| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.42283298097251587, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.008456659619450317, | |
| "eval_loss": 1.399917721748352, | |
| "eval_runtime": 15.6201, | |
| "eval_samples_per_second": 12.74, | |
| "eval_steps_per_second": 1.601, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02536997885835095, | |
| "grad_norm": 9.886091232299805, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3114, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.042283298097251586, | |
| "eval_loss": 0.19794023036956787, | |
| "eval_runtime": 15.2884, | |
| "eval_samples_per_second": 13.016, | |
| "eval_steps_per_second": 1.635, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0507399577167019, | |
| "grad_norm": 1.1028578281402588, | |
| "learning_rate": 6e-05, | |
| "loss": 0.4389, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.07610993657505286, | |
| "grad_norm": 1.941910743713379, | |
| "learning_rate": 9e-05, | |
| "loss": 0.1821, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.08456659619450317, | |
| "eval_loss": 0.17988403141498566, | |
| "eval_runtime": 15.3453, | |
| "eval_samples_per_second": 12.968, | |
| "eval_steps_per_second": 1.629, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1014799154334038, | |
| "grad_norm": 0.35268205404281616, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 0.1627, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.12684989429175475, | |
| "grad_norm": 0.3946090042591095, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.1838, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.12684989429175475, | |
| "eval_loss": 0.17440293729305267, | |
| "eval_runtime": 15.3905, | |
| "eval_samples_per_second": 12.93, | |
| "eval_steps_per_second": 1.624, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1522198731501057, | |
| "grad_norm": 0.7058480978012085, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.1601, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.16913319238900634, | |
| "eval_loss": 0.15480488538742065, | |
| "eval_runtime": 15.3939, | |
| "eval_samples_per_second": 12.927, | |
| "eval_steps_per_second": 1.624, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.17758985200845667, | |
| "grad_norm": 0.1734049916267395, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.1484, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.2029598308668076, | |
| "grad_norm": 2.5953726768493652, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.1632, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.21141649048625794, | |
| "eval_loss": 0.1542930006980896, | |
| "eval_runtime": 15.404, | |
| "eval_samples_per_second": 12.919, | |
| "eval_steps_per_second": 1.623, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.22832980972515857, | |
| "grad_norm": 0.17785046994686127, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.1548, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.2536997885835095, | |
| "grad_norm": 0.15654253959655762, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1491, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2536997885835095, | |
| "eval_loss": 0.1547684371471405, | |
| "eval_runtime": 15.4124, | |
| "eval_samples_per_second": 12.912, | |
| "eval_steps_per_second": 1.622, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.27906976744186046, | |
| "grad_norm": 0.9842440485954285, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.1461, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.2959830866807611, | |
| "eval_loss": 0.15374921262264252, | |
| "eval_runtime": 15.4033, | |
| "eval_samples_per_second": 12.919, | |
| "eval_steps_per_second": 1.623, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.3044397463002114, | |
| "grad_norm": 0.3132956624031067, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.1481, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.3298097251585624, | |
| "grad_norm": 0.3042653501033783, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.1665, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.3382663847780127, | |
| "eval_loss": 0.15339957177639008, | |
| "eval_runtime": 15.4031, | |
| "eval_samples_per_second": 12.92, | |
| "eval_steps_per_second": 1.623, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.35517970401691334, | |
| "grad_norm": 1.0083894729614258, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.1557, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.38054968287526425, | |
| "grad_norm": 0.3310617506504059, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.1498, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.38054968287526425, | |
| "eval_loss": 0.15380239486694336, | |
| "eval_runtime": 15.4088, | |
| "eval_samples_per_second": 12.915, | |
| "eval_steps_per_second": 1.622, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.4059196617336152, | |
| "grad_norm": 0.22873643040657043, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.1727, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.42283298097251587, | |
| "eval_loss": 0.15397053956985474, | |
| "eval_runtime": 15.4033, | |
| "eval_samples_per_second": 12.919, | |
| "eval_steps_per_second": 1.623, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 50, | |
| "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": 7.54256789372928e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |