| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.6118546845124283, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.030592734225621414, | |
| "grad_norm": 81.92265319824219, | |
| "learning_rate": 1e-05, | |
| "loss": 46.1719, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.030592734225621414, | |
| "eval_loss": 2.8948862552642822, | |
| "eval_runtime": 9.1276, | |
| "eval_samples_per_second": 6.026, | |
| "eval_steps_per_second": 3.068, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06118546845124283, | |
| "grad_norm": 80.7544174194336, | |
| "learning_rate": 2e-05, | |
| "loss": 43.8281, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.09177820267686425, | |
| "grad_norm": 82.4859390258789, | |
| "learning_rate": 3e-05, | |
| "loss": 45.8125, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.12237093690248566, | |
| "grad_norm": 74.56072998046875, | |
| "learning_rate": 4e-05, | |
| "loss": 44.6797, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.15296367112810708, | |
| "grad_norm": 82.29596710205078, | |
| "learning_rate": 5e-05, | |
| "loss": 45.2969, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.15296367112810708, | |
| "eval_loss": 2.674999952316284, | |
| "eval_runtime": 0.4404, | |
| "eval_samples_per_second": 124.898, | |
| "eval_steps_per_second": 63.585, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.1835564053537285, | |
| "grad_norm": 73.7712173461914, | |
| "learning_rate": 6e-05, | |
| "loss": 43.0312, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.21414913957934992, | |
| "grad_norm": 75.59838104248047, | |
| "learning_rate": 7e-05, | |
| "loss": 41.9766, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.2447418738049713, | |
| "grad_norm": 70.16816711425781, | |
| "learning_rate": 8e-05, | |
| "loss": 40.8359, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.27533460803059273, | |
| "grad_norm": 68.68514251708984, | |
| "learning_rate": 9e-05, | |
| "loss": 41.7578, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.30592734225621415, | |
| "grad_norm": 64.07447814941406, | |
| "learning_rate": 0.0001, | |
| "loss": 34.5, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.30592734225621415, | |
| "eval_loss": 2.095170497894287, | |
| "eval_runtime": 0.4386, | |
| "eval_samples_per_second": 125.394, | |
| "eval_steps_per_second": 63.837, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3365200764818356, | |
| "grad_norm": 64.02308654785156, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 34.875, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.367112810707457, | |
| "grad_norm": 63.0142707824707, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 31.0859, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.3977055449330784, | |
| "grad_norm": 74.16285705566406, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 29.8125, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.42829827915869984, | |
| "grad_norm": 61.39728927612305, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 26.4844, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.4588910133843212, | |
| "grad_norm": 63.88688659667969, | |
| "learning_rate": 5e-05, | |
| "loss": 22.7109, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.4588910133843212, | |
| "eval_loss": 1.4044034481048584, | |
| "eval_runtime": 0.465, | |
| "eval_samples_per_second": 118.289, | |
| "eval_steps_per_second": 60.22, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.4894837476099426, | |
| "grad_norm": 57.453948974609375, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 21.2578, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.5200764818355641, | |
| "grad_norm": 53.67185592651367, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 21.0703, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.5506692160611855, | |
| "grad_norm": 55.76949691772461, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 21.2773, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.5812619502868069, | |
| "grad_norm": 46.442745208740234, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 16.8008, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.6118546845124283, | |
| "grad_norm": 54.886348724365234, | |
| "learning_rate": 0.0, | |
| "loss": 20.1484, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.6118546845124283, | |
| "eval_loss": 1.225568175315857, | |
| "eval_runtime": 0.4457, | |
| "eval_samples_per_second": 123.411, | |
| "eval_steps_per_second": 62.827, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2390125141032960.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |