| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.0, | |
| "eval_steps": 500, | |
| "global_step": 4770, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9968553459119497, | |
| "grad_norm": 0.01754792593419552, | |
| "learning_rate": 0.000776604254595881, | |
| "loss": 0.5977, | |
| "step": 317 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.0064516129032258064, | |
| "eval_loss": 0.6016289591789246, | |
| "eval_runtime": 5.4207, | |
| "eval_samples_per_second": 571.886, | |
| "eval_steps_per_second": 11.991, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.9937106918238994, | |
| "grad_norm": 0.013247409835457802, | |
| "learning_rate": 0.0007213193795213482, | |
| "loss": 0.6075, | |
| "step": 634 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.6012811064720154, | |
| "eval_runtime": 5.4611, | |
| "eval_samples_per_second": 567.648, | |
| "eval_steps_per_second": 11.902, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 2.990566037735849, | |
| "grad_norm": 0.012003601528704166, | |
| "learning_rate": 0.0006660345044468155, | |
| "loss": 0.6077, | |
| "step": 951 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.6012278199195862, | |
| "eval_runtime": 5.4297, | |
| "eval_samples_per_second": 570.937, | |
| "eval_steps_per_second": 11.971, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.9874213836477987, | |
| "grad_norm": 0.010578780435025692, | |
| "learning_rate": 0.0006107496293722828, | |
| "loss": 0.6072, | |
| "step": 1268 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.6011785268783569, | |
| "eval_runtime": 5.4145, | |
| "eval_samples_per_second": 572.54, | |
| "eval_steps_per_second": 12.005, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.984276729559748, | |
| "grad_norm": 0.01368038635700941, | |
| "learning_rate": 0.00055546475429775, | |
| "loss": 0.608, | |
| "step": 1585 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.6011925339698792, | |
| "eval_runtime": 5.4677, | |
| "eval_samples_per_second": 566.968, | |
| "eval_steps_per_second": 11.888, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 5.981132075471698, | |
| "grad_norm": 0.01561660785228014, | |
| "learning_rate": 0.0005001798792232172, | |
| "loss": 0.6076, | |
| "step": 1902 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.6011727452278137, | |
| "eval_runtime": 5.4651, | |
| "eval_samples_per_second": 567.234, | |
| "eval_steps_per_second": 11.894, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.977987421383648, | |
| "grad_norm": 0.012493623420596123, | |
| "learning_rate": 0.00044489500414868454, | |
| "loss": 0.607, | |
| "step": 2219 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.6012096405029297, | |
| "eval_runtime": 5.3992, | |
| "eval_samples_per_second": 574.154, | |
| "eval_steps_per_second": 12.039, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.9748427672955975, | |
| "grad_norm": 0.014549925923347473, | |
| "learning_rate": 0.00038961012907415184, | |
| "loss": 0.6078, | |
| "step": 2536 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.6011644601821899, | |
| "eval_runtime": 5.4037, | |
| "eval_samples_per_second": 573.681, | |
| "eval_steps_per_second": 12.029, | |
| "step": 2544 | |
| }, | |
| { | |
| "epoch": 8.971698113207546, | |
| "grad_norm": 0.02066616527736187, | |
| "learning_rate": 0.0003343252539996191, | |
| "loss": 0.6075, | |
| "step": 2853 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.6011798977851868, | |
| "eval_runtime": 5.475, | |
| "eval_samples_per_second": 566.208, | |
| "eval_steps_per_second": 11.872, | |
| "step": 2862 | |
| }, | |
| { | |
| "epoch": 9.968553459119496, | |
| "grad_norm": 0.010850013233721256, | |
| "learning_rate": 0.00027904037892508633, | |
| "loss": 0.6076, | |
| "step": 3170 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.6012014746665955, | |
| "eval_runtime": 5.4388, | |
| "eval_samples_per_second": 569.982, | |
| "eval_steps_per_second": 11.951, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 10.965408805031446, | |
| "grad_norm": 0.01570860482752323, | |
| "learning_rate": 0.00022375550385055363, | |
| "loss": 0.6077, | |
| "step": 3487 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.6011857390403748, | |
| "eval_runtime": 5.4108, | |
| "eval_samples_per_second": 572.928, | |
| "eval_steps_per_second": 12.013, | |
| "step": 3498 | |
| }, | |
| { | |
| "epoch": 11.962264150943396, | |
| "grad_norm": 0.007834916934370995, | |
| "learning_rate": 0.00016847062877602088, | |
| "loss": 0.6072, | |
| "step": 3804 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.6011877655982971, | |
| "eval_runtime": 5.4133, | |
| "eval_samples_per_second": 572.658, | |
| "eval_steps_per_second": 12.007, | |
| "step": 3816 | |
| }, | |
| { | |
| "epoch": 12.959119496855346, | |
| "grad_norm": 0.015295589342713356, | |
| "learning_rate": 0.00011318575370148815, | |
| "loss": 0.6081, | |
| "step": 4121 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.6011701226234436, | |
| "eval_runtime": 5.4283, | |
| "eval_samples_per_second": 571.085, | |
| "eval_steps_per_second": 11.974, | |
| "step": 4134 | |
| }, | |
| { | |
| "epoch": 13.955974842767295, | |
| "grad_norm": 0.014071750454604626, | |
| "learning_rate": 5.790087862695542e-05, | |
| "loss": 0.607, | |
| "step": 4438 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.6011898517608643, | |
| "eval_runtime": 5.4906, | |
| "eval_samples_per_second": 564.599, | |
| "eval_steps_per_second": 11.838, | |
| "step": 4452 | |
| }, | |
| { | |
| "epoch": 14.952830188679245, | |
| "grad_norm": 0.011002879589796066, | |
| "learning_rate": 2.6160035524226845e-06, | |
| "loss": 0.6074, | |
| "step": 4755 | |
| } | |
| ], | |
| "logging_steps": 317, | |
| "max_steps": 4770, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "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": 1236646073993904.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.3738796870532265, | |
| "learning_rate": 0.0008318891296704137, | |
| "lr_scheduler_type": "linear", | |
| "num_train_epochs": 15, | |
| "temperature": 10.526715944436805, | |
| "weight_decay": 0.22690902065757862 | |
| } | |
| } | |