| { | |
| "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": 0.007859878242015839, | |
| "learning_rate": 0.0007934093547399718, | |
| "loss": 0.5931, | |
| "step": 317 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.5877403020858765, | |
| "eval_runtime": 5.3692, | |
| "eval_samples_per_second": 577.366, | |
| "eval_steps_per_second": 12.106, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.9937106918238994, | |
| "grad_norm": 0.013005654327571392, | |
| "learning_rate": 0.0007676597858899992, | |
| "loss": 0.5933, | |
| "step": 634 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.5876849293708801, | |
| "eval_runtime": 5.3979, | |
| "eval_samples_per_second": 574.293, | |
| "eval_steps_per_second": 12.042, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 2.990566037735849, | |
| "grad_norm": 0.011686289682984352, | |
| "learning_rate": 0.0007259882616863973, | |
| "loss": 0.5936, | |
| "step": 951 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.5876944661140442, | |
| "eval_runtime": 5.3376, | |
| "eval_samples_per_second": 580.786, | |
| "eval_steps_per_second": 12.178, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.9874213836477987, | |
| "grad_norm": 0.010250881314277649, | |
| "learning_rate": 0.0006702046329072582, | |
| "loss": 0.5932, | |
| "step": 1268 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.5876566171646118, | |
| "eval_runtime": 5.3623, | |
| "eval_samples_per_second": 578.11, | |
| "eval_steps_per_second": 12.122, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.984276729559748, | |
| "grad_norm": 0.013469184748828411, | |
| "learning_rate": 0.0006027316581600536, | |
| "loss": 0.594, | |
| "step": 1585 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.5876731276512146, | |
| "eval_runtime": 5.3565, | |
| "eval_samples_per_second": 578.738, | |
| "eval_steps_per_second": 12.135, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 5.981132075471698, | |
| "grad_norm": 0.015176467597484589, | |
| "learning_rate": 0.0005264997801914848, | |
| "loss": 0.5936, | |
| "step": 1902 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.5876610279083252, | |
| "eval_runtime": 5.3967, | |
| "eval_samples_per_second": 574.42, | |
| "eval_steps_per_second": 12.044, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.977987421383648, | |
| "grad_norm": 0.01225706934928894, | |
| "learning_rate": 0.0004448198527870465, | |
| "loss": 0.593, | |
| "step": 2219 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.5876200795173645, | |
| "eval_runtime": 5.386, | |
| "eval_samples_per_second": 575.564, | |
| "eval_steps_per_second": 12.068, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.9748427672955975, | |
| "grad_norm": 0.014281037263572216, | |
| "learning_rate": 0.00036123934590356535, | |
| "loss": 0.5938, | |
| "step": 2536 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.5876643061637878, | |
| "eval_runtime": 5.34, | |
| "eval_samples_per_second": 580.525, | |
| "eval_steps_per_second": 12.172, | |
| "step": 2544 | |
| }, | |
| { | |
| "epoch": 8.971698113207546, | |
| "grad_norm": 0.020413335412740707, | |
| "learning_rate": 0.0002793882742407039, | |
| "loss": 0.5934, | |
| "step": 2853 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.5876378417015076, | |
| "eval_runtime": 5.3343, | |
| "eval_samples_per_second": 581.141, | |
| "eval_steps_per_second": 12.185, | |
| "step": 2862 | |
| }, | |
| { | |
| "epoch": 9.968553459119496, | |
| "grad_norm": 0.010601122863590717, | |
| "learning_rate": 0.00020282154078240177, | |
| "loss": 0.5935, | |
| "step": 3170 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.5876396298408508, | |
| "eval_runtime": 5.3546, | |
| "eval_samples_per_second": 578.942, | |
| "eval_steps_per_second": 12.139, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 10.965408805031446, | |
| "grad_norm": 0.015482204966247082, | |
| "learning_rate": 0.00013486454254193946, | |
| "loss": 0.5936, | |
| "step": 3487 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.5875952839851379, | |
| "eval_runtime": 5.3966, | |
| "eval_samples_per_second": 574.436, | |
| "eval_steps_per_second": 12.045, | |
| "step": 3498 | |
| }, | |
| { | |
| "epoch": 11.962264150943396, | |
| "grad_norm": 0.007544202264398336, | |
| "learning_rate": 7.846874406237966e-05, | |
| "loss": 0.5932, | |
| "step": 3804 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.587660014629364, | |
| "eval_runtime": 5.388, | |
| "eval_samples_per_second": 575.35, | |
| "eval_steps_per_second": 12.064, | |
| "step": 3816 | |
| }, | |
| { | |
| "epoch": 12.959119496855346, | |
| "grad_norm": 0.015037346631288528, | |
| "learning_rate": 3.608349131102299e-05, | |
| "loss": 0.594, | |
| "step": 4121 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.5875839591026306, | |
| "eval_runtime": 5.3407, | |
| "eval_samples_per_second": 580.45, | |
| "eval_steps_per_second": 12.171, | |
| "step": 4134 | |
| }, | |
| { | |
| "epoch": 13.955974842767295, | |
| "grad_norm": 0.013815987855196, | |
| "learning_rate": 9.549633264184268e-06, | |
| "loss": 0.593, | |
| "step": 4438 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.03225806451612903, | |
| "eval_loss": 0.5875993371009827, | |
| "eval_runtime": 5.3398, | |
| "eval_samples_per_second": 580.541, | |
| "eval_steps_per_second": 12.173, | |
| "step": 4452 | |
| }, | |
| { | |
| "epoch": 14.952830188679245, | |
| "grad_norm": 0.010695732198655605, | |
| "learning_rate": 1.9571341049241364e-08, | |
| "loss": 0.5934, | |
| "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": 0.5457565605433671, | |
| "learning_rate": 0.0008021186295599815, | |
| "lr_scheduler_type": "cosine", | |
| "num_train_epochs": 15, | |
| "temperature": 13.83793993486481, | |
| "weight_decay": 0.09136269626429569 | |
| } | |
| } | |