| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.6923076923076925, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 1.2540255213989835, | |
| "learning_rate": 4e-05, | |
| "loss": 2.461, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 1.2283767946925122, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3764, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "eval_loss": 1.8175537586212158, | |
| "eval_runtime": 2.6465, | |
| "eval_samples_per_second": 7.557, | |
| "eval_steps_per_second": 1.134, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 0.5222933366890941, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 1.6306, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "eval_loss": 1.2812074422836304, | |
| "eval_runtime": 1.3813, | |
| "eval_samples_per_second": 14.48, | |
| "eval_steps_per_second": 2.172, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 0.40273382098746935, | |
| "learning_rate": 0.0001766044443118978, | |
| "loss": 1.2054, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "eval_loss": 0.9901388883590698, | |
| "eval_runtime": 1.3581, | |
| "eval_samples_per_second": 14.726, | |
| "eval_steps_per_second": 2.209, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 0.5929366580267508, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.9679, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "eval_loss": 0.8994835019111633, | |
| "eval_runtime": 1.3574, | |
| "eval_samples_per_second": 14.734, | |
| "eval_steps_per_second": 2.21, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 0.34496621838110464, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 0.8649, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "eval_loss": 0.867924690246582, | |
| "eval_runtime": 1.355, | |
| "eval_samples_per_second": 14.76, | |
| "eval_steps_per_second": 2.214, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 0.2439958306613565, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 0.8062, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "eval_loss": 0.8346365690231323, | |
| "eval_runtime": 1.3475, | |
| "eval_samples_per_second": 14.843, | |
| "eval_steps_per_second": 2.226, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.384615384615385, | |
| "grad_norm": 0.4208553153399377, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 0.8035, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 5.384615384615385, | |
| "eval_loss": 0.8252410888671875, | |
| "eval_runtime": 1.3524, | |
| "eval_samples_per_second": 14.789, | |
| "eval_steps_per_second": 2.218, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 0.21047147413474968, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 0.7083, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "eval_loss": 0.8200315237045288, | |
| "eval_runtime": 1.3534, | |
| "eval_samples_per_second": 14.778, | |
| "eval_steps_per_second": 2.217, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.923076923076923, | |
| "grad_norm": 0.25331336485020006, | |
| "learning_rate": 6.030737921409169e-06, | |
| "loss": 0.703, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 6.923076923076923, | |
| "eval_loss": 0.8187538981437683, | |
| "eval_runtime": 1.348, | |
| "eval_samples_per_second": 14.837, | |
| "eval_steps_per_second": 2.226, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 0.2588396412267122, | |
| "learning_rate": 0.0, | |
| "loss": 0.6968, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "eval_loss": 0.8178739547729492, | |
| "eval_runtime": 1.3503, | |
| "eval_samples_per_second": 14.811, | |
| "eval_steps_per_second": 2.222, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "step": 50, | |
| "total_flos": 828551528448.0, | |
| "train_loss": 1.0779932928085327, | |
| "train_runtime": 175.8451, | |
| "train_samples_per_second": 4.549, | |
| "train_steps_per_second": 0.284 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 828551528448.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |