| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.436636818940658, | |
| "eval_steps": 750, | |
| "global_step": 98000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15176809834572771, | |
| "grad_norm": 1.347341537475586, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 2.0649, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.30353619669145543, | |
| "grad_norm": 1.299472689628601, | |
| "learning_rate": 1.984469638142569e-05, | |
| "loss": 1.7903, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.45530429503718317, | |
| "grad_norm": 1.5234107971191406, | |
| "learning_rate": 1.9534089144277063e-05, | |
| "loss": 1.6734, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6070723933829109, | |
| "grad_norm": 6.629986763000488, | |
| "learning_rate": 1.922348190712844e-05, | |
| "loss": 1.6014, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.7588404917286387, | |
| "grad_norm": 1.8827601671218872, | |
| "learning_rate": 1.8912874669979814e-05, | |
| "loss": 1.5582, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.9106085900743663, | |
| "grad_norm": 1.4713941812515259, | |
| "learning_rate": 1.8602267432831186e-05, | |
| "loss": 1.5393, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.4226312637329102, | |
| "eval_rouge1": 15.9571, | |
| "eval_rouge2": 5.6803, | |
| "eval_rougeL": 13.7071, | |
| "eval_rougeLsum": 13.8127, | |
| "eval_runtime": 167.9928, | |
| "eval_samples_per_second": 29.763, | |
| "eval_steps_per_second": 3.72, | |
| "step": 13178 | |
| }, | |
| { | |
| "epoch": 1.062376688420094, | |
| "grad_norm": 1.397419810295105, | |
| "learning_rate": 1.829166019568256e-05, | |
| "loss": 1.5108, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.214144786765822, | |
| "grad_norm": 1.282456398010254, | |
| "learning_rate": 1.7981052958533937e-05, | |
| "loss": 1.4931, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.3659128851115496, | |
| "grad_norm": 1.5143241882324219, | |
| "learning_rate": 1.767044572138531e-05, | |
| "loss": 1.4687, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.5176809834572773, | |
| "grad_norm": 1.0350826978683472, | |
| "learning_rate": 1.7359838484236684e-05, | |
| "loss": 1.4543, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.669449081803005, | |
| "grad_norm": 1.2214738130569458, | |
| "learning_rate": 1.704923124708806e-05, | |
| "loss": 1.4466, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.8212171801487327, | |
| "grad_norm": 1.226135492324829, | |
| "learning_rate": 1.6738624009939432e-05, | |
| "loss": 1.4314, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.9729852784944604, | |
| "grad_norm": 1.2347540855407715, | |
| "learning_rate": 1.6428016772790807e-05, | |
| "loss": 1.4255, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.3427051305770874, | |
| "eval_rouge1": 16.3039, | |
| "eval_rouge2": 6.0757, | |
| "eval_rougeL": 14.0031, | |
| "eval_rougeLsum": 14.1114, | |
| "eval_runtime": 168.2423, | |
| "eval_samples_per_second": 29.719, | |
| "eval_steps_per_second": 3.715, | |
| "step": 26356 | |
| }, | |
| { | |
| "epoch": 2.124753376840188, | |
| "grad_norm": 1.024688482284546, | |
| "learning_rate": 1.6117409535642183e-05, | |
| "loss": 1.4107, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.2765214751859157, | |
| "grad_norm": 1.1046956777572632, | |
| "learning_rate": 1.5806802298493555e-05, | |
| "loss": 1.3999, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.428289573531644, | |
| "grad_norm": 1.4118067026138306, | |
| "learning_rate": 1.549619506134493e-05, | |
| "loss": 1.394, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.5800576718773716, | |
| "grad_norm": 1.0044879913330078, | |
| "learning_rate": 1.5185587824196304e-05, | |
| "loss": 1.3894, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 2.7318257702230992, | |
| "grad_norm": 3.0287246704101562, | |
| "learning_rate": 1.4874980587047681e-05, | |
| "loss": 1.3823, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.883593868568827, | |
| "grad_norm": 0.8824607133865356, | |
| "learning_rate": 1.4564373349899055e-05, | |
| "loss": 1.3747, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.2928217649459839, | |
| "eval_rouge1": 16.2852, | |
| "eval_rouge2": 6.139, | |
| "eval_rougeL": 14.0119, | |
| "eval_rougeLsum": 14.1209, | |
| "eval_runtime": 172.2837, | |
| "eval_samples_per_second": 29.022, | |
| "eval_steps_per_second": 3.628, | |
| "step": 39534 | |
| }, | |
| { | |
| "epoch": 3.0353619669145546, | |
| "grad_norm": 1.7139147520065308, | |
| "learning_rate": 1.4253766112750429e-05, | |
| "loss": 1.365, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 3.1871300652602823, | |
| "grad_norm": 0.9839210510253906, | |
| "learning_rate": 1.3943158875601804e-05, | |
| "loss": 1.3631, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 3.33889816360601, | |
| "grad_norm": 2.022289514541626, | |
| "learning_rate": 1.3632551638453178e-05, | |
| "loss": 1.3497, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 3.4906662619517377, | |
| "grad_norm": 4.369687080383301, | |
| "learning_rate": 1.3321944401304551e-05, | |
| "loss": 1.3536, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 3.6424343602974654, | |
| "grad_norm": 1.046391487121582, | |
| "learning_rate": 1.3011337164155927e-05, | |
| "loss": 1.3455, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 3.794202458643193, | |
| "grad_norm": 3.8603522777557373, | |
| "learning_rate": 1.27007299270073e-05, | |
| "loss": 1.3396, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 3.9459705569889207, | |
| "grad_norm": 1.02574622631073, | |
| "learning_rate": 1.2390122689858674e-05, | |
| "loss": 1.3347, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.2622406482696533, | |
| "eval_rouge1": 16.4481, | |
| "eval_rouge2": 6.2714, | |
| "eval_rougeL": 14.1706, | |
| "eval_rougeLsum": 14.2806, | |
| "eval_runtime": 167.9003, | |
| "eval_samples_per_second": 29.78, | |
| "eval_steps_per_second": 3.722, | |
| "step": 52712 | |
| }, | |
| { | |
| "epoch": 4.097738655334648, | |
| "grad_norm": 1.366310715675354, | |
| "learning_rate": 1.2079515452710048e-05, | |
| "loss": 1.3304, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 4.249506753680376, | |
| "grad_norm": 1.1469073295593262, | |
| "learning_rate": 1.1768908215561424e-05, | |
| "loss": 1.3271, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 4.401274852026104, | |
| "grad_norm": 1.0787475109100342, | |
| "learning_rate": 1.1458300978412797e-05, | |
| "loss": 1.3199, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 4.5530429503718315, | |
| "grad_norm": 1.045688271522522, | |
| "learning_rate": 1.1147693741264171e-05, | |
| "loss": 1.3229, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 4.704811048717559, | |
| "grad_norm": 1.0128060579299927, | |
| "learning_rate": 1.0837086504115546e-05, | |
| "loss": 1.3156, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 4.856579147063288, | |
| "grad_norm": 1.1346766948699951, | |
| "learning_rate": 1.052647926696692e-05, | |
| "loss": 1.3186, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.23964262008667, | |
| "eval_rouge1": 16.5213, | |
| "eval_rouge2": 6.4307, | |
| "eval_rougeL": 14.289, | |
| "eval_rougeLsum": 14.3853, | |
| "eval_runtime": 170.4893, | |
| "eval_samples_per_second": 29.327, | |
| "eval_steps_per_second": 3.666, | |
| "step": 65890 | |
| }, | |
| { | |
| "epoch": 5.008347245409015, | |
| "grad_norm": 1.000510334968567, | |
| "learning_rate": 1.0215872029818294e-05, | |
| "loss": 1.3074, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 5.160115343754743, | |
| "grad_norm": 1.4083774089813232, | |
| "learning_rate": 9.90526479266967e-06, | |
| "loss": 1.3079, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 5.311883442100471, | |
| "grad_norm": 1.065021276473999, | |
| "learning_rate": 9.594657555521045e-06, | |
| "loss": 1.3055, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 5.4636515404461985, | |
| "grad_norm": 0.9340164065361023, | |
| "learning_rate": 9.284050318372419e-06, | |
| "loss": 1.305, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 5.615419638791926, | |
| "grad_norm": 0.9457820653915405, | |
| "learning_rate": 8.973443081223792e-06, | |
| "loss": 1.3015, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 5.767187737137654, | |
| "grad_norm": 0.8897130489349365, | |
| "learning_rate": 8.662835844075168e-06, | |
| "loss": 1.2985, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 5.9189558354833816, | |
| "grad_norm": 1.2775472402572632, | |
| "learning_rate": 8.352228606926543e-06, | |
| "loss": 1.2973, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.2244175672531128, | |
| "eval_rouge1": 16.4244, | |
| "eval_rouge2": 6.384, | |
| "eval_rougeL": 14.2167, | |
| "eval_rougeLsum": 14.3188, | |
| "eval_runtime": 170.4105, | |
| "eval_samples_per_second": 29.341, | |
| "eval_steps_per_second": 3.668, | |
| "step": 79068 | |
| }, | |
| { | |
| "epoch": 6.070723933829109, | |
| "grad_norm": 1.0759906768798828, | |
| "learning_rate": 8.041621369777917e-06, | |
| "loss": 1.2908, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 6.222492032174837, | |
| "grad_norm": 1.315941333770752, | |
| "learning_rate": 7.73101413262929e-06, | |
| "loss": 1.2927, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 6.374260130520565, | |
| "grad_norm": 0.936198353767395, | |
| "learning_rate": 7.420406895480665e-06, | |
| "loss": 1.2945, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 6.526028228866292, | |
| "grad_norm": 1.233934998512268, | |
| "learning_rate": 7.10979965833204e-06, | |
| "loss": 1.285, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 6.67779632721202, | |
| "grad_norm": 1.1760342121124268, | |
| "learning_rate": 6.7991924211834135e-06, | |
| "loss": 1.2877, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 6.829564425557748, | |
| "grad_norm": 2.0586724281311035, | |
| "learning_rate": 6.488585184034788e-06, | |
| "loss": 1.283, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 6.981332523903475, | |
| "grad_norm": 0.8292114734649658, | |
| "learning_rate": 6.1779779468861636e-06, | |
| "loss": 1.2817, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.21384859085083, | |
| "eval_rouge1": 16.5727, | |
| "eval_rouge2": 6.4685, | |
| "eval_rougeL": 14.3558, | |
| "eval_rougeLsum": 14.4703, | |
| "eval_runtime": 168.1885, | |
| "eval_samples_per_second": 29.729, | |
| "eval_steps_per_second": 3.716, | |
| "step": 92246 | |
| }, | |
| { | |
| "epoch": 7.133100622249203, | |
| "grad_norm": 0.9742059111595154, | |
| "learning_rate": 5.867370709737537e-06, | |
| "loss": 1.2858, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 7.284868720594931, | |
| "grad_norm": 1.0256426334381104, | |
| "learning_rate": 5.556763472588912e-06, | |
| "loss": 1.2804, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 7.436636818940658, | |
| "grad_norm": 1.0933358669281006, | |
| "learning_rate": 5.2461562354402865e-06, | |
| "loss": 1.2772, | |
| "step": 98000 | |
| } | |
| ], | |
| "logging_steps": 2000, | |
| "max_steps": 131780, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 1.0610228799838618e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |