| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.963369963369964, | |
| "eval_steps": 500, | |
| "global_step": 680, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14652014652014653, | |
| "grad_norm": 0.301513671875, | |
| "learning_rate": 0.00019989329748023725, | |
| "loss": 0.8924, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.29304029304029305, | |
| "grad_norm": 0.32275390625, | |
| "learning_rate": 0.00019957341762950344, | |
| "loss": 0.8199, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.43956043956043955, | |
| "grad_norm": 0.353515625, | |
| "learning_rate": 0.0001990410430875205, | |
| "loss": 0.8061, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5860805860805861, | |
| "grad_norm": 0.445068359375, | |
| "learning_rate": 0.0001982973099683902, | |
| "loss": 0.8483, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.7326007326007326, | |
| "grad_norm": 0.396728515625, | |
| "learning_rate": 0.0001973438054360693, | |
| "loss": 0.8471, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.8791208791208791, | |
| "grad_norm": 0.35693359375, | |
| "learning_rate": 0.00019618256431728194, | |
| "loss": 0.8433, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.0256410256410255, | |
| "grad_norm": 0.360595703125, | |
| "learning_rate": 0.0001948160647590966, | |
| "loss": 0.8381, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.1721611721611722, | |
| "grad_norm": 0.37109375, | |
| "learning_rate": 0.00019324722294043558, | |
| "loss": 0.7955, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.3186813186813187, | |
| "grad_norm": 0.388671875, | |
| "learning_rate": 0.0001914793868488021, | |
| "loss": 0.8182, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.4652014652014653, | |
| "grad_norm": 0.45458984375, | |
| "learning_rate": 0.00018951632913550626, | |
| "loss": 0.7914, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.6117216117216118, | |
| "grad_norm": 0.51025390625, | |
| "learning_rate": 0.00018736223906463696, | |
| "loss": 0.8114, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.7582417582417582, | |
| "grad_norm": 0.52001953125, | |
| "learning_rate": 0.00018502171357296144, | |
| "loss": 0.8434, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 0.433349609375, | |
| "learning_rate": 0.00018249974745983023, | |
| "loss": 0.7869, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.051282051282051, | |
| "grad_norm": 0.62255859375, | |
| "learning_rate": 0.000179801722728024, | |
| "loss": 0.7767, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.197802197802198, | |
| "grad_norm": 0.469970703125, | |
| "learning_rate": 0.00017693339709828792, | |
| "loss": 0.7711, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.3443223443223444, | |
| "grad_norm": 0.447998046875, | |
| "learning_rate": 0.00017390089172206592, | |
| "loss": 0.7823, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.490842490842491, | |
| "grad_norm": 0.41015625, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.7747, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.6373626373626373, | |
| "grad_norm": 0.466796875, | |
| "learning_rate": 0.00016736956436465573, | |
| "loss": 0.7877, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.7838827838827838, | |
| "grad_norm": 0.48046875, | |
| "learning_rate": 0.00016388468056519612, | |
| "loss": 0.7778, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.9304029304029307, | |
| "grad_norm": 0.47998046875, | |
| "learning_rate": 0.00016026346363792567, | |
| "loss": 0.754, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 0.50244140625, | |
| "learning_rate": 0.0001565136414422592, | |
| "loss": 0.8089, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.2234432234432235, | |
| "grad_norm": 0.475830078125, | |
| "learning_rate": 0.0001526432162877356, | |
| "loss": 0.75, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.36996336996337, | |
| "grad_norm": 0.464111328125, | |
| "learning_rate": 0.00014866044785668563, | |
| "loss": 0.762, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.5164835164835164, | |
| "grad_norm": 0.61474609375, | |
| "learning_rate": 0.00014457383557765386, | |
| "loss": 0.7423, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.663003663003663, | |
| "grad_norm": 0.4873046875, | |
| "learning_rate": 0.00014039210048718949, | |
| "loss": 0.764, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 0.5546875, | |
| "learning_rate": 0.00013612416661871533, | |
| "loss": 0.7422, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.956043956043956, | |
| "grad_norm": 0.473388671875, | |
| "learning_rate": 0.00013177914195819016, | |
| "loss": 0.7702, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.102564102564102, | |
| "grad_norm": 0.476318359375, | |
| "learning_rate": 0.0001273662990072083, | |
| "loss": 0.7237, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.249084249084249, | |
| "grad_norm": 0.471923828125, | |
| "learning_rate": 0.0001228950549950134, | |
| "loss": 0.7334, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.395604395604396, | |
| "grad_norm": 0.50537109375, | |
| "learning_rate": 0.00011837495178165706, | |
| "loss": 0.7274, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.542124542124542, | |
| "grad_norm": 0.5, | |
| "learning_rate": 0.00011381563549518823, | |
| "loss": 0.7142, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.688644688644689, | |
| "grad_norm": 0.48779296875, | |
| "learning_rate": 0.00010922683594633021, | |
| "loss": 0.7166, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 4.835164835164835, | |
| "grad_norm": 0.513671875, | |
| "learning_rate": 0.00010461834586457398, | |
| "loss": 0.76, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.981684981684982, | |
| "grad_norm": 0.5048828125, | |
| "learning_rate": 0.0001, | |
| "loss": 0.7571, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 5.128205128205128, | |
| "grad_norm": 0.5166015625, | |
| "learning_rate": 9.538165413542607e-05, | |
| "loss": 0.7046, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 5.274725274725275, | |
| "grad_norm": 0.51416015625, | |
| "learning_rate": 9.077316405366981e-05, | |
| "loss": 0.6916, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 5.4212454212454215, | |
| "grad_norm": 0.61279296875, | |
| "learning_rate": 8.61843645048118e-05, | |
| "loss": 0.7281, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 5.5677655677655675, | |
| "grad_norm": 0.54736328125, | |
| "learning_rate": 8.162504821834295e-05, | |
| "loss": 0.6963, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 0.6103515625, | |
| "learning_rate": 7.710494500498662e-05, | |
| "loss": 0.7323, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 5.860805860805861, | |
| "grad_norm": 0.60595703125, | |
| "learning_rate": 7.263370099279172e-05, | |
| "loss": 0.7342, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.007326007326007, | |
| "grad_norm": 0.55078125, | |
| "learning_rate": 6.822085804180984e-05, | |
| "loss": 0.7018, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 0.55078125, | |
| "learning_rate": 6.387583338128471e-05, | |
| "loss": 0.6998, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 6.3003663003663, | |
| "grad_norm": 0.591796875, | |
| "learning_rate": 5.960789951281052e-05, | |
| "loss": 0.6639, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 6.446886446886447, | |
| "grad_norm": 0.60107421875, | |
| "learning_rate": 5.542616442234618e-05, | |
| "loss": 0.7004, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 6.593406593406593, | |
| "grad_norm": 0.55810546875, | |
| "learning_rate": 5.1339552143314384e-05, | |
| "loss": 0.7155, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 6.73992673992674, | |
| "grad_norm": 0.609375, | |
| "learning_rate": 4.735678371226441e-05, | |
| "loss": 0.7091, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 6.886446886446887, | |
| "grad_norm": 0.53662109375, | |
| "learning_rate": 4.3486358557740814e-05, | |
| "loss": 0.6888, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 7.032967032967033, | |
| "grad_norm": 0.5029296875, | |
| "learning_rate": 3.973653636207437e-05, | |
| "loss": 0.6888, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 7.17948717948718, | |
| "grad_norm": 0.53662109375, | |
| "learning_rate": 3.6115319434803894e-05, | |
| "loss": 0.6845, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 7.326007326007326, | |
| "grad_norm": 0.59130859375, | |
| "learning_rate": 3.263043563534428e-05, | |
| "loss": 0.6731, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.472527472527473, | |
| "grad_norm": 0.564453125, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.6708, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 7.619047619047619, | |
| "grad_norm": 0.62158203125, | |
| "learning_rate": 2.6099108277934103e-05, | |
| "loss": 0.6906, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 7.7655677655677655, | |
| "grad_norm": 0.572265625, | |
| "learning_rate": 2.3066602901712108e-05, | |
| "loss": 0.6841, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 7.912087912087912, | |
| "grad_norm": 0.5771484375, | |
| "learning_rate": 2.0198277271976052e-05, | |
| "loss": 0.6664, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 8.058608058608058, | |
| "grad_norm": 0.587890625, | |
| "learning_rate": 1.750025254016978e-05, | |
| "loss": 0.6899, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 8.205128205128204, | |
| "grad_norm": 0.55810546875, | |
| "learning_rate": 1.4978286427038601e-05, | |
| "loss": 0.6723, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 8.351648351648352, | |
| "grad_norm": 0.560546875, | |
| "learning_rate": 1.2637760935363053e-05, | |
| "loss": 0.6494, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 8.498168498168498, | |
| "grad_norm": 0.568359375, | |
| "learning_rate": 1.0483670864493778e-05, | |
| "loss": 0.6958, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 8.644688644688644, | |
| "grad_norm": 0.5947265625, | |
| "learning_rate": 8.520613151197898e-06, | |
| "loss": 0.6637, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 8.791208791208792, | |
| "grad_norm": 0.5556640625, | |
| "learning_rate": 6.75277705956443e-06, | |
| "loss": 0.6876, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 8.937728937728938, | |
| "grad_norm": 0.60791015625, | |
| "learning_rate": 5.183935240903414e-06, | |
| "loss": 0.6656, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 9.084249084249084, | |
| "grad_norm": 0.51416015625, | |
| "learning_rate": 3.817435682718096e-06, | |
| "loss": 0.6786, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 0.56689453125, | |
| "learning_rate": 2.656194563930714e-06, | |
| "loss": 0.668, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 9.377289377289378, | |
| "grad_norm": 0.5615234375, | |
| "learning_rate": 1.7026900316098215e-06, | |
| "loss": 0.6681, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 9.523809523809524, | |
| "grad_norm": 0.587890625, | |
| "learning_rate": 9.589569124794916e-07, | |
| "loss": 0.6496, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 9.67032967032967, | |
| "grad_norm": 0.57373046875, | |
| "learning_rate": 4.2658237049655323e-07, | |
| "loss": 0.6891, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 9.816849816849818, | |
| "grad_norm": 0.572265625, | |
| "learning_rate": 1.0670251976275803e-07, | |
| "loss": 0.6689, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 9.963369963369964, | |
| "grad_norm": 0.57470703125, | |
| "learning_rate": 0.0, | |
| "loss": 0.6729, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 9.963369963369964, | |
| "step": 680, | |
| "total_flos": 3.318222409433088e+16, | |
| "train_loss": 0.735039118458243, | |
| "train_runtime": 593.4308, | |
| "train_samples_per_second": 4.6, | |
| "train_steps_per_second": 1.146 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 680, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 3.318222409433088e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |