| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.616088947024199, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.052321778940483975, | |
| "grad_norm": 50.25, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 3.2827, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.10464355788096795, | |
| "grad_norm": 282.0, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "loss": 3.2733, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.15696533682145192, | |
| "grad_norm": 390.0, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 3.1571, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2092871157619359, | |
| "grad_norm": 322.0, | |
| "learning_rate": 3.2000000000000003e-06, | |
| "loss": 3.393, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2616088947024199, | |
| "grad_norm": 700.0, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 3.4911, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.31393067364290383, | |
| "grad_norm": 1704.0, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 3.9171, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.36625245258338784, | |
| "grad_norm": 214.0, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 4.6157, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.4185742315238718, | |
| "grad_norm": 1608.0, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 4.779, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.4708960104643558, | |
| "grad_norm": 162.0, | |
| "learning_rate": 7.2000000000000005e-06, | |
| "loss": 4.2508, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5232177894048398, | |
| "grad_norm": 149.0, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 4.6611, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5755395683453237, | |
| "grad_norm": 214.0, | |
| "learning_rate": 8.8e-06, | |
| "loss": 3.9426, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.6278613472858077, | |
| "grad_norm": 408.0, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 3.7952, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.6801831262262917, | |
| "grad_norm": 71.0, | |
| "learning_rate": 1.04e-05, | |
| "loss": 3.647, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.7325049051667757, | |
| "grad_norm": 59.0, | |
| "learning_rate": 1.1200000000000001e-05, | |
| "loss": 3.6275, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.7848266841072596, | |
| "grad_norm": 45.25, | |
| "learning_rate": 1.2e-05, | |
| "loss": 3.4353, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8371484630477436, | |
| "grad_norm": 47.0, | |
| "learning_rate": 1.2800000000000001e-05, | |
| "loss": 3.69, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.8894702419882276, | |
| "grad_norm": 33.25, | |
| "learning_rate": 1.3600000000000002e-05, | |
| "loss": 3.6108, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.9417920209287116, | |
| "grad_norm": 134.0, | |
| "learning_rate": 1.4400000000000001e-05, | |
| "loss": 3.7036, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.9941137998691956, | |
| "grad_norm": 45.25, | |
| "learning_rate": 1.5200000000000002e-05, | |
| "loss": 3.4621, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.0464355788096795, | |
| "grad_norm": 48.25, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 3.4607, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0987573577501635, | |
| "grad_norm": 42.0, | |
| "learning_rate": 1.6800000000000002e-05, | |
| "loss": 18.2652, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.1510791366906474, | |
| "grad_norm": 41.0, | |
| "learning_rate": 1.76e-05, | |
| "loss": 3.6676, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.2034009156311314, | |
| "grad_norm": 25.375, | |
| "learning_rate": 1.8400000000000003e-05, | |
| "loss": 3.3636, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.2557226945716153, | |
| "grad_norm": 35.25, | |
| "learning_rate": 1.9200000000000003e-05, | |
| "loss": 3.1784, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.3080444735120995, | |
| "grad_norm": 42.0, | |
| "learning_rate": 2e-05, | |
| "loss": 3.7673, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.3603662524525835, | |
| "grad_norm": 32.75, | |
| "learning_rate": 1.9999025240093045e-05, | |
| "loss": 3.7099, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.4126880313930674, | |
| "grad_norm": 27.625, | |
| "learning_rate": 1.9996101150403543e-05, | |
| "loss": 3.3748, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.4650098103335514, | |
| "grad_norm": 24.375, | |
| "learning_rate": 1.9991228300988586e-05, | |
| "loss": 3.1567, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.5173315892740353, | |
| "grad_norm": 26.625, | |
| "learning_rate": 1.9984407641819812e-05, | |
| "loss": 3.1535, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.5696533682145193, | |
| "grad_norm": 20.125, | |
| "learning_rate": 1.9975640502598243e-05, | |
| "loss": 3.0614, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.6219751471550032, | |
| "grad_norm": 23.125, | |
| "learning_rate": 1.9964928592495046e-05, | |
| "loss": 3.071, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.6742969260954872, | |
| "grad_norm": 32.25, | |
| "learning_rate": 1.9952273999818312e-05, | |
| "loss": 3.1193, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.7266187050359711, | |
| "grad_norm": 19.375, | |
| "learning_rate": 1.9937679191605964e-05, | |
| "loss": 3.0596, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.7789404839764553, | |
| "grad_norm": 36.75, | |
| "learning_rate": 1.9921147013144782e-05, | |
| "loss": 2.9772, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.831262262916939, | |
| "grad_norm": 27.25, | |
| "learning_rate": 1.9902680687415704e-05, | |
| "loss": 3.0119, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.8835840418574232, | |
| "grad_norm": 26.375, | |
| "learning_rate": 1.988228381446553e-05, | |
| "loss": 2.8278, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.9359058207979072, | |
| "grad_norm": 27.75, | |
| "learning_rate": 1.985996037070505e-05, | |
| "loss": 2.6952, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.9882275997383911, | |
| "grad_norm": 27.5, | |
| "learning_rate": 1.983571470813386e-05, | |
| "loss": 2.5663, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.040549378678875, | |
| "grad_norm": 27.375, | |
| "learning_rate": 1.9809551553491918e-05, | |
| "loss": 2.2019, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.092871157619359, | |
| "grad_norm": 15.75, | |
| "learning_rate": 1.9781476007338058e-05, | |
| "loss": 2.2746, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.145192936559843, | |
| "grad_norm": 38.0, | |
| "learning_rate": 1.9751493543055634e-05, | |
| "loss": 2.4245, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.197514715500327, | |
| "grad_norm": 30.125, | |
| "learning_rate": 1.9719610005785466e-05, | |
| "loss": 2.1032, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.249836494440811, | |
| "grad_norm": 18.5, | |
| "learning_rate": 1.9685831611286312e-05, | |
| "loss": 2.3059, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.302158273381295, | |
| "grad_norm": 26.5, | |
| "learning_rate": 1.9650164944723116e-05, | |
| "loss": 2.025, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.354480052321779, | |
| "grad_norm": 18.625, | |
| "learning_rate": 1.961261695938319e-05, | |
| "loss": 2.0708, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.4068018312622628, | |
| "grad_norm": 28.25, | |
| "learning_rate": 1.9573194975320672e-05, | |
| "loss": 1.926, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.459123610202747, | |
| "grad_norm": 24.75, | |
| "learning_rate": 1.9531906677929472e-05, | |
| "loss": 1.9348, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.5114453891432307, | |
| "grad_norm": 48.5, | |
| "learning_rate": 1.9488760116444966e-05, | |
| "loss": 1.8206, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.563767168083715, | |
| "grad_norm": 19.875, | |
| "learning_rate": 1.944376370237481e-05, | |
| "loss": 1.8137, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.616088947024199, | |
| "grad_norm": 25.125, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 1.8683, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 14, | |
| "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": 1.9128047871039898e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |