| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9979315099977017, | |
| "eval_steps": 500, | |
| "global_step": 3261, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09193288899103655, | |
| "grad_norm": 0.15536943078041077, | |
| "learning_rate": 0.00019988036932954271, | |
| "loss": 0.8236, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1838657779820731, | |
| "grad_norm": 0.16779352724552155, | |
| "learning_rate": 0.00019892504066072438, | |
| "loss": 0.6327, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2757986669731096, | |
| "grad_norm": 0.21519789099693298, | |
| "learning_rate": 0.0001970235207829469, | |
| "loss": 0.5599, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.3677315559641462, | |
| "grad_norm": 0.1756363958120346, | |
| "learning_rate": 0.0001941939972186009, | |
| "loss": 0.5278, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.4596644449551827, | |
| "grad_norm": 0.19388951361179352, | |
| "learning_rate": 0.00019046353359388504, | |
| "loss": 0.4978, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5515973339462192, | |
| "grad_norm": 0.2049034982919693, | |
| "learning_rate": 0.00018586781078255458, | |
| "loss": 0.4743, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.6435302229372558, | |
| "grad_norm": 0.1463390439748764, | |
| "learning_rate": 0.00018045078562803203, | |
| "loss": 0.4456, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.7354631119282924, | |
| "grad_norm": 0.19917741417884827, | |
| "learning_rate": 0.0001742642705081106, | |
| "loss": 0.4407, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.8273960009193289, | |
| "grad_norm": 0.19318071007728577, | |
| "learning_rate": 0.00016736743776359978, | |
| "loss": 0.421, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.9193288899103654, | |
| "grad_norm": 0.1719323843717575, | |
| "learning_rate": 0.00015982625373091875, | |
| "loss": 0.409, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.011261778901402, | |
| "grad_norm": 0.2220071256160736, | |
| "learning_rate": 0.00015171284779196334, | |
| "loss": 0.3985, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.1031946678924385, | |
| "grad_norm": 0.14137883484363556, | |
| "learning_rate": 0.00014310482247611208, | |
| "loss": 0.3805, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.1951275568834752, | |
| "grad_norm": 0.14456775784492493, | |
| "learning_rate": 0.00013408451121306046, | |
| "loss": 0.3747, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.2870604458745116, | |
| "grad_norm": 0.1589755117893219, | |
| "learning_rate": 0.0001247381908358749, | |
| "loss": 0.3716, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.378993334865548, | |
| "grad_norm": 0.20999275147914886, | |
| "learning_rate": 0.00011515525636646231, | |
| "loss": 0.3618, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.4709262238565848, | |
| "grad_norm": 0.15207038819789886, | |
| "learning_rate": 0.00010542736597640826, | |
| "loss": 0.3595, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.5628591128476212, | |
| "grad_norm": 0.13484518229961395, | |
| "learning_rate": 9.564756430140164e-05, | |
| "loss": 0.3581, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.6547920018386577, | |
| "grad_norm": 0.1453435719013214, | |
| "learning_rate": 8.590939249450595e-05, | |
| "loss": 0.3561, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.7467248908296944, | |
| "grad_norm": 0.16422341763973236, | |
| "learning_rate": 7.630599353037633e-05, | |
| "loss": 0.3564, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.8386577798207309, | |
| "grad_norm": 0.13742196559906006, | |
| "learning_rate": 6.692922131794517e-05, | |
| "loss": 0.353, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.9305906688117673, | |
| "grad_norm": 0.1561811864376068, | |
| "learning_rate": 5.7868762142672204e-05, | |
| "loss": 0.3531, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.022523557802804, | |
| "grad_norm": 0.13941729068756104, | |
| "learning_rate": 4.9211276841525744e-05, | |
| "loss": 0.3438, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.1144564467938407, | |
| "grad_norm": 0.13483913242816925, | |
| "learning_rate": 4.103957191555876e-05, | |
| "loss": 0.339, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.206389335784877, | |
| "grad_norm": 0.16717399656772614, | |
| "learning_rate": 3.343180750816377e-05, | |
| "loss": 0.3386, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.2983222247759136, | |
| "grad_norm": 0.14586015045642853, | |
| "learning_rate": 2.6460749824479912e-05, | |
| "loss": 0.3313, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.3902551137669503, | |
| "grad_norm": 0.14884348213672638, | |
| "learning_rate": 2.019307514235741e-05, | |
| "loss": 0.3344, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.4821880027579866, | |
| "grad_norm": 0.15428389608860016, | |
| "learning_rate": 1.4688732071827094e-05, | |
| "loss": 0.3321, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.5741208917490233, | |
| "grad_norm": 0.1362064927816391, | |
| "learning_rate": 1.0000368162888795e-05, | |
| "loss": 0.3364, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.66605378074006, | |
| "grad_norm": 0.1446692794561386, | |
| "learning_rate": 6.1728263459614796e-06, | |
| "loss": 0.3364, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.757986669731096, | |
| "grad_norm": 0.14172622561454773, | |
| "learning_rate": 3.2427160214043793e-06, | |
| "loss": 0.3358, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.849919558722133, | |
| "grad_norm": 0.14800885319709778, | |
| "learning_rate": 1.2380629005222855e-06, | |
| "loss": 0.3341, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.9418524477131696, | |
| "grad_norm": 0.1342545449733734, | |
| "learning_rate": 1.7804094723098408e-07, | |
| "loss": 0.3333, | |
| "step": 3200 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 3261, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.206758990180516e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |