| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.903614457831326, | |
| "eval_steps": 500, | |
| "global_step": 822, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12048192771084337, | |
| "grad_norm": 2.6048758029937744, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 1.2322, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.24096385542168675, | |
| "grad_norm": 2.073004722595215, | |
| "learning_rate": 9.523809523809524e-05, | |
| "loss": 0.4195, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3614457831325301, | |
| "grad_norm": 1.133252501487732, | |
| "learning_rate": 0.00014285714285714287, | |
| "loss": 0.1977, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4819277108433735, | |
| "grad_norm": 0.7059364914894104, | |
| "learning_rate": 0.00019047619047619048, | |
| "loss": 0.1374, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6024096385542169, | |
| "grad_norm": 0.7211808562278748, | |
| "learning_rate": 0.0001999480933568615, | |
| "loss": 0.1198, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.7228915662650602, | |
| "grad_norm": 0.5436549782752991, | |
| "learning_rate": 0.00019973731496914914, | |
| "loss": 0.0994, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8433734939759037, | |
| "grad_norm": 0.612585186958313, | |
| "learning_rate": 0.00019936476229183133, | |
| "loss": 0.096, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.963855421686747, | |
| "grad_norm": 0.6169544458389282, | |
| "learning_rate": 0.00019883103960671305, | |
| "loss": 0.086, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0843373493975903, | |
| "grad_norm": 0.5241321921348572, | |
| "learning_rate": 0.00019813701261394136, | |
| "loss": 0.0838, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.2048192771084336, | |
| "grad_norm": 0.3530048429965973, | |
| "learning_rate": 0.00019728380702783643, | |
| "loss": 0.0685, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.3253012048192772, | |
| "grad_norm": 0.4779577851295471, | |
| "learning_rate": 0.00019627280675097908, | |
| "loss": 0.0667, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.4457831325301205, | |
| "grad_norm": 0.7595181465148926, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.0626, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.5662650602409638, | |
| "grad_norm": 0.4125242531299591, | |
| "learning_rate": 0.00019378423479332046, | |
| "loss": 0.0563, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.6867469879518073, | |
| "grad_norm": 0.3203200697898865, | |
| "learning_rate": 0.0001923106995853349, | |
| "loss": 0.0621, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.8072289156626506, | |
| "grad_norm": 0.3810570240020752, | |
| "learning_rate": 0.00019068743608505455, | |
| "loss": 0.0582, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.927710843373494, | |
| "grad_norm": 0.2916155457496643, | |
| "learning_rate": 0.00018891707723181294, | |
| "loss": 0.0598, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.0481927710843375, | |
| "grad_norm": 0.5267001986503601, | |
| "learning_rate": 0.00018700249455414394, | |
| "loss": 0.0541, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.1686746987951806, | |
| "grad_norm": 0.33983558416366577, | |
| "learning_rate": 0.0001849467935121521, | |
| "loss": 0.0524, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.289156626506024, | |
| "grad_norm": 0.36935296654701233, | |
| "learning_rate": 0.000182753308460445, | |
| "loss": 0.0526, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.4096385542168672, | |
| "grad_norm": 0.4720439016819, | |
| "learning_rate": 0.0001804255972397977, | |
| "loss": 0.0478, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.5301204819277108, | |
| "grad_norm": 0.572353720664978, | |
| "learning_rate": 0.00017796743540632223, | |
| "loss": 0.0504, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.6506024096385543, | |
| "grad_norm": 0.35790249705314636, | |
| "learning_rate": 0.0001753828101075017, | |
| "loss": 0.0476, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.7710843373493974, | |
| "grad_norm": 0.3537259101867676, | |
| "learning_rate": 0.00017267591361502232, | |
| "loss": 0.0456, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.891566265060241, | |
| "grad_norm": 0.34377923607826233, | |
| "learning_rate": 0.00016985113652489374, | |
| "loss": 0.0473, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.0120481927710845, | |
| "grad_norm": 0.32719871401786804, | |
| "learning_rate": 0.00016691306063588583, | |
| "loss": 0.046, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.1325301204819276, | |
| "grad_norm": 0.30426010489463806, | |
| "learning_rate": 0.0001638664515178348, | |
| "loss": 0.045, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.253012048192771, | |
| "grad_norm": 0.4393266439437866, | |
| "learning_rate": 0.00016071625078187114, | |
| "loss": 0.0407, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.3734939759036147, | |
| "grad_norm": 0.48441895842552185, | |
| "learning_rate": 0.00015746756806510838, | |
| "loss": 0.0403, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.4939759036144578, | |
| "grad_norm": 0.2893245220184326, | |
| "learning_rate": 0.00015412567274279316, | |
| "loss": 0.039, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.6144578313253013, | |
| "grad_norm": 0.33335432410240173, | |
| "learning_rate": 0.00015069598538135906, | |
| "loss": 0.0374, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.734939759036145, | |
| "grad_norm": 0.36576047539711, | |
| "learning_rate": 0.0001471840689462482, | |
| "loss": 0.0424, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.855421686746988, | |
| "grad_norm": 0.2801503539085388, | |
| "learning_rate": 0.00014359561977876102, | |
| "loss": 0.0332, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.9759036144578315, | |
| "grad_norm": 0.29735037684440613, | |
| "learning_rate": 0.00013993645835656953, | |
| "loss": 0.0355, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.096385542168675, | |
| "grad_norm": 0.43135517835617065, | |
| "learning_rate": 0.0001362125198528817, | |
| "loss": 0.0442, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.216867469879518, | |
| "grad_norm": 0.38527563214302063, | |
| "learning_rate": 0.00013242984450956828, | |
| "loss": 0.0356, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.337349397590361, | |
| "grad_norm": 0.3635711073875427, | |
| "learning_rate": 0.00012859456783986893, | |
| "loss": 0.041, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.457831325301205, | |
| "grad_norm": 0.2927173972129822, | |
| "learning_rate": 0.00012471291067656697, | |
| "loss": 0.034, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.578313253012048, | |
| "grad_norm": 0.407322496175766, | |
| "learning_rate": 0.00012079116908177593, | |
| "loss": 0.0327, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.698795180722891, | |
| "grad_norm": 0.33329063653945923, | |
| "learning_rate": 0.00011683570413470383, | |
| "loss": 0.0339, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.8192771084337345, | |
| "grad_norm": 0.2710469663143158, | |
| "learning_rate": 0.00011285293161395946, | |
| "loss": 0.0299, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.9397590361445785, | |
| "grad_norm": 0.308614045381546, | |
| "learning_rate": 0.00010884931159113586, | |
| "loss": 0.0386, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 5.0602409638554215, | |
| "grad_norm": 0.3434526324272156, | |
| "learning_rate": 0.00010483133795255071, | |
| "loss": 0.0367, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 5.180722891566265, | |
| "grad_norm": 0.3069072961807251, | |
| "learning_rate": 0.00010080552786613899, | |
| "loss": 0.0356, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 5.301204819277109, | |
| "grad_norm": 0.2935517430305481, | |
| "learning_rate": 9.677841121058273e-05, | |
| "loss": 0.0304, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 5.421686746987952, | |
| "grad_norm": 0.3041512072086334, | |
| "learning_rate": 9.275651998382377e-05, | |
| "loss": 0.0289, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.542168674698795, | |
| "grad_norm": 0.30504339933395386, | |
| "learning_rate": 8.874637770813946e-05, | |
| "loss": 0.0302, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.662650602409639, | |
| "grad_norm": 0.20522096753120422, | |
| "learning_rate": 8.475448884896547e-05, | |
| "loss": 0.029, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 5.783132530120482, | |
| "grad_norm": 0.2943919897079468, | |
| "learning_rate": 8.078732826462915e-05, | |
| "loss": 0.0288, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.903614457831325, | |
| "grad_norm": 0.328308641910553, | |
| "learning_rate": 7.685133070410571e-05, | |
| "loss": 0.0237, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 6.024096385542169, | |
| "grad_norm": 0.425624281167984, | |
| "learning_rate": 7.295288036983163e-05, | |
| "loss": 0.0303, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.144578313253012, | |
| "grad_norm": 0.2587184011936188, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.0246, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 6.265060240963855, | |
| "grad_norm": 0.35053542256355286, | |
| "learning_rate": 6.52938434246697e-05, | |
| "loss": 0.0273, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 6.385542168674699, | |
| "grad_norm": 0.2893577516078949, | |
| "learning_rate": 6.154567979971493e-05, | |
| "loss": 0.026, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 6.506024096385542, | |
| "grad_norm": 0.22769379615783691, | |
| "learning_rate": 5.785988922274711e-05, | |
| "loss": 0.029, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 6.626506024096385, | |
| "grad_norm": 0.3005225956439972, | |
| "learning_rate": 5.424245005956048e-05, | |
| "loss": 0.0246, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 6.746987951807229, | |
| "grad_norm": 0.2721225321292877, | |
| "learning_rate": 5.069922980970626e-05, | |
| "loss": 0.0256, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 6.867469879518072, | |
| "grad_norm": 0.23859107494354248, | |
| "learning_rate": 4.723597558938672e-05, | |
| "loss": 0.0295, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 6.9879518072289155, | |
| "grad_norm": 0.2128613144159317, | |
| "learning_rate": 4.385830480961192e-05, | |
| "loss": 0.0271, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 7.108433734939759, | |
| "grad_norm": 0.15234726667404175, | |
| "learning_rate": 4.057169606473827e-05, | |
| "loss": 0.0209, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 7.228915662650603, | |
| "grad_norm": 0.1998305469751358, | |
| "learning_rate": 3.738148024616863e-05, | |
| "loss": 0.0236, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.349397590361446, | |
| "grad_norm": 0.16628368198871613, | |
| "learning_rate": 3.429283189562694e-05, | |
| "loss": 0.022, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 7.469879518072289, | |
| "grad_norm": 0.22349688410758972, | |
| "learning_rate": 3.131076081203247e-05, | |
| "loss": 0.0218, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 7.590361445783133, | |
| "grad_norm": 0.1814289093017578, | |
| "learning_rate": 2.84401039255879e-05, | |
| "loss": 0.0201, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 7.710843373493976, | |
| "grad_norm": 0.228335902094841, | |
| "learning_rate": 2.5685517452260567e-05, | |
| "loss": 0.0224, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 7.831325301204819, | |
| "grad_norm": 0.1545294225215912, | |
| "learning_rate": 2.3051469341383402e-05, | |
| "loss": 0.0238, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 7.951807228915663, | |
| "grad_norm": 0.1964937150478363, | |
| "learning_rate": 2.0542232028624586e-05, | |
| "loss": 0.0232, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 8.072289156626505, | |
| "grad_norm": 0.15069320797920227, | |
| "learning_rate": 1.8161875506081293e-05, | |
| "loss": 0.0206, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 8.19277108433735, | |
| "grad_norm": 0.13602782785892487, | |
| "learning_rate": 1.5914260720737795e-05, | |
| "loss": 0.0187, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 8.313253012048193, | |
| "grad_norm": 0.19349129498004913, | |
| "learning_rate": 1.3803033311995072e-05, | |
| "loss": 0.0192, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 8.433734939759036, | |
| "grad_norm": 0.1513078659772873, | |
| "learning_rate": 1.1831617698430609e-05, | |
| "loss": 0.0171, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 8.55421686746988, | |
| "grad_norm": 0.15297076106071472, | |
| "learning_rate": 1.0003211523378796e-05, | |
| "loss": 0.0187, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 8.674698795180722, | |
| "grad_norm": 0.1606190949678421, | |
| "learning_rate": 8.32078046834176e-06, | |
| "loss": 0.0198, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 8.795180722891565, | |
| "grad_norm": 0.11450409889221191, | |
| "learning_rate": 6.787053442643232e-06, | |
| "loss": 0.0196, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 8.91566265060241, | |
| "grad_norm": 0.23408889770507812, | |
| "learning_rate": 5.40451815712748e-06, | |
| "loss": 0.0197, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 9.036144578313253, | |
| "grad_norm": 0.19213688373565674, | |
| "learning_rate": 4.175417089083378e-06, | |
| "loss": 0.0193, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 9.156626506024097, | |
| "grad_norm": 0.13061603903770447, | |
| "learning_rate": 3.1017438449379434e-06, | |
| "loss": 0.0217, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 9.27710843373494, | |
| "grad_norm": 0.20497524738311768, | |
| "learning_rate": 2.1852399266194314e-06, | |
| "loss": 0.0206, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 9.397590361445783, | |
| "grad_norm": 0.16401517391204834, | |
| "learning_rate": 1.4273919068349184e-06, | |
| "loss": 0.0218, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 9.518072289156626, | |
| "grad_norm": 0.08886178582906723, | |
| "learning_rate": 8.294290178437969e-07, | |
| "loss": 0.0181, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 9.638554216867469, | |
| "grad_norm": 0.1821727752685547, | |
| "learning_rate": 3.923211576387087e-07, | |
| "loss": 0.0182, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 9.759036144578314, | |
| "grad_norm": 0.17254865169525146, | |
| "learning_rate": 1.1677731676733584e-07, | |
| "loss": 0.0193, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 9.879518072289157, | |
| "grad_norm": 0.18676318228244781, | |
| "learning_rate": 3.244428347204398e-09, | |
| "loss": 0.0161, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 9.903614457831326, | |
| "step": 822, | |
| "total_flos": 1.0767742262918784e+17, | |
| "train_loss": 0.05996827910361934, | |
| "train_runtime": 931.6223, | |
| "train_samples_per_second": 56.469, | |
| "train_steps_per_second": 0.882 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 822, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 10000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0767742262918784e+17, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |