| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 10, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0016, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.0005397433415055275, | |
| "eval_runtime": 0.8141, | |
| "eval_samples_per_second": 1.228, | |
| "eval_steps_per_second": 1.228, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.07040310180048692, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 0.0014, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 9.640214557293802e-05, | |
| "eval_runtime": 1.4789, | |
| "eval_samples_per_second": 0.676, | |
| "eval_steps_per_second": 0.676, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.07040310180048692, | |
| "learning_rate": 0.0001766044443118978, | |
| "loss": 0.0001, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 9.640214557293802e-05, | |
| "eval_runtime": 0.6475, | |
| "eval_samples_per_second": 1.544, | |
| "eval_steps_per_second": 1.544, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.004378194380085889, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.0001, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 4.272457954357378e-05, | |
| "eval_runtime": 0.6471, | |
| "eval_samples_per_second": 1.545, | |
| "eval_steps_per_second": 1.545, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.004378194380085889, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 4.272457954357378e-05, | |
| "eval_runtime": 0.6411, | |
| "eval_samples_per_second": 1.56, | |
| "eval_steps_per_second": 1.56, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.00021419215401051602, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.9796718081342988e-05, | |
| "eval_runtime": 0.6523, | |
| "eval_samples_per_second": 1.533, | |
| "eval_steps_per_second": 1.533, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.00021419215401051602, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 2.9796718081342988e-05, | |
| "eval_runtime": 0.648, | |
| "eval_samples_per_second": 1.543, | |
| "eval_steps_per_second": 1.543, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.00011325198881217911, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.337109253858216e-05, | |
| "eval_runtime": 0.6473, | |
| "eval_samples_per_second": 1.545, | |
| "eval_steps_per_second": 1.545, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.00011325198881217911, | |
| "learning_rate": 6.030737921409169e-06, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 2.337109253858216e-05, | |
| "eval_runtime": 0.6455, | |
| "eval_samples_per_second": 1.549, | |
| "eval_steps_per_second": 1.549, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 8.057820187760094e-05, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 2.355557808186859e-05, | |
| "eval_runtime": 0.6631, | |
| "eval_samples_per_second": 1.508, | |
| "eval_steps_per_second": 1.508, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 10, | |
| "total_flos": 4718728839168.0, | |
| "train_loss": 0.0003110202618358926, | |
| "train_runtime": 115.7621, | |
| "train_samples_per_second": 0.777, | |
| "train_steps_per_second": 0.086 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 10, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4718728839168.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |