| { | |
| "best_global_step": 200, | |
| "best_metric": 0.22340504825115204, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 200, | |
| "global_step": 939, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.3194888178913738, | |
| "grad_norm": 14.664009094238281, | |
| "learning_rate": 9.9e-06, | |
| "loss": 0.6825, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6389776357827476, | |
| "grad_norm": 9.503365516662598, | |
| "learning_rate": 1.9900000000000003e-05, | |
| "loss": 0.3076, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6389776357827476, | |
| "eval_accuracy": 0.913, | |
| "eval_loss": 0.22340504825115204, | |
| "eval_runtime": 14.3197, | |
| "eval_samples_per_second": 69.834, | |
| "eval_steps_per_second": 4.4, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.9584664536741214, | |
| "grad_norm": 16.2436466217041, | |
| "learning_rate": 1.732070365358593e-05, | |
| "loss": 0.2755, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.2779552715654952, | |
| "grad_norm": 16.86052894592285, | |
| "learning_rate": 1.4614343707713128e-05, | |
| "loss": 0.2064, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2779552715654952, | |
| "eval_accuracy": 0.917, | |
| "eval_loss": 0.2431255728006363, | |
| "eval_runtime": 14.3011, | |
| "eval_samples_per_second": 69.925, | |
| "eval_steps_per_second": 4.405, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.5974440894568689, | |
| "grad_norm": 17.355743408203125, | |
| "learning_rate": 1.1907983761840325e-05, | |
| "loss": 0.201, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.9169329073482428, | |
| "grad_norm": 15.128535270690918, | |
| "learning_rate": 9.201623815967524e-06, | |
| "loss": 0.198, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.9169329073482428, | |
| "eval_accuracy": 0.928, | |
| "eval_loss": 0.253868043422699, | |
| "eval_runtime": 14.166, | |
| "eval_samples_per_second": 70.591, | |
| "eval_steps_per_second": 4.447, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.236421725239617, | |
| "grad_norm": 0.2765483260154724, | |
| "learning_rate": 6.495263870094724e-06, | |
| "loss": 0.1217, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.5559105431309903, | |
| "grad_norm": 0.1613696962594986, | |
| "learning_rate": 3.7889039242219215e-06, | |
| "loss": 0.0941, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.5559105431309903, | |
| "eval_accuracy": 0.935, | |
| "eval_loss": 0.2665516436100006, | |
| "eval_runtime": 14.2928, | |
| "eval_samples_per_second": 69.965, | |
| "eval_steps_per_second": 4.408, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.8753993610223643, | |
| "grad_norm": 0.09486523270606995, | |
| "learning_rate": 1.0825439783491206e-06, | |
| "loss": 0.1242, | |
| "step": 900 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 939, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 400, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1973332915200000.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |