| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.3679890560875512, | |
| "eval_steps": 500, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.022799817601459188, | |
| "grad_norm": 0.765625, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.2602, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.045599635202918376, | |
| "grad_norm": 0.546875, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.1504, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.06839945280437756, | |
| "grad_norm": 0.44140625, | |
| "learning_rate": 0.0001, | |
| "loss": 0.137, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.09119927040583675, | |
| "grad_norm": 0.5390625, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 0.1296, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.11399908800729594, | |
| "grad_norm": 0.84765625, | |
| "learning_rate": 0.0001666666666666667, | |
| "loss": 0.1308, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.13679890560875513, | |
| "grad_norm": 0.376953125, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0811, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.15959872321021432, | |
| "grad_norm": 0.439453125, | |
| "learning_rate": 0.00019988560970029743, | |
| "loss": 0.0842, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.1823985408116735, | |
| "grad_norm": 0.5, | |
| "learning_rate": 0.00019977121940059484, | |
| "loss": 0.0772, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.2051983584131327, | |
| "grad_norm": 0.44921875, | |
| "learning_rate": 0.00019965682910089226, | |
| "loss": 0.0799, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.22799817601459188, | |
| "grad_norm": 0.5078125, | |
| "learning_rate": 0.00019954243880118967, | |
| "loss": 0.0733, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2507979936160511, | |
| "grad_norm": 0.4140625, | |
| "learning_rate": 0.0001994280485014871, | |
| "loss": 0.1053, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.27359781121751026, | |
| "grad_norm": 0.431640625, | |
| "learning_rate": 0.00019931365820178448, | |
| "loss": 0.0868, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.29639762881896947, | |
| "grad_norm": 0.87109375, | |
| "learning_rate": 0.00019919926790208192, | |
| "loss": 0.0824, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.31919744642042863, | |
| "grad_norm": 1.0390625, | |
| "learning_rate": 0.00019908487760237934, | |
| "loss": 0.0667, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.34199726402188785, | |
| "grad_norm": 0.8671875, | |
| "learning_rate": 0.00019897048730267673, | |
| "loss": 0.0742, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.364797081623347, | |
| "grad_norm": 0.5703125, | |
| "learning_rate": 0.00019885609700297417, | |
| "loss": 0.0815, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.3875968992248062, | |
| "grad_norm": 0.8203125, | |
| "learning_rate": 0.00019874170670327156, | |
| "loss": 0.0695, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.4103967168262654, | |
| "grad_norm": 0.56640625, | |
| "learning_rate": 0.000198627316403569, | |
| "loss": 0.0564, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.4331965344277246, | |
| "grad_norm": 0.427734375, | |
| "learning_rate": 0.0001985129261038664, | |
| "loss": 0.051, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.45599635202918376, | |
| "grad_norm": 0.83203125, | |
| "learning_rate": 0.0001983985358041638, | |
| "loss": 0.0745, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.478796169630643, | |
| "grad_norm": 0.625, | |
| "learning_rate": 0.00019828414550446125, | |
| "loss": 0.0949, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.5015959872321022, | |
| "grad_norm": 0.337890625, | |
| "learning_rate": 0.00019816975520475864, | |
| "loss": 0.1331, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.5243958048335613, | |
| "grad_norm": 0.828125, | |
| "learning_rate": 0.00019805536490505606, | |
| "loss": 0.1326, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.5471956224350205, | |
| "grad_norm": 0.78515625, | |
| "learning_rate": 0.00019794097460535347, | |
| "loss": 0.1135, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.5699954400364797, | |
| "grad_norm": 0.66015625, | |
| "learning_rate": 0.0001978265843056509, | |
| "loss": 0.1085, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.5927952576379389, | |
| "grad_norm": 0.625, | |
| "learning_rate": 0.0001977121940059483, | |
| "loss": 0.1347, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.615595075239398, | |
| "grad_norm": 0.73828125, | |
| "learning_rate": 0.00019759780370624572, | |
| "loss": 0.102, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.6383948928408573, | |
| "grad_norm": 0.578125, | |
| "learning_rate": 0.00019748341340654314, | |
| "loss": 0.0991, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.6611947104423165, | |
| "grad_norm": 0.25390625, | |
| "learning_rate": 0.00019736902310684055, | |
| "loss": 0.1125, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.6839945280437757, | |
| "grad_norm": 0.87109375, | |
| "learning_rate": 0.00019725463280713797, | |
| "loss": 0.1183, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.7067943456452348, | |
| "grad_norm": 0.734375, | |
| "learning_rate": 0.00019714024250743536, | |
| "loss": 0.105, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.729594163246694, | |
| "grad_norm": 0.49609375, | |
| "learning_rate": 0.0001970258522077328, | |
| "loss": 0.1315, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.7523939808481532, | |
| "grad_norm": 0.59765625, | |
| "learning_rate": 0.00019691146190803022, | |
| "loss": 0.1133, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.7751937984496124, | |
| "grad_norm": 0.6484375, | |
| "learning_rate": 0.00019679707160832763, | |
| "loss": 0.1111, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.7979936160510716, | |
| "grad_norm": 0.671875, | |
| "learning_rate": 0.00019668268130862505, | |
| "loss": 0.1277, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.8207934336525308, | |
| "grad_norm": 0.392578125, | |
| "learning_rate": 0.00019656829100892244, | |
| "loss": 0.1145, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.84359325125399, | |
| "grad_norm": 0.63671875, | |
| "learning_rate": 0.00019645390070921988, | |
| "loss": 0.1144, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.8663930688554492, | |
| "grad_norm": 0.69921875, | |
| "learning_rate": 0.00019633951040951727, | |
| "loss": 0.0995, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.8891928864569083, | |
| "grad_norm": 0.8984375, | |
| "learning_rate": 0.00019622512010981468, | |
| "loss": 0.0972, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.9119927040583675, | |
| "grad_norm": 0.306640625, | |
| "learning_rate": 0.00019611072981011213, | |
| "loss": 0.105, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9347925216598267, | |
| "grad_norm": 0.419921875, | |
| "learning_rate": 0.00019599633951040952, | |
| "loss": 0.1285, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.957592339261286, | |
| "grad_norm": 0.52734375, | |
| "learning_rate": 0.00019588194921070696, | |
| "loss": 0.1059, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.9803921568627451, | |
| "grad_norm": 0.79296875, | |
| "learning_rate": 0.00019576755891100435, | |
| "loss": 0.1013, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 1.0031919744642044, | |
| "grad_norm": 0.55078125, | |
| "learning_rate": 0.00019565316861130177, | |
| "loss": 0.0989, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.0259917920656634, | |
| "grad_norm": 0.84765625, | |
| "learning_rate": 0.00019553877831159918, | |
| "loss": 0.1024, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.0487916096671226, | |
| "grad_norm": 0.439453125, | |
| "learning_rate": 0.0001954243880118966, | |
| "loss": 0.0818, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.0715914272685818, | |
| "grad_norm": 0.59765625, | |
| "learning_rate": 0.000195309997712194, | |
| "loss": 0.0847, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 1.094391244870041, | |
| "grad_norm": 0.6640625, | |
| "learning_rate": 0.00019519560741249143, | |
| "loss": 0.0817, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.1171910624715002, | |
| "grad_norm": 0.61328125, | |
| "learning_rate": 0.00019508121711278885, | |
| "loss": 0.0742, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 1.1399908800729595, | |
| "grad_norm": 0.2265625, | |
| "learning_rate": 0.00019496682681308626, | |
| "loss": 0.0493, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.1627906976744187, | |
| "grad_norm": 0.4765625, | |
| "learning_rate": 0.00019485243651338368, | |
| "loss": 0.049, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 1.1855905152758779, | |
| "grad_norm": 0.451171875, | |
| "learning_rate": 0.0001947380462136811, | |
| "loss": 0.04, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.2083903328773369, | |
| "grad_norm": 0.39453125, | |
| "learning_rate": 0.0001946236559139785, | |
| "loss": 0.0446, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 1.231190150478796, | |
| "grad_norm": 0.5546875, | |
| "learning_rate": 0.00019450926561427593, | |
| "loss": 0.0398, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.2539899680802553, | |
| "grad_norm": 0.353515625, | |
| "learning_rate": 0.00019439487531457334, | |
| "loss": 0.0547, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 1.2767897856817145, | |
| "grad_norm": 0.83203125, | |
| "learning_rate": 0.00019428048501487076, | |
| "loss": 0.0503, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.2995896032831737, | |
| "grad_norm": 0.328125, | |
| "learning_rate": 0.00019416609471516815, | |
| "loss": 0.0378, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 1.322389420884633, | |
| "grad_norm": 0.54296875, | |
| "learning_rate": 0.0001940517044154656, | |
| "loss": 0.04, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.3451892384860922, | |
| "grad_norm": 0.82421875, | |
| "learning_rate": 0.00019393731411576298, | |
| "loss": 0.0362, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 1.3679890560875512, | |
| "grad_norm": 1.7890625, | |
| "learning_rate": 0.0001938229238160604, | |
| "loss": 0.0445, | |
| "step": 3000 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 87720, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "save_steps": 200, | |
| "total_flos": 0.0, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |