| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5555555555555556, | |
| "eval_steps": 30, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.011111111111111112, | |
| "grad_norm": 0.6721189022064209, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 0.4836, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.022222222222222223, | |
| "grad_norm": 0.06062021106481552, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.0217, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03333333333333333, | |
| "grad_norm": 0.034004177898168564, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 0.0043, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03333333333333333, | |
| "eval_loss": 0.004450319334864616, | |
| "eval_runtime": 393.1048, | |
| "eval_samples_per_second": 5.088, | |
| "eval_steps_per_second": 0.254, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.044444444444444446, | |
| "grad_norm": 0.014907660894095898, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.0043, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05555555555555555, | |
| "grad_norm": 0.007317671552300453, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.0039, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06666666666666667, | |
| "grad_norm": 0.03429165109992027, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.0041, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.06666666666666667, | |
| "eval_loss": 0.003960395231842995, | |
| "eval_runtime": 404.1276, | |
| "eval_samples_per_second": 4.949, | |
| "eval_steps_per_second": 0.247, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07777777777777778, | |
| "grad_norm": 0.7178479433059692, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.0042, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08888888888888889, | |
| "grad_norm": 0.08405599743127823, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.0039, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.01760088838636875, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.0042, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.003934871405363083, | |
| "eval_runtime": 392.97, | |
| "eval_samples_per_second": 5.089, | |
| "eval_steps_per_second": 0.254, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1111111111111111, | |
| "grad_norm": 0.009712091647088528, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0039, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.12222222222222222, | |
| "grad_norm": 0.00771428132429719, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.0039, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.13333333333333333, | |
| "grad_norm": 0.023663334548473358, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.0038, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.13333333333333333, | |
| "eval_loss": 0.00376720423810184, | |
| "eval_runtime": 391.7943, | |
| "eval_samples_per_second": 5.105, | |
| "eval_steps_per_second": 0.255, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.14444444444444443, | |
| "grad_norm": 0.004410896450281143, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 0.0039, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.15555555555555556, | |
| "grad_norm": 0.014762775972485542, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0038, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.16666666666666666, | |
| "grad_norm": 0.016578922048211098, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.0036, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16666666666666666, | |
| "eval_loss": 0.0037360445130616426, | |
| "eval_runtime": 394.3937, | |
| "eval_samples_per_second": 5.071, | |
| "eval_steps_per_second": 0.254, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.17777777777777778, | |
| "grad_norm": 0.012592969462275505, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.0038, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.18888888888888888, | |
| "grad_norm": 0.013751073740422726, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.0039, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.01368357427418232, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0038, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.003743910463526845, | |
| "eval_runtime": 397.88, | |
| "eval_samples_per_second": 5.027, | |
| "eval_steps_per_second": 0.251, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.2111111111111111, | |
| "grad_norm": 0.004404901526868343, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 0.0038, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 0.030225256457924843, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0039, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.23333333333333334, | |
| "grad_norm": 0.009082062169909477, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 0.0039, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.23333333333333334, | |
| "eval_loss": 0.0037830513902008533, | |
| "eval_runtime": 397.6789, | |
| "eval_samples_per_second": 5.029, | |
| "eval_steps_per_second": 0.251, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.24444444444444444, | |
| "grad_norm": 0.009521303698420525, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.0038, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.25555555555555554, | |
| "grad_norm": 0.007651821710169315, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.0038, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.26666666666666666, | |
| "grad_norm": 0.009911688044667244, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.0038, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.26666666666666666, | |
| "eval_loss": 0.003697082633152604, | |
| "eval_runtime": 397.7693, | |
| "eval_samples_per_second": 5.028, | |
| "eval_steps_per_second": 0.251, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.2777777777777778, | |
| "grad_norm": 0.022675497457385063, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.0037, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.28888888888888886, | |
| "grad_norm": 0.012265544384717941, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.0036, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.05352924019098282, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.0034, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.0031155734322965145, | |
| "eval_runtime": 397.2379, | |
| "eval_samples_per_second": 5.035, | |
| "eval_steps_per_second": 0.252, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.3111111111111111, | |
| "grad_norm": 0.02856474742293358, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.0029, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.32222222222222224, | |
| "grad_norm": 0.07519371062517166, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.0026, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 0.02294323220849037, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0032, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "eval_loss": 0.0026329518295824528, | |
| "eval_runtime": 394.8427, | |
| "eval_samples_per_second": 5.065, | |
| "eval_steps_per_second": 0.253, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.34444444444444444, | |
| "grad_norm": 0.013690228573977947, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 0.0029, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 0.02574954181909561, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.0023, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.36666666666666664, | |
| "grad_norm": 0.02459203265607357, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 0.0027, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.36666666666666664, | |
| "eval_loss": 0.0024726453702896833, | |
| "eval_runtime": 396.4992, | |
| "eval_samples_per_second": 5.044, | |
| "eval_steps_per_second": 0.252, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.37777777777777777, | |
| "grad_norm": 0.01764465682208538, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.0025, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.3888888888888889, | |
| "grad_norm": 0.013812178745865822, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.0022, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.03405527397990227, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.0022, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.002381704282015562, | |
| "eval_runtime": 396.4549, | |
| "eval_samples_per_second": 5.045, | |
| "eval_steps_per_second": 0.252, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.4111111111111111, | |
| "grad_norm": 0.020748196169734, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 0.0024, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.4222222222222222, | |
| "grad_norm": 0.013957368209958076, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.0025, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.43333333333333335, | |
| "grad_norm": 0.03190489485859871, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 0.002, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.43333333333333335, | |
| "eval_loss": 0.0022392040118575096, | |
| "eval_runtime": 395.9075, | |
| "eval_samples_per_second": 5.052, | |
| "eval_steps_per_second": 0.253, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 0.02455313876271248, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.0023, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.45555555555555555, | |
| "grad_norm": 0.012559180147945881, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.0023, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.4666666666666667, | |
| "grad_norm": 0.018574975430965424, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 0.0025, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.4666666666666667, | |
| "eval_loss": 0.002195958746597171, | |
| "eval_runtime": 400.9447, | |
| "eval_samples_per_second": 4.988, | |
| "eval_steps_per_second": 0.249, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.4777777777777778, | |
| "grad_norm": 0.035168472677469254, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 0.0023, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.4888888888888889, | |
| "grad_norm": 0.025580240413546562, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.002, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.018627658486366272, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0023, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.0020566468592733145, | |
| "eval_runtime": 395.5489, | |
| "eval_samples_per_second": 5.056, | |
| "eval_steps_per_second": 0.253, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.5111111111111111, | |
| "grad_norm": 0.025099800899624825, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 0.0022, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.5222222222222223, | |
| "grad_norm": 0.017753081396222115, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.002, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.5333333333333333, | |
| "grad_norm": 0.01829727366566658, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 0.0015, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.5333333333333333, | |
| "eval_loss": 0.0018324151169508696, | |
| "eval_runtime": 392.5068, | |
| "eval_samples_per_second": 5.095, | |
| "eval_steps_per_second": 0.255, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.5444444444444444, | |
| "grad_norm": 0.020837554708123207, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 0.0018, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 0.027814585715532303, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.002, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 900, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.222904495304868e+17, | |
| "train_batch_size": 10, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |