| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 60.0, | |
| "eval_steps": 8874, | |
| "global_step": 88740, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.8650519847869873, | |
| "learning_rate": 9.001014198782962e-06, | |
| "loss": 2.4122, | |
| "step": 8874 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 3.9393680095672607, | |
| "eval_runtime": 31.1663, | |
| "eval_samples_per_second": 311.715, | |
| "eval_steps_per_second": 7.091, | |
| "step": 8874 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 11.994268417358398, | |
| "learning_rate": 8.001239576290288e-06, | |
| "loss": 2.2167, | |
| "step": 17748 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 4.1646857261657715, | |
| "eval_runtime": 31.5415, | |
| "eval_samples_per_second": 308.007, | |
| "eval_steps_per_second": 7.007, | |
| "step": 17748 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 20.35555076599121, | |
| "learning_rate": 7.001690331304937e-06, | |
| "loss": 2.0965, | |
| "step": 26622 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 4.430016994476318, | |
| "eval_runtime": 31.5349, | |
| "eval_samples_per_second": 308.071, | |
| "eval_steps_per_second": 7.008, | |
| "step": 26622 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 3.9490978717803955, | |
| "learning_rate": 6.002141086319586e-06, | |
| "loss": 2.0238, | |
| "step": 35496 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 4.574044227600098, | |
| "eval_runtime": 31.6749, | |
| "eval_samples_per_second": 306.709, | |
| "eval_steps_per_second": 6.977, | |
| "step": 35496 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 9.400227546691895, | |
| "learning_rate": 5.0025918413342355e-06, | |
| "loss": 1.9938, | |
| "step": 44370 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 4.62649393081665, | |
| "eval_runtime": 31.7226, | |
| "eval_samples_per_second": 306.249, | |
| "eval_steps_per_second": 6.967, | |
| "step": 44370 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "grad_norm": 5.468742370605469, | |
| "learning_rate": 4.003042596348885e-06, | |
| "loss": 1.973, | |
| "step": 53244 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_loss": 4.671382427215576, | |
| "eval_runtime": 31.6041, | |
| "eval_samples_per_second": 307.396, | |
| "eval_steps_per_second": 6.993, | |
| "step": 53244 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "grad_norm": 0.6607534289360046, | |
| "learning_rate": 3.003606040117197e-06, | |
| "loss": 1.9583, | |
| "step": 62118 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_loss": 4.793060302734375, | |
| "eval_runtime": 31.7099, | |
| "eval_samples_per_second": 306.372, | |
| "eval_steps_per_second": 6.969, | |
| "step": 62118 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "grad_norm": 1.4975688457489014, | |
| "learning_rate": 2.004056795131846e-06, | |
| "loss": 1.9466, | |
| "step": 70992 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_loss": 4.7913103103637695, | |
| "eval_runtime": 33.0493, | |
| "eval_samples_per_second": 293.955, | |
| "eval_steps_per_second": 6.687, | |
| "step": 70992 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "grad_norm": 2.9576919078826904, | |
| "learning_rate": 1.0045075501464953e-06, | |
| "loss": 1.9415, | |
| "step": 79866 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_loss": 4.844839572906494, | |
| "eval_runtime": 31.7934, | |
| "eval_samples_per_second": 305.566, | |
| "eval_steps_per_second": 6.951, | |
| "step": 79866 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 0.5801959037780762, | |
| "learning_rate": 5.070993914807302e-09, | |
| "loss": 1.9369, | |
| "step": 88740 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_loss": 4.886190891265869, | |
| "eval_runtime": 31.6305, | |
| "eval_samples_per_second": 307.141, | |
| "eval_steps_per_second": 6.987, | |
| "step": 88740 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "step": 88740, | |
| "total_flos": 1.4015777943683174e+18, | |
| "train_loss": 2.04994026615675, | |
| "train_runtime": 30910.218, | |
| "train_samples_per_second": 114.776, | |
| "train_steps_per_second": 2.871 | |
| } | |
| ], | |
| "logging_steps": 8874, | |
| "max_steps": 88740, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 60, | |
| "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": 1.4015777943683174e+18, | |
| "train_batch_size": 40, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |