| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.028070175438596492, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0005614035087719298, | |
| "eval_loss": 7.9864583015441895, | |
| "eval_runtime": 79.61, | |
| "eval_samples_per_second": 9.421, | |
| "eval_steps_per_second": 4.71, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002807017543859649, | |
| "grad_norm": 8.250601768493652, | |
| "learning_rate": 5e-05, | |
| "loss": 8.2384, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.005614035087719298, | |
| "grad_norm": 19.211210250854492, | |
| "learning_rate": 0.0001, | |
| "loss": 5.4542, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.005614035087719298, | |
| "eval_loss": 1.5920395851135254, | |
| "eval_runtime": 81.4876, | |
| "eval_samples_per_second": 9.204, | |
| "eval_steps_per_second": 4.602, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008421052631578947, | |
| "grad_norm": 0.5418584942817688, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.4308, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.011228070175438596, | |
| "grad_norm": 0.024581190198659897, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.0007, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.011228070175438596, | |
| "eval_loss": 0.0002460002724546939, | |
| "eval_runtime": 80.9649, | |
| "eval_samples_per_second": 9.263, | |
| "eval_steps_per_second": 4.632, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.014035087719298246, | |
| "grad_norm": 0.005365356337279081, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.0002, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.016842105263157894, | |
| "grad_norm": 0.003963032737374306, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0001, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.016842105263157894, | |
| "eval_loss": 4.475672176340595e-05, | |
| "eval_runtime": 81.0386, | |
| "eval_samples_per_second": 9.255, | |
| "eval_steps_per_second": 4.627, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.019649122807017545, | |
| "grad_norm": 0.0038028114940971136, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.02245614035087719, | |
| "grad_norm": 0.005309837870299816, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.02245614035087719, | |
| "eval_loss": 2.7809644961962476e-05, | |
| "eval_runtime": 75.0644, | |
| "eval_samples_per_second": 9.991, | |
| "eval_steps_per_second": 4.996, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.02526315789473684, | |
| "grad_norm": 0.0019041873747482896, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.028070175438596492, | |
| "grad_norm": 0.0022254472132772207, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.028070175438596492, | |
| "eval_loss": 2.6173887818004005e-05, | |
| "eval_runtime": 75.5745, | |
| "eval_samples_per_second": 9.924, | |
| "eval_steps_per_second": 4.962, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9247821240729600.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |