| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 11070, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09033423667570009, | |
| "grad_norm": 4.842950820922852, | |
| "learning_rate": 9.5483288166215e-05, | |
| "loss": 4.3861, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.18066847335140018, | |
| "grad_norm": 4.639570713043213, | |
| "learning_rate": 9.096657633242999e-05, | |
| "loss": 4.037, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.27100271002710025, | |
| "grad_norm": 3.680816650390625, | |
| "learning_rate": 8.644986449864499e-05, | |
| "loss": 3.9132, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.36133694670280037, | |
| "grad_norm": 3.527472496032715, | |
| "learning_rate": 8.193315266485999e-05, | |
| "loss": 3.8609, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.45167118337850043, | |
| "grad_norm": 4.149468421936035, | |
| "learning_rate": 7.741644083107498e-05, | |
| "loss": 3.8017, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.5420054200542005, | |
| "grad_norm": 4.817673683166504, | |
| "learning_rate": 7.289972899728998e-05, | |
| "loss": 3.7356, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6323396567299007, | |
| "grad_norm": 5.4917497634887695, | |
| "learning_rate": 6.838301716350496e-05, | |
| "loss": 3.6767, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.7226738934056007, | |
| "grad_norm": 5.168456077575684, | |
| "learning_rate": 6.386630532971996e-05, | |
| "loss": 3.6308, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.8130081300813008, | |
| "grad_norm": 5.315395832061768, | |
| "learning_rate": 5.9349593495934964e-05, | |
| "loss": 3.6409, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.9033423667570009, | |
| "grad_norm": 4.9155988693237305, | |
| "learning_rate": 5.483288166214996e-05, | |
| "loss": 3.5936, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.993676603432701, | |
| "grad_norm": 5.8576812744140625, | |
| "learning_rate": 5.031616982836496e-05, | |
| "loss": 3.5826, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.084010840108401, | |
| "grad_norm": 7.912364482879639, | |
| "learning_rate": 4.579945799457995e-05, | |
| "loss": 3.2283, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.174345076784101, | |
| "grad_norm": 7.648754596710205, | |
| "learning_rate": 4.1282746160794944e-05, | |
| "loss": 3.1922, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.2646793134598013, | |
| "grad_norm": 7.921971321105957, | |
| "learning_rate": 3.6766034327009944e-05, | |
| "loss": 3.1329, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.3550135501355014, | |
| "grad_norm": 8.775491714477539, | |
| "learning_rate": 3.224932249322493e-05, | |
| "loss": 3.151, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.4453477868112015, | |
| "grad_norm": 8.102723121643066, | |
| "learning_rate": 2.7732610659439927e-05, | |
| "loss": 3.1504, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.5356820234869015, | |
| "grad_norm": 7.8450751304626465, | |
| "learning_rate": 2.3215898825654924e-05, | |
| "loss": 3.1267, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.6260162601626016, | |
| "grad_norm": 8.472620010375977, | |
| "learning_rate": 1.869918699186992e-05, | |
| "loss": 3.1105, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.7163504968383019, | |
| "grad_norm": 7.973145961761475, | |
| "learning_rate": 1.4182475158084915e-05, | |
| "loss": 3.0822, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.8066847335140017, | |
| "grad_norm": 8.880084991455078, | |
| "learning_rate": 9.66576332429991e-06, | |
| "loss": 3.0406, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.897018970189702, | |
| "grad_norm": 9.18106460571289, | |
| "learning_rate": 5.149051490514906e-06, | |
| "loss": 3.0481, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.9873532068654018, | |
| "grad_norm": 9.302742004394531, | |
| "learning_rate": 6.323396567299007e-07, | |
| "loss": 3.0543, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 11070, | |
| "total_flos": 3.971230131335731e+17, | |
| "train_loss": 3.4603993375342283, | |
| "train_runtime": 12312.9591, | |
| "train_samples_per_second": 14.385, | |
| "train_steps_per_second": 0.899 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 11070, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 3.971230131335731e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |