_per_second": 0.2, + "step": 1905 + }, + { + "epoch": 0.2467553484157038, + "grad_norm": 5.322937685840939, + "learning_rate": 1.169831428421128e-06, + "loss": 3.0752, + "step": 1906 + }, + { + "epoch": 0.24688481082305724, + "grad_norm": 5.759678630636752, + "learning_rate": 1.169692963152058e-06, + "loss": 3.2339, + "step": 1907 + }, + { + "epoch": 0.24701427323041072, + "grad_norm": 3.206134709658607, + "learning_rate": 1.16955438661841e-06, + "loss": 2.9053, + "step": 1908 + }, + { + "epoch": 0.24714373563776418, + "grad_norm": 3.530220628237768, + "learning_rate": 1.1694156988484909e-06, + "loss": 3.4482, + "step": 1909 + }, + { + "epoch": 0.24727319804511766, + "grad_norm": 3.483503154963089, + "learning_rate": 1.1692768998706306e-06, + "loss": 3.0337, + "step": 1910 + }, + { + "epoch": 0.24727319804511766, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6439098119735718, + "eval_runtime": 14.2668, + "eval_samples_per_second": 3.084, + "eval_steps_per_second": 0.21, + "step": 1910 + }, + { + "epoch": 0.2474026604524711, + "grad_norm": 4.352787043277252, + "learning_rate": 1.1691379897131814e-06, + "loss": 3.0293, + "step": 1911 + }, + { + "epoch": 0.24753212285982457, + "grad_norm": 2.3934729366186818, + "learning_rate": 1.1689989684045186e-06, + "loss": 2.9399, + "step": 1912 + }, + { + "epoch": 0.24766158526717805, + "grad_norm": 2.2565153309055437, + "learning_rate": 1.1688598359730401e-06, + "loss": 3.0776, + "step": 1913 + }, + { + "epoch": 0.2477910476745315, + "grad_norm": 2.215045688960175, + "learning_rate": 1.1687205924471668e-06, + "loss": 2.6819, + "step": 1914 + }, + { + "epoch": 0.24792051008188498, + "grad_norm": 2.571773481207664, + "learning_rate": 1.1685812378553416e-06, + "loss": 2.9763, + "step": 1915 + }, + { + "epoch": 0.24792051008188498, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.645751953125, + "eval_runtime": 13.988, + "eval_samples_per_second": 3.146, + "eval_steps_per_second": 0.214, + "step": 1915 + }, + { + "epoch": 0.24804997248923844, + "grad_norm": 4.5284448031533815, + "learning_rate": 1.1684417722260306e-06, + "loss": 2.9907, + "step": 1916 + }, + { + "epoch": 0.2481794348965919, + "grad_norm": 4.239696600057911, + "learning_rate": 1.1683021955877226e-06, + "loss": 3.3342, + "step": 1917 + }, + { + "epoch": 0.24830889730394537, + "grad_norm": 1.4685864204533403, + "learning_rate": 1.168162507968929e-06, + "loss": 2.498, + "step": 1918 + }, + { + "epoch": 0.24843835971129882, + "grad_norm": 3.123063655653541, + "learning_rate": 1.1680227093981838e-06, + "loss": 3.2268, + "step": 1919 + }, + { + "epoch": 0.2485678221186523, + "grad_norm": 4.049662140273082, + "learning_rate": 1.1678827999040434e-06, + "loss": 3.3848, + "step": 1920 + }, + { + "epoch": 0.2485678221186523, + "eval_PRM Accuracy": 0.2358490566037736, + "eval_PRM F1": 0.047058823529411764, + "eval_PRM F1 AUC": 0.5120481927710844, + "eval_PRM F1 Neg": 0.36220472440944884, + "eval_PRM NPV": 0.22115384615384615, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.024096385542168676, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6492587327957153, + "eval_runtime": 13.6457, + "eval_samples_per_second": 3.224, + "eval_steps_per_second": 0.22, + "step": 1920 + }, + { + "epoch": 0.24869728452600576, + "grad_norm": 3.1032801751811925, + "learning_rate": 1.1677427795150874e-06, + "loss": 2.9585, + "step": 1921 + }, + { + "epoch": 0.24882674693335924, + "grad_norm": 5.033863510907717, + "learning_rate": 1.1676026482599181e-06, + "loss": 3.1479, + "step": 1922 + }, + { + "epoch": 0.2489562093407127, + "grad_norm": 2.329811089621362, + "learning_rate": 1.1674624061671597e-06, + "loss": 3.0049, + "step": 1923 + }, + { + "epoch": 0.24908567174806615, + "grad_norm": 3.6429543909690683, + "learning_rate": 1.1673220532654596e-06, + "loss": 3.3096, + "step": 1924 + }, + { + "epoch": 0.24921513415541963, + "grad_norm": 3.425761242416327, + "learning_rate": 1.1671815895834878e-06, + "loss": 2.9817, + "step": 1925 + }, + { + "epoch": 0.24921513415541963, + "eval_PRM Accuracy": 0.2358490566037736, + "eval_PRM F1": 0.047058823529411764, + "eval_PRM F1 AUC": 0.5120481927710844, + "eval_PRM F1 Neg": 0.36220472440944884, + "eval_PRM NPV": 0.22115384615384615, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.024096385542168676, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.64990234375, + "eval_runtime": 13.9752, + "eval_samples_per_second": 3.148, + "eval_steps_per_second": 0.215, + "step": 1925 + }, + { + "epoch": 0.24934459656277308, + "grad_norm": 2.3724580370443267, + "learning_rate": 1.1670410151499367e-06, + "loss": 2.9553, + "step": 1926 + }, + { + "epoch": 0.24947405897012656, + "grad_norm": 3.7303337256752234, + "learning_rate": 1.1669003299935216e-06, + "loss": 2.9834, + "step": 1927 + }, + { + "epoch": 0.24960352137748001, + "grad_norm": 1.8399631563706738, + "learning_rate": 1.1667595341429804e-06, + "loss": 2.5762, + "step": 1928 + }, + { + "epoch": 0.24973298378483347, + "grad_norm": 2.4346348505121376, + "learning_rate": 1.1666186276270732e-06, + "loss": 2.8062, + "step": 1929 + }, + { + "epoch": 0.24986244619218695, + "grad_norm": 3.129114112134674, + "learning_rate": 1.1664776104745833e-06, + "loss": 2.9463, + "step": 1930 + }, + { + "epoch": 0.24986244619218695, + "eval_PRM Accuracy": 0.2358490566037736, + "eval_PRM F1": 0.047058823529411764, + "eval_PRM F1 AUC": 0.5120481927710844, + "eval_PRM F1 Neg": 0.36220472440944884, + "eval_PRM NPV": 0.22115384615384615, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.024096385542168676, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.654296875, + "eval_runtime": 14.3271, + "eval_samples_per_second": 3.071, + "eval_steps_per_second": 0.209, + "step": 1930 + }, + { + "epoch": 0.2499919085995404, + "grad_norm": 4.006686316110688, + "learning_rate": 1.1663364827143164e-06, + "loss": 3.2124, + "step": 1931 + }, + { + "epoch": 0.25012137100689386, + "grad_norm": 2.148758338978397, + "learning_rate": 1.1661952443751003e-06, + "loss": 3.1199, + "step": 1932 + }, + { + "epoch": 0.25025083341424736, + "grad_norm": 6.715887482720353, + "learning_rate": 1.166053895485786e-06, + "loss": 3.7588, + "step": 1933 + }, + { + "epoch": 0.2503802958216008, + "grad_norm": 1.7416847813457657, + "learning_rate": 1.165912436075247e-06, + "loss": 2.707, + "step": 1934 + }, + { + "epoch": 0.25050975822895427, + "grad_norm": 2.9003109139957655, + "learning_rate": 1.165770866172379e-06, + "loss": 2.9849, + "step": 1935 + }, + { + "epoch": 0.25050975822895427, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6504572629928589, + "eval_runtime": 14.3051, + "eval_samples_per_second": 3.076, + "eval_steps_per_second": 0.21, + "step": 1935 + }, + { + "epoch": 0.2506392206363077, + "grad_norm": 4.424929347402928, + "learning_rate": 1.1656291858061007e-06, + "loss": 3.0566, + "step": 1936 + }, + { + "epoch": 0.2507686830436612, + "grad_norm": 3.890277300436681, + "learning_rate": 1.165487395005353e-06, + "loss": 3.7363, + "step": 1937 + }, + { + "epoch": 0.2508981454510147, + "grad_norm": 4.396164022571692, + "learning_rate": 1.1653454937990998e-06, + "loss": 3.1294, + "step": 1938 + }, + { + "epoch": 0.25102760785836814, + "grad_norm": 2.7710371824078117, + "learning_rate": 1.1652034822163274e-06, + "loss": 2.6621, + "step": 1939 + }, + { + "epoch": 0.2511570702657216, + "grad_norm": 4.940295711783432, + "learning_rate": 1.1650613602860442e-06, + "loss": 3.7119, + "step": 1940 + }, + { + "epoch": 0.2511570702657216, + "eval_PRM Accuracy": 0.2358490566037736, + "eval_PRM F1": 0.047058823529411764, + "eval_PRM F1 AUC": 0.5120481927710844, + "eval_PRM F1 Neg": 0.36220472440944884, + "eval_PRM NPV": 0.22115384615384615, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.024096385542168676, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6569157838821411, + "eval_runtime": 14.436, + "eval_samples_per_second": 3.048, + "eval_steps_per_second": 0.208, + "step": 1940 + }, + { + "epoch": 0.25128653267307505, + "grad_norm": 5.206647762855925, + "learning_rate": 1.1649191280372815e-06, + "loss": 2.8794, + "step": 1941 + }, + { + "epoch": 0.2514159950804285, + "grad_norm": 3.819919695496243, + "learning_rate": 1.1647767854990934e-06, + "loss": 3.0488, + "step": 1942 + }, + { + "epoch": 0.251545457487782, + "grad_norm": 6.460455605701821, + "learning_rate": 1.164634332700556e-06, + "loss": 3.4609, + "step": 1943 + }, + { + "epoch": 0.25167491989513546, + "grad_norm": 2.791025876629594, + "learning_rate": 1.1644917696707683e-06, + "loss": 2.5996, + "step": 1944 + }, + { + "epoch": 0.2518043823024889, + "grad_norm": 5.080405747007676, + "learning_rate": 1.1643490964388516e-06, + "loss": 2.8545, + "step": 1945 + }, + { + "epoch": 0.2518043823024889, + "eval_PRM Accuracy": 0.24528301886792453, + "eval_PRM F1": 0.06976744186046512, + "eval_PRM F1 AUC": 0.5180722891566265, + "eval_PRM F1 Neg": 0.36507936507936506, + "eval_PRM NPV": 0.22330097087378642, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.03614457831325301, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6588023900985718, + "eval_runtime": 14.35, + "eval_samples_per_second": 3.066, + "eval_steps_per_second": 0.209, + "step": 1945 + }, + { + "epoch": 0.25193384470984237, + "grad_norm": 4.32648225151238, + "learning_rate": 1.1642063130339499e-06, + "loss": 3.1641, + "step": 1946 + }, + { + "epoch": 0.2520633071171958, + "grad_norm": 8.471658791981463, + "learning_rate": 1.1640634194852296e-06, + "loss": 4.6089, + "step": 1947 + }, + { + "epoch": 0.25219276952454933, + "grad_norm": 2.5457007924993973, + "learning_rate": 1.1639204158218793e-06, + "loss": 3.0625, + "step": 1948 + }, + { + "epoch": 0.2523222319319028, + "grad_norm": 3.822627469940579, + "learning_rate": 1.1637773020731108e-06, + "loss": 3.0161, + "step": 1949 + }, + { + "epoch": 0.25245169433925624, + "grad_norm": 5.080701908629334, + "learning_rate": 1.1636340782681578e-06, + "loss": 3.28, + "step": 1950 + }, + { + "epoch": 0.25245169433925624, + "eval_PRM Accuracy": 0.2358490566037736, + "eval_PRM F1": 0.047058823529411764, + "eval_PRM F1 AUC": 0.5120481927710844, + "eval_PRM F1 Neg": 0.36220472440944884, + "eval_PRM NPV": 0.22115384615384615, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.024096385542168676, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6610440015792847, + "eval_runtime": 14.35, + "eval_samples_per_second": 3.066, + "eval_steps_per_second": 0.209, + "step": 1950 + }, + { + "epoch": 0.2525811567466097, + "grad_norm": 2.6845066594555678, + "learning_rate": 1.1634907444362767e-06, + "loss": 3.0469, + "step": 1951 + }, + { + "epoch": 0.25271061915396315, + "grad_norm": 3.1470433673561042, + "learning_rate": 1.1633473006067464e-06, + "loss": 2.9512, + "step": 1952 + }, + { + "epoch": 0.25284008156131665, + "grad_norm": 2.1104458544901044, + "learning_rate": 1.1632037468088682e-06, + "loss": 2.8396, + "step": 1953 + }, + { + "epoch": 0.2529695439686701, + "grad_norm": 3.0694203591952545, + "learning_rate": 1.1630600830719657e-06, + "loss": 3.1455, + "step": 1954 + }, + { + "epoch": 0.25309900637602356, + "grad_norm": 3.054022978685594, + "learning_rate": 1.1629163094253853e-06, + "loss": 3.1145, + "step": 1955 + }, + { + "epoch": 0.25309900637602356, + "eval_PRM Accuracy": 0.2358490566037736, + "eval_PRM F1": 0.047058823529411764, + "eval_PRM F1 AUC": 0.5120481927710844, + "eval_PRM F1 Neg": 0.36220472440944884, + "eval_PRM NPV": 0.22115384615384615, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.024096385542168676, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6629971265792847, + "eval_runtime": 14.6532, + "eval_samples_per_second": 3.003, + "eval_steps_per_second": 0.205, + "step": 1955 + }, + { + "epoch": 0.253228468783377, + "grad_norm": 2.59793924117539, + "learning_rate": 1.1627724258984956e-06, + "loss": 2.6399, + "step": 1956 + }, + { + "epoch": 0.25335793119073047, + "grad_norm": 2.9695030332991554, + "learning_rate": 1.1626284325206878e-06, + "loss": 2.9143, + "step": 1957 + }, + { + "epoch": 0.253487393598084, + "grad_norm": 3.5658172072979224, + "learning_rate": 1.1624843293213754e-06, + "loss": 2.9331, + "step": 1958 + }, + { + "epoch": 0.25361685600543743, + "grad_norm": 6.02672060693311, + "learning_rate": 1.1623401163299944e-06, + "loss": 3.7141, + "step": 1959 + }, + { + "epoch": 0.2537463184127909, + "grad_norm": 2.5545131595119943, + "learning_rate": 1.1621957935760033e-06, + "loss": 2.7351, + "step": 1960 + }, + { + "epoch": 0.2537463184127909, + "eval_PRM Accuracy": 0.2358490566037736, + "eval_PRM F1": 0.047058823529411764, + "eval_PRM F1 AUC": 0.5120481927710844, + "eval_PRM F1 Neg": 0.36220472440944884, + "eval_PRM NPV": 0.22115384615384615, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.024096385542168676, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6679909229278564, + "eval_runtime": 15.1648, + "eval_samples_per_second": 2.901, + "eval_steps_per_second": 0.198, + "step": 1960 + }, + { + "epoch": 0.25387578082014434, + "grad_norm": 2.827112950035107, + "learning_rate": 1.1620513610888829e-06, + "loss": 3.3955, + "step": 1961 + }, + { + "epoch": 0.2540052432274978, + "grad_norm": 3.1636385852004514, + "learning_rate": 1.1619068188981366e-06, + "loss": 2.9683, + "step": 1962 + }, + { + "epoch": 0.2541347056348513, + "grad_norm": 3.7404378662813853, + "learning_rate": 1.1617621670332896e-06, + "loss": 3.314, + "step": 1963 + }, + { + "epoch": 0.25426416804220475, + "grad_norm": 4.352457415208915, + "learning_rate": 1.1616174055238907e-06, + "loss": 3.3213, + "step": 1964 + }, + { + "epoch": 0.2543936304495582, + "grad_norm": 2.1336260965762457, + "learning_rate": 1.1614725343995096e-06, + "loss": 2.5703, + "step": 1965 + }, + { + "epoch": 0.2543936304495582, + "eval_PRM Accuracy": 0.2358490566037736, + "eval_PRM F1": 0.047058823529411764, + "eval_PRM F1 AUC": 0.5120481927710844, + "eval_PRM F1 Neg": 0.36220472440944884, + "eval_PRM NPV": 0.22115384615384615, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.024096385542168676, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6664817333221436, + "eval_runtime": 13.9433, + "eval_samples_per_second": 3.156, + "eval_steps_per_second": 0.215, + "step": 1965 + }, + { + "epoch": 0.25452309285691166, + "grad_norm": 2.4535077176087063, + "learning_rate": 1.1613275536897397e-06, + "loss": 3.0042, + "step": 1966 + }, + { + "epoch": 0.2546525552642651, + "grad_norm": 2.031682185170567, + "learning_rate": 1.1611824634241962e-06, + "loss": 2.9478, + "step": 1967 + }, + { + "epoch": 0.2547820176716186, + "grad_norm": 3.261890774656847, + "learning_rate": 1.1610372636325165e-06, + "loss": 2.9756, + "step": 1968 + }, + { + "epoch": 0.2549114800789721, + "grad_norm": 4.095008951501019, + "learning_rate": 1.1608919543443608e-06, + "loss": 2.8286, + "step": 1969 + }, + { + "epoch": 0.25504094248632553, + "grad_norm": 3.7551506961258863, + "learning_rate": 1.1607465355894114e-06, + "loss": 3.4907, + "step": 1970 + }, + { + "epoch": 0.25504094248632553, + "eval_PRM Accuracy": 0.2358490566037736, + "eval_PRM F1": 0.047058823529411764, + "eval_PRM F1 AUC": 0.5120481927710844, + "eval_PRM F1 Neg": 0.36220472440944884, + "eval_PRM NPV": 0.22115384615384615, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.024096385542168676, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6677912473678589, + "eval_runtime": 14.5067, + "eval_samples_per_second": 3.033, + "eval_steps_per_second": 0.207, + "step": 1970 + }, + { + "epoch": 0.255170404893679, + "grad_norm": 5.452217070822066, + "learning_rate": 1.160601007397373e-06, + "loss": 3.9702, + "step": 1971 + }, + { + "epoch": 0.2552998673010325, + "grad_norm": 1.9282831671714697, + "learning_rate": 1.1604553697979725e-06, + "loss": 2.7742, + "step": 1972 + }, + { + "epoch": 0.25542932970838594, + "grad_norm": 2.7401274261259125, + "learning_rate": 1.1603096228209599e-06, + "loss": 2.7227, + "step": 1973 + }, + { + "epoch": 0.2555587921157394, + "grad_norm": 2.872327736649136, + "learning_rate": 1.1601637664961063e-06, + "loss": 2.9717, + "step": 1974 + }, + { + "epoch": 0.25568825452309285, + "grad_norm": 4.785932472762122, + "learning_rate": 1.1600178008532062e-06, + "loss": 3.0, + "step": 1975 + }, + { + "epoch": 0.25568825452309285, + "eval_PRM Accuracy": 0.2358490566037736, + "eval_PRM F1": 0.047058823529411764, + "eval_PRM F1 AUC": 0.5120481927710844, + "eval_PRM F1 Neg": 0.36220472440944884, + "eval_PRM NPV": 0.22115384615384615, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.024096385542168676, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6714088916778564, + "eval_runtime": 15.8787, + "eval_samples_per_second": 2.771, + "eval_steps_per_second": 0.189, + "step": 1975 + }, + { + "epoch": 0.2558177169304463, + "grad_norm": 3.7662031622478596, + "learning_rate": 1.1598717259220762e-06, + "loss": 2.936, + "step": 1976 + }, + { + "epoch": 0.2559471793377998, + "grad_norm": 2.055785163278057, + "learning_rate": 1.1597255417325544e-06, + "loss": 2.6257, + "step": 1977 + }, + { + "epoch": 0.25607664174515327, + "grad_norm": 2.3292145666544517, + "learning_rate": 1.1595792483145027e-06, + "loss": 2.823, + "step": 1978 + }, + { + "epoch": 0.2562061041525067, + "grad_norm": 2.5003043444882813, + "learning_rate": 1.1594328456978038e-06, + "loss": 2.6716, + "step": 1979 + }, + { + "epoch": 0.2563355665598602, + "grad_norm": 2.226384507862159, + "learning_rate": 1.159286333912364e-06, + "loss": 2.7947, + "step": 1980 + }, + { + "epoch": 0.2563355665598602, + "eval_PRM Accuracy": 0.2358490566037736, + "eval_PRM F1": 0.047058823529411764, + "eval_PRM F1 AUC": 0.5120481927710844, + "eval_PRM F1 Neg": 0.36220472440944884, + "eval_PRM NPV": 0.22115384615384615, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.024096385542168676, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.671431064605713, + "eval_runtime": 14.8882, + "eval_samples_per_second": 2.955, + "eval_steps_per_second": 0.202, + "step": 1980 + }, + { + "epoch": 0.2564650289672136, + "grad_norm": 2.371023306951928, + "learning_rate": 1.1591397129881106e-06, + "loss": 2.7693, + "step": 1981 + }, + { + "epoch": 0.25659449137456714, + "grad_norm": 9.343558328232445, + "learning_rate": 1.1589929829549947e-06, + "loss": 4.0332, + "step": 1982 + }, + { + "epoch": 0.2567239537819206, + "grad_norm": 4.712377500802992, + "learning_rate": 1.1588461438429882e-06, + "loss": 2.9692, + "step": 1983 + }, + { + "epoch": 0.25685341618927404, + "grad_norm": 2.62681756261129, + "learning_rate": 1.1586991956820864e-06, + "loss": 3.2104, + "step": 1984 + }, + { + "epoch": 0.2569828785966275, + "grad_norm": 4.675799683507873, + "learning_rate": 1.1585521385023062e-06, + "loss": 3.2573, + "step": 1985 + }, + { + "epoch": 0.2569828785966275, + "eval_PRM Accuracy": 0.2358490566037736, + "eval_PRM F1": 0.047058823529411764, + "eval_PRM F1 AUC": 0.5120481927710844, + "eval_PRM F1 Neg": 0.36220472440944884, + "eval_PRM NPV": 0.22115384615384615, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.024096385542168676, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6739612817764282, + "eval_runtime": 14.5772, + "eval_samples_per_second": 3.018, + "eval_steps_per_second": 0.206, + "step": 1985 + }, + { + "epoch": 0.25711234100398095, + "grad_norm": 4.361973393201989, + "learning_rate": 1.1584049723336874e-06, + "loss": 3.1494, + "step": 1986 + }, + { + "epoch": 0.25724180341133446, + "grad_norm": 2.5864916173219825, + "learning_rate": 1.1582576972062912e-06, + "loss": 2.7046, + "step": 1987 + }, + { + "epoch": 0.2573712658186879, + "grad_norm": 4.057478533906205, + "learning_rate": 1.1581103131502016e-06, + "loss": 3.0259, + "step": 1988 + }, + { + "epoch": 0.25750072822604136, + "grad_norm": 2.3452153576282764, + "learning_rate": 1.1579628201955248e-06, + "loss": 2.9102, + "step": 1989 + }, + { + "epoch": 0.2576301906333948, + "grad_norm": 4.00776815918346, + "learning_rate": 1.1578152183723895e-06, + "loss": 3.085, + "step": 1990 + }, + { + "epoch": 0.2576301906333948, + "eval_PRM Accuracy": 0.2358490566037736, + "eval_PRM F1": 0.047058823529411764, + "eval_PRM F1 AUC": 0.5120481927710844, + "eval_PRM F1 Neg": 0.36220472440944884, + "eval_PRM NPV": 0.22115384615384615, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.024096385542168676, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6740056276321411, + "eval_runtime": 14.3788, + "eval_samples_per_second": 3.06, + "eval_steps_per_second": 0.209, + "step": 1990 + }, + { + "epoch": 0.25775965304074827, + "grad_norm": 1.627012938499294, + "learning_rate": 1.1576675077109462e-06, + "loss": 2.7305, + "step": 1991 + }, + { + "epoch": 0.2578891154481018, + "grad_norm": 2.5493138688048798, + "learning_rate": 1.1575196882413677e-06, + "loss": 3.1897, + "step": 1992 + }, + { + "epoch": 0.25801857785545523, + "grad_norm": 2.7078753517169165, + "learning_rate": 1.1573717599938492e-06, + "loss": 2.6023, + "step": 1993 + }, + { + "epoch": 0.2581480402628087, + "grad_norm": 4.831892795699603, + "learning_rate": 1.157223722998608e-06, + "loss": 3.3555, + "step": 1994 + }, + { + "epoch": 0.25827750267016214, + "grad_norm": 3.2439017894440947, + "learning_rate": 1.1570755772858837e-06, + "loss": 3.2395, + "step": 1995 + }, + { + "epoch": 0.25827750267016214, + "eval_PRM Accuracy": 0.2358490566037736, + "eval_PRM F1": 0.047058823529411764, + "eval_PRM F1 AUC": 0.5120481927710844, + "eval_PRM F1 Neg": 0.36220472440944884, + "eval_PRM NPV": 0.22115384615384615, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.024096385542168676, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.676957607269287, + "eval_runtime": 14.4971, + "eval_samples_per_second": 3.035, + "eval_steps_per_second": 0.207, + "step": 1995 + }, + { + "epoch": 0.2584069650775156, + "grad_norm": 1.9346674939905308, + "learning_rate": 1.1569273228859381e-06, + "loss": 2.9404, + "step": 1996 + }, + { + "epoch": 0.2585364274848691, + "grad_norm": 2.999529166710086, + "learning_rate": 1.1567789598290552e-06, + "loss": 3.043, + "step": 1997 + }, + { + "epoch": 0.25866588989222256, + "grad_norm": 3.5759652702235187, + "learning_rate": 1.156630488145541e-06, + "loss": 3.0403, + "step": 1998 + }, + { + "epoch": 0.258795352299576, + "grad_norm": 4.313763829851741, + "learning_rate": 1.1564819078657238e-06, + "loss": 3.0625, + "step": 1999 + }, + { + "epoch": 0.25892481470692946, + "grad_norm": 2.9395046851351374, + "learning_rate": 1.1563332190199544e-06, + "loss": 2.6731, + "step": 2000 + }, + { + "epoch": 0.25892481470692946, + "eval_PRM Accuracy": 0.2358490566037736, + "eval_PRM F1": 0.047058823529411764, + "eval_PRM F1 AUC": 0.5120481927710844, + "eval_PRM F1 Neg": 0.36220472440944884, + "eval_PRM NPV": 0.22115384615384615, + "eval_PRM Precision": 1.0, + "eval_PRM Recall": 0.024096385542168676, + "eval_PRM Specificty": 1.0, + "eval_loss": 1.6755592823028564, + "eval_runtime": 14.1564, + "eval_samples_per_second": 3.108, + "eval_steps_per_second": 0.212, + "step": 2000 } ], "logging_steps": 1, @@ -15342,7 +20442,7 @@ "attributes": {} } }, - "total_flos": 1259448129847296.0, + "total_flos": 1679157807448064.0, "train_batch_size": 4, "trial_name": null, "trial_params": null