| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 1000, | |
| "global_step": 2184, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.045787545787545784, | |
| "grad_norm": 8.169079780578613, | |
| "learning_rate": 1.9896720173943657e-05, | |
| "loss": 0.8583, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.09157509157509157, | |
| "grad_norm": 22.747098922729492, | |
| "learning_rate": 1.9589014040268678e-05, | |
| "loss": 0.759, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.13736263736263737, | |
| "grad_norm": 3.93502140045166, | |
| "learning_rate": 1.908323756616754e-05, | |
| "loss": 0.7104, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.18315018315018314, | |
| "grad_norm": 10.182877540588379, | |
| "learning_rate": 1.838983805289396e-05, | |
| "loss": 0.7041, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.22893772893772893, | |
| "grad_norm": 2.300295352935791, | |
| "learning_rate": 1.7523138336671628e-05, | |
| "loss": 0.7295, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.27472527472527475, | |
| "grad_norm": 1.7253031730651855, | |
| "learning_rate": 1.6501040936687444e-05, | |
| "loss": 0.7295, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.32051282051282054, | |
| "grad_norm": 1.2007420063018799, | |
| "learning_rate": 1.5344658261278013e-05, | |
| "loss": 0.6888, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.3663003663003663, | |
| "grad_norm": 2.432363986968994, | |
| "learning_rate": 1.4077876510757502e-05, | |
| "loss": 0.6686, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.41208791208791207, | |
| "grad_norm": 1.7616468667984009, | |
| "learning_rate": 1.2726862284894939e-05, | |
| "loss": 0.779, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.45787545787545786, | |
| "grad_norm": 4.531213760375977, | |
| "learning_rate": 1.1319522086539666e-05, | |
| "loss": 0.711, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.45787545787545786, | |
| "eval_accuracy": 0.6213872832369942, | |
| "eval_loss": 0.7638348937034607, | |
| "eval_runtime": 12.3349, | |
| "eval_samples_per_second": 28.051, | |
| "eval_steps_per_second": 9.404, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5036630036630036, | |
| "grad_norm": 4.445523738861084, | |
| "learning_rate": 9.884925885869326e-06, | |
| "loss": 0.6899, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.5494505494505495, | |
| "grad_norm": 5.363688945770264, | |
| "learning_rate": 8.452706652097187e-06, | |
| "loss": 0.6949, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.5952380952380952, | |
| "grad_norm": 11.651291847229004, | |
| "learning_rate": 7.052448255890958e-06, | |
| "loss": 0.6604, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.6410256410256411, | |
| "grad_norm": 10.71049690246582, | |
| "learning_rate": 5.713074385969457e-06, | |
| "loss": 0.667, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.6868131868131868, | |
| "grad_norm": 5.3541436195373535, | |
| "learning_rate": 4.462251102394669e-06, | |
| "loss": 0.6748, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.7326007326007326, | |
| "grad_norm": 7.145966529846191, | |
| "learning_rate": 3.325815367397557e-06, | |
| "loss": 0.665, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.7783882783882784, | |
| "grad_norm": 6.2031168937683105, | |
| "learning_rate": 2.327241357985063e-06, | |
| "loss": 0.735, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.8241758241758241, | |
| "grad_norm": 3.511324882507324, | |
| "learning_rate": 1.4871555841564889e-06, | |
| "loss": 0.6776, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.86996336996337, | |
| "grad_norm": 5.494836330413818, | |
| "learning_rate": 8.229108284305176e-07, | |
| "loss": 0.6689, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.9157509157509157, | |
| "grad_norm": 5.310685157775879, | |
| "learning_rate": 3.482277073731988e-07, | |
| "loss": 0.6407, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9157509157509157, | |
| "eval_accuracy": 0.6329479768786127, | |
| "eval_loss": 0.6654821634292603, | |
| "eval_runtime": 12.2552, | |
| "eval_samples_per_second": 28.233, | |
| "eval_steps_per_second": 9.465, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9615384615384616, | |
| "grad_norm": 11.22006893157959, | |
| "learning_rate": 7.291125901946027e-08, | |
| "loss": 0.6801, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 2184, | |
| "total_flos": 0.0, | |
| "train_loss": 0.7032692458603408, | |
| "train_runtime": 2584.7937, | |
| "train_samples_per_second": 2.534, | |
| "train_steps_per_second": 0.845 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 2184, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 0.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |