| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7090909090909091, | |
| "eval_steps": 13, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01818181818181818, | |
| "eval_loss": 0.7239479422569275, | |
| "eval_runtime": 9.6694, | |
| "eval_samples_per_second": 9.618, | |
| "eval_steps_per_second": 1.241, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05454545454545454, | |
| "grad_norm": 1.006616234779358, | |
| "learning_rate": 3e-05, | |
| "loss": 0.7497, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.10909090909090909, | |
| "grad_norm": 0.7354311347007751, | |
| "learning_rate": 6e-05, | |
| "loss": 0.6608, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.16363636363636364, | |
| "grad_norm": 0.5442305207252502, | |
| "learning_rate": 9e-05, | |
| "loss": 0.5711, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.21818181818181817, | |
| "grad_norm": 0.4888646900653839, | |
| "learning_rate": 9.994965332706573e-05, | |
| "loss": 0.6065, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.23636363636363636, | |
| "eval_loss": 0.5811941027641296, | |
| "eval_runtime": 9.3174, | |
| "eval_samples_per_second": 9.981, | |
| "eval_steps_per_second": 1.288, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.2727272727272727, | |
| "grad_norm": 0.43622222542762756, | |
| "learning_rate": 9.968561049466214e-05, | |
| "loss": 0.545, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.32727272727272727, | |
| "grad_norm": 0.4785720407962799, | |
| "learning_rate": 9.919647942993148e-05, | |
| "loss": 0.552, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.38181818181818183, | |
| "grad_norm": 0.5042471289634705, | |
| "learning_rate": 9.848447601883435e-05, | |
| "loss": 0.5068, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.43636363636363634, | |
| "grad_norm": 0.5013201832771301, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.5388, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.4727272727272727, | |
| "eval_loss": 0.4993453621864319, | |
| "eval_runtime": 9.3279, | |
| "eval_samples_per_second": 9.97, | |
| "eval_steps_per_second": 1.286, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.4909090909090909, | |
| "grad_norm": 0.45883774757385254, | |
| "learning_rate": 9.640574942595196e-05, | |
| "loss": 0.4867, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "grad_norm": 0.546485185623169, | |
| "learning_rate": 9.504844339512095e-05, | |
| "loss": 0.4773, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.5516665577888489, | |
| "learning_rate": 9.348705665778478e-05, | |
| "loss": 0.4846, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.6545454545454545, | |
| "grad_norm": 0.5257235765457153, | |
| "learning_rate": 9.172866268606513e-05, | |
| "loss": 0.4445, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.7090909090909091, | |
| "grad_norm": 0.5515545010566711, | |
| "learning_rate": 8.978122744408906e-05, | |
| "loss": 0.4303, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.7090909090909091, | |
| "eval_loss": 0.41172003746032715, | |
| "eval_runtime": 9.3482, | |
| "eval_samples_per_second": 9.948, | |
| "eval_steps_per_second": 1.284, | |
| "step": 39 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 150, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.711665917257318e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |