| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.0, | |
| "eval_steps": 500, | |
| "global_step": 2862, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9968553459119497, | |
| "grad_norm": 0.9016032814979553, | |
| "learning_rate": 1.7784765897973445e-05, | |
| "loss": 0.8053, | |
| "step": 317 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6696774193548387, | |
| "eval_loss": 0.40919938683509827, | |
| "eval_runtime": 1.4322, | |
| "eval_samples_per_second": 2164.57, | |
| "eval_steps_per_second": 45.386, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.9937106918238994, | |
| "grad_norm": 0.7222766876220703, | |
| "learning_rate": 1.556953179594689e-05, | |
| "loss": 0.3073, | |
| "step": 634 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.844516129032258, | |
| "eval_loss": 0.13953416049480438, | |
| "eval_runtime": 1.4538, | |
| "eval_samples_per_second": 2132.319, | |
| "eval_steps_per_second": 44.71, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 2.990566037735849, | |
| "grad_norm": 0.6133838891983032, | |
| "learning_rate": 1.3354297693920338e-05, | |
| "loss": 0.1428, | |
| "step": 951 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8964516129032258, | |
| "eval_loss": 0.0722731500864029, | |
| "eval_runtime": 1.4484, | |
| "eval_samples_per_second": 2140.365, | |
| "eval_steps_per_second": 44.879, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.9874213836477987, | |
| "grad_norm": 0.5777781009674072, | |
| "learning_rate": 1.1139063591893781e-05, | |
| "loss": 0.0917, | |
| "step": 1268 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9180645161290323, | |
| "eval_loss": 0.050948865711688995, | |
| "eval_runtime": 1.645, | |
| "eval_samples_per_second": 1884.542, | |
| "eval_steps_per_second": 39.515, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.984276729559748, | |
| "grad_norm": 0.4380759596824646, | |
| "learning_rate": 8.923829489867226e-06, | |
| "loss": 0.0715, | |
| "step": 1585 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9232258064516129, | |
| "eval_loss": 0.042863838374614716, | |
| "eval_runtime": 1.6378, | |
| "eval_samples_per_second": 1892.819, | |
| "eval_steps_per_second": 39.688, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 5.981132075471698, | |
| "grad_norm": 0.46454158425331116, | |
| "learning_rate": 6.708595387840672e-06, | |
| "loss": 0.0617, | |
| "step": 1902 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9261290322580645, | |
| "eval_loss": 0.03817891329526901, | |
| "eval_runtime": 1.6408, | |
| "eval_samples_per_second": 1889.326, | |
| "eval_steps_per_second": 39.615, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.977987421383648, | |
| "grad_norm": 0.34600403904914856, | |
| "learning_rate": 4.4933612858141165e-06, | |
| "loss": 0.0562, | |
| "step": 2219 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9303225806451613, | |
| "eval_loss": 0.035540465265512466, | |
| "eval_runtime": 1.6395, | |
| "eval_samples_per_second": 1890.84, | |
| "eval_steps_per_second": 39.647, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.9748427672955975, | |
| "grad_norm": 0.32331326603889465, | |
| "learning_rate": 2.2781271837875614e-06, | |
| "loss": 0.053, | |
| "step": 2536 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.932258064516129, | |
| "eval_loss": 0.034126147627830505, | |
| "eval_runtime": 1.6425, | |
| "eval_samples_per_second": 1887.356, | |
| "eval_steps_per_second": 39.574, | |
| "step": 2544 | |
| }, | |
| { | |
| "epoch": 8.971698113207546, | |
| "grad_norm": 0.3256802558898926, | |
| "learning_rate": 6.289308176100629e-08, | |
| "loss": 0.0511, | |
| "step": 2853 | |
| } | |
| ], | |
| "logging_steps": 317, | |
| "max_steps": 2862, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 1000000000.0, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 743272879539288.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.7270619849890861, | |
| "num_train_epochs": 9, | |
| "temperature": 2 | |
| } | |
| } | |