| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.048951048951049, | |
| "eval_steps": 9, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.013986013986013986, | |
| "grad_norm": 0.01746705360710621, | |
| "learning_rate": 1e-05, | |
| "loss": 10.3794, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.013986013986013986, | |
| "eval_loss": 10.378853797912598, | |
| "eval_runtime": 0.2703, | |
| "eval_samples_per_second": 887.793, | |
| "eval_steps_per_second": 29.593, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.027972027972027972, | |
| "grad_norm": 0.017648639157414436, | |
| "learning_rate": 2e-05, | |
| "loss": 10.3791, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.04195804195804196, | |
| "grad_norm": 0.014356564730405807, | |
| "learning_rate": 3e-05, | |
| "loss": 10.3763, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.055944055944055944, | |
| "grad_norm": 0.014636906795203686, | |
| "learning_rate": 4e-05, | |
| "loss": 10.3774, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.06993006993006994, | |
| "grad_norm": 0.01596330665051937, | |
| "learning_rate": 5e-05, | |
| "loss": 10.378, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.08391608391608392, | |
| "grad_norm": 0.015443079173564911, | |
| "learning_rate": 6e-05, | |
| "loss": 10.3777, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0979020979020979, | |
| "grad_norm": 0.016583969816565514, | |
| "learning_rate": 7e-05, | |
| "loss": 10.3779, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.11188811188811189, | |
| "grad_norm": 0.016844727098941803, | |
| "learning_rate": 8e-05, | |
| "loss": 10.3803, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.1258741258741259, | |
| "grad_norm": 0.015567583031952381, | |
| "learning_rate": 9e-05, | |
| "loss": 10.3784, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.1258741258741259, | |
| "eval_loss": 10.378643989562988, | |
| "eval_runtime": 0.2702, | |
| "eval_samples_per_second": 888.337, | |
| "eval_steps_per_second": 29.611, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.13986013986013987, | |
| "grad_norm": 0.016822688281536102, | |
| "learning_rate": 0.0001, | |
| "loss": 10.3798, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 0.016712935641407967, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 10.3767, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.16783216783216784, | |
| "grad_norm": 0.017638981342315674, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 10.3783, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 0.017003795132040977, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 10.3771, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.1958041958041958, | |
| "grad_norm": 0.014297555200755596, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 10.3767, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.2097902097902098, | |
| "grad_norm": 0.015179406851530075, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 10.377, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.22377622377622378, | |
| "grad_norm": 0.01621691696345806, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 10.378, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.23776223776223776, | |
| "grad_norm": 0.015583082102239132, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 10.3775, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.2517482517482518, | |
| "grad_norm": 0.014875374734401703, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 10.3774, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2517482517482518, | |
| "eval_loss": 10.378125190734863, | |
| "eval_runtime": 0.2728, | |
| "eval_samples_per_second": 879.689, | |
| "eval_steps_per_second": 29.323, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.26573426573426573, | |
| "grad_norm": 0.014841767027974129, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 10.3763, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.27972027972027974, | |
| "grad_norm": 0.01593184284865856, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 10.3778, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2937062937062937, | |
| "grad_norm": 0.016356591135263443, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 10.3782, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 0.016125470399856567, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 10.3769, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.32167832167832167, | |
| "grad_norm": 0.014987400732934475, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 10.378, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.3356643356643357, | |
| "grad_norm": 0.015122763812541962, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 10.3761, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.34965034965034963, | |
| "grad_norm": 0.016353605315089226, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 10.376, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.01983753964304924, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 10.3778, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.3776223776223776, | |
| "grad_norm": 0.018628032878041267, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 10.3774, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.3776223776223776, | |
| "eval_loss": 10.377577781677246, | |
| "eval_runtime": 0.2695, | |
| "eval_samples_per_second": 890.527, | |
| "eval_steps_per_second": 29.684, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.3916083916083916, | |
| "grad_norm": 0.01745915226638317, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 10.3772, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.40559440559440557, | |
| "grad_norm": 0.018339073285460472, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 10.3751, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.4195804195804196, | |
| "grad_norm": 0.015603472478687763, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 10.3766, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.43356643356643354, | |
| "grad_norm": 0.020217476412653923, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 10.377, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.44755244755244755, | |
| "grad_norm": 0.020117416977882385, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 10.3755, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "grad_norm": 0.01999656856060028, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 10.3776, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.4755244755244755, | |
| "grad_norm": 0.021135497838258743, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 10.376, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.48951048951048953, | |
| "grad_norm": 0.020801398903131485, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 10.3757, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.5034965034965035, | |
| "grad_norm": 0.023042971268296242, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 10.3766, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.5034965034965035, | |
| "eval_loss": 10.376969337463379, | |
| "eval_runtime": 0.2663, | |
| "eval_samples_per_second": 901.284, | |
| "eval_steps_per_second": 30.043, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.5174825174825175, | |
| "grad_norm": 0.021229099482297897, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 10.3766, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.5314685314685315, | |
| "grad_norm": 0.022592877969145775, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 10.3744, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "grad_norm": 0.02457505464553833, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 10.377, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.5594405594405595, | |
| "grad_norm": 0.023601705208420753, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 10.3756, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5734265734265734, | |
| "grad_norm": 0.02284586802124977, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 10.3771, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.5874125874125874, | |
| "grad_norm": 0.021345047280192375, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 10.3756, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.6013986013986014, | |
| "grad_norm": 0.02439739555120468, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 10.375, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 0.02309122122824192, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 10.3761, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.6293706293706294, | |
| "grad_norm": 0.02600119635462761, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 10.3761, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6293706293706294, | |
| "eval_loss": 10.376312255859375, | |
| "eval_runtime": 0.2799, | |
| "eval_samples_per_second": 857.336, | |
| "eval_steps_per_second": 28.578, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6433566433566433, | |
| "grad_norm": 0.02472044713795185, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 10.3767, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.6573426573426573, | |
| "grad_norm": 0.026171980425715446, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 10.3751, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.6713286713286714, | |
| "grad_norm": 0.025832511484622955, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 10.3763, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.6853146853146853, | |
| "grad_norm": 0.028169607743620872, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 10.3757, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.6993006993006993, | |
| "grad_norm": 0.029072731733322144, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 10.3751, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.7132867132867133, | |
| "grad_norm": 0.025323813781142235, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 10.3755, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.030344007536768913, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 10.3769, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.7412587412587412, | |
| "grad_norm": 0.026691097766160965, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 10.3738, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.7552447552447552, | |
| "grad_norm": 0.030549872666597366, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 10.3772, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.7552447552447552, | |
| "eval_loss": 10.375645637512207, | |
| "eval_runtime": 0.2681, | |
| "eval_samples_per_second": 895.207, | |
| "eval_steps_per_second": 29.84, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 0.028555618599057198, | |
| "learning_rate": 5e-05, | |
| "loss": 10.376, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.7832167832167832, | |
| "grad_norm": 0.029428837820887566, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 10.3741, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.7972027972027972, | |
| "grad_norm": 0.029308155179023743, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 10.3746, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.8111888111888111, | |
| "grad_norm": 0.032606493681669235, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 10.3758, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.8251748251748252, | |
| "grad_norm": 0.02967746928334236, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 10.3742, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.8391608391608392, | |
| "grad_norm": 0.03057865984737873, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 10.3758, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8531468531468531, | |
| "grad_norm": 0.032224319875240326, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 10.3755, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.8671328671328671, | |
| "grad_norm": 0.040901295840740204, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 10.3744, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.8811188811188811, | |
| "grad_norm": 0.03291678801178932, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 10.3743, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.8811188811188811, | |
| "eval_loss": 10.37505054473877, | |
| "eval_runtime": 0.2661, | |
| "eval_samples_per_second": 901.766, | |
| "eval_steps_per_second": 30.059, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.8951048951048951, | |
| "grad_norm": 0.038531407713890076, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 10.3745, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 0.03420910984277725, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 10.3745, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 0.03833569958806038, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 10.374, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.9370629370629371, | |
| "grad_norm": 0.03805769979953766, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 10.375, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.951048951048951, | |
| "grad_norm": 0.0376318022608757, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 10.3774, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.965034965034965, | |
| "grad_norm": 0.03903169184923172, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 10.3733, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.9790209790209791, | |
| "grad_norm": 0.03841084614396095, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 10.3736, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.993006993006993, | |
| "grad_norm": 0.02962353080511093, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 10.3753, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 1.006993006993007, | |
| "grad_norm": 0.04900306835770607, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 15.5217, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.006993006993007, | |
| "eval_loss": 10.374591827392578, | |
| "eval_runtime": 0.2844, | |
| "eval_samples_per_second": 843.76, | |
| "eval_steps_per_second": 28.125, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.020979020979021, | |
| "grad_norm": 0.04284500703215599, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 10.2186, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 1.034965034965035, | |
| "grad_norm": 0.041206490248441696, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 10.4779, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.048951048951049, | |
| "grad_norm": 0.03888395428657532, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 10.0933, | |
| "step": 75 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 23457647296512.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |