| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9627085377821394, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03925417075564279, | |
| "grad_norm": 81.5, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 0.3818, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07850834151128558, | |
| "grad_norm": 322.0, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "loss": 0.367, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11776251226692837, | |
| "grad_norm": 784.0, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 0.4472, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.15701668302257116, | |
| "grad_norm": 350.0, | |
| "learning_rate": 3.2000000000000003e-06, | |
| "loss": 0.6806, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.19627085377821393, | |
| "grad_norm": 186.0, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.9139, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.23552502453385674, | |
| "grad_norm": 199.0, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 0.8205, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2747791952894995, | |
| "grad_norm": 118.0, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 0.9017, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.3140333660451423, | |
| "grad_norm": 616.0, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 3.1966, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.35328753680078506, | |
| "grad_norm": 692.0, | |
| "learning_rate": 7.2000000000000005e-06, | |
| "loss": 1.0627, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.39254170755642787, | |
| "grad_norm": 75.5, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 7.3782, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.43179587831207067, | |
| "grad_norm": 908.0, | |
| "learning_rate": 8.8e-06, | |
| "loss": 1.1091, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.47105004906771347, | |
| "grad_norm": 61.5, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 0.9262, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5103042198233563, | |
| "grad_norm": 45.25, | |
| "learning_rate": 1.04e-05, | |
| "loss": 0.6774, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.549558390578999, | |
| "grad_norm": 39.75, | |
| "learning_rate": 1.1200000000000001e-05, | |
| "loss": 0.8204, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.5888125613346418, | |
| "grad_norm": 38.0, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.6443, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6280667320902846, | |
| "grad_norm": 31.25, | |
| "learning_rate": 1.2800000000000001e-05, | |
| "loss": 0.607, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.6673209028459274, | |
| "grad_norm": 33.5, | |
| "learning_rate": 1.3600000000000002e-05, | |
| "loss": 0.674, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.7065750736015701, | |
| "grad_norm": 37.5, | |
| "learning_rate": 1.4400000000000001e-05, | |
| "loss": 0.6806, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.745829244357213, | |
| "grad_norm": 36.25, | |
| "learning_rate": 1.5200000000000002e-05, | |
| "loss": 0.6131, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.7850834151128557, | |
| "grad_norm": 43.25, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.5914, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.8243375858684985, | |
| "grad_norm": 27.75, | |
| "learning_rate": 1.6800000000000002e-05, | |
| "loss": 0.5488, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.8635917566241413, | |
| "grad_norm": 47.75, | |
| "learning_rate": 1.76e-05, | |
| "loss": 0.7107, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.9028459273797841, | |
| "grad_norm": 16.375, | |
| "learning_rate": 1.8400000000000003e-05, | |
| "loss": 1.2491, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.9421000981354269, | |
| "grad_norm": 35.5, | |
| "learning_rate": 1.9200000000000003e-05, | |
| "loss": 0.6448, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.9813542688910697, | |
| "grad_norm": 22.875, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5682, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.0206084396467126, | |
| "grad_norm": 32.75, | |
| "learning_rate": 1.9999025240093045e-05, | |
| "loss": 0.5689, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.0598626104023552, | |
| "grad_norm": 28.375, | |
| "learning_rate": 1.9996101150403543e-05, | |
| "loss": 0.4836, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.099116781157998, | |
| "grad_norm": 18.25, | |
| "learning_rate": 1.9991228300988586e-05, | |
| "loss": 0.519, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.138370951913641, | |
| "grad_norm": 26.125, | |
| "learning_rate": 1.9984407641819812e-05, | |
| "loss": 0.4941, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.1776251226692835, | |
| "grad_norm": 20.5, | |
| "learning_rate": 1.9975640502598243e-05, | |
| "loss": 0.4484, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.2168792934249264, | |
| "grad_norm": 24.625, | |
| "learning_rate": 1.9964928592495046e-05, | |
| "loss": 0.5169, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.2561334641805693, | |
| "grad_norm": 14.125, | |
| "learning_rate": 1.9952273999818312e-05, | |
| "loss": 0.4417, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.295387634936212, | |
| "grad_norm": 26.25, | |
| "learning_rate": 1.9937679191605964e-05, | |
| "loss": 0.4361, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.3346418056918548, | |
| "grad_norm": 23.125, | |
| "learning_rate": 1.9921147013144782e-05, | |
| "loss": 0.4454, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.3738959764474976, | |
| "grad_norm": 23.625, | |
| "learning_rate": 1.9902680687415704e-05, | |
| "loss": 0.446, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.4131501472031402, | |
| "grad_norm": 13.9375, | |
| "learning_rate": 1.988228381446553e-05, | |
| "loss": 0.4006, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.452404317958783, | |
| "grad_norm": 31.5, | |
| "learning_rate": 1.985996037070505e-05, | |
| "loss": 0.4577, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.491658488714426, | |
| "grad_norm": 47.5, | |
| "learning_rate": 1.983571470813386e-05, | |
| "loss": 0.4207, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.5309126594700686, | |
| "grad_norm": 12.3125, | |
| "learning_rate": 1.9809551553491918e-05, | |
| "loss": 0.4251, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.5701668302257115, | |
| "grad_norm": 20.75, | |
| "learning_rate": 1.9781476007338058e-05, | |
| "loss": 0.3968, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.6094210009813543, | |
| "grad_norm": 15.0, | |
| "learning_rate": 1.9751493543055634e-05, | |
| "loss": 0.4019, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.648675171736997, | |
| "grad_norm": 11.5, | |
| "learning_rate": 1.9719610005785466e-05, | |
| "loss": 0.3894, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.6879293424926398, | |
| "grad_norm": 20.375, | |
| "learning_rate": 1.9685831611286312e-05, | |
| "loss": 0.3923, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.7271835132482827, | |
| "grad_norm": 16.75, | |
| "learning_rate": 1.9650164944723116e-05, | |
| "loss": 0.374, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.7664376840039253, | |
| "grad_norm": 21.375, | |
| "learning_rate": 1.961261695938319e-05, | |
| "loss": 0.402, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.8056918547595682, | |
| "grad_norm": 14.25, | |
| "learning_rate": 1.9573194975320672e-05, | |
| "loss": 0.3646, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.844946025515211, | |
| "grad_norm": 15.0625, | |
| "learning_rate": 1.9531906677929472e-05, | |
| "loss": 0.3451, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.8842001962708537, | |
| "grad_norm": 12.75, | |
| "learning_rate": 1.9488760116444966e-05, | |
| "loss": 0.3373, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.9234543670264965, | |
| "grad_norm": 12.5, | |
| "learning_rate": 1.944376370237481e-05, | |
| "loss": 0.3693, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.9627085377821394, | |
| "grad_norm": 2.34375, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.3595, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2500, | |
| "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": 1.91328310788096e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |