| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.896, | |
| "eval_steps": 500, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 6.713689804077148, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.234, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.880593299865723, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 1.2808, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 6.147163391113281, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2447, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.915257453918457, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 1.1614, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 6.076275825500488, | |
| "learning_rate": 9.86522435289912e-06, | |
| "loss": 1.1953, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 5.434309959411621, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 1.1858, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 5.233426570892334, | |
| "learning_rate": 9.468163201617063e-06, | |
| "loss": 1.1388, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 4.256894111633301, | |
| "learning_rate": 9.177439057064684e-06, | |
| "loss": 1.1322, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.779238700866699, | |
| "learning_rate": 8.83022221559489e-06, | |
| "loss": 1.0899, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.9273000955581665, | |
| "learning_rate": 8.43120818934367e-06, | |
| "loss": 1.0839, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.064, | |
| "grad_norm": 3.337892532348633, | |
| "learning_rate": 7.985792958513932e-06, | |
| "loss": 1.6738, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 1.6545937061309814, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 1.0706, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.256, | |
| "grad_norm": 1.280295729637146, | |
| "learning_rate": 6.980398830195785e-06, | |
| "loss": 1.0189, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3519999999999999, | |
| "grad_norm": 1.170376181602478, | |
| "learning_rate": 6.434016163555452e-06, | |
| "loss": 1.028, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.448, | |
| "grad_norm": 0.9753493666648865, | |
| "learning_rate": 5.8682408883346535e-06, | |
| "loss": 0.9001, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.544, | |
| "grad_norm": 1.2441856861114502, | |
| "learning_rate": 5.290724144552379e-06, | |
| "loss": 1.0615, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 1.053038477897644, | |
| "learning_rate": 4.7092758554476215e-06, | |
| "loss": 0.9473, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.736, | |
| "grad_norm": 0.8429034948348999, | |
| "learning_rate": 4.131759111665349e-06, | |
| "loss": 0.9851, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8319999999999999, | |
| "grad_norm": 0.8916655778884888, | |
| "learning_rate": 3.5659838364445505e-06, | |
| "loss": 0.8426, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.928, | |
| "grad_norm": 1.2656738758087158, | |
| "learning_rate": 3.019601169804216e-06, | |
| "loss": 1.0051, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.032, | |
| "grad_norm": 1.7887486219406128, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 1.6283, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.128, | |
| "grad_norm": 0.8325550556182861, | |
| "learning_rate": 2.0142070414860704e-06, | |
| "loss": 0.9316, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.224, | |
| "grad_norm": 0.7157667875289917, | |
| "learning_rate": 1.5687918106563326e-06, | |
| "loss": 0.9276, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 0.6169079542160034, | |
| "learning_rate": 1.1697777844051105e-06, | |
| "loss": 0.8732, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.416, | |
| "grad_norm": 0.6843702793121338, | |
| "learning_rate": 8.225609429353187e-07, | |
| "loss": 0.9338, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.512, | |
| "grad_norm": 0.5900323987007141, | |
| "learning_rate": 5.318367983829393e-07, | |
| "loss": 0.908, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.608, | |
| "grad_norm": 0.7179703712463379, | |
| "learning_rate": 3.015368960704584e-07, | |
| "loss": 0.9636, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.7039999999999997, | |
| "grad_norm": 0.6371797323226929, | |
| "learning_rate": 1.3477564710088097e-07, | |
| "loss": 0.941, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.5585090517997742, | |
| "learning_rate": 3.3808211290284886e-08, | |
| "loss": 0.8554, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.896, | |
| "grad_norm": 0.6148462295532227, | |
| "learning_rate": 0.0, | |
| "loss": 1.0298, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.896, | |
| "step": 30, | |
| "total_flos": 29476257431552.0, | |
| "train_loss": 1.0757389744122823, | |
| "train_runtime": 1864.5646, | |
| "train_samples_per_second": 1.609, | |
| "train_steps_per_second": 0.016 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 29476257431552.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |