| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 200, | |
| "global_step": 3330, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6006006006006006, | |
| "grad_norm": 10.51630973815918, | |
| "learning_rate": 1.1951951951951951e-05, | |
| "loss": 2.6385, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6006006006006006, | |
| "eval_train_loss": 2.488961696624756, | |
| "eval_train_runtime": 15.8109, | |
| "eval_train_samples_per_second": 338.943, | |
| "eval_train_steps_per_second": 10.626, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.2012012012012012, | |
| "grad_norm": 9.947551727294922, | |
| "learning_rate": 1.9559559559559563e-05, | |
| "loss": 2.3324, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2012012012012012, | |
| "eval_train_loss": 2.419891119003296, | |
| "eval_train_runtime": 15.4172, | |
| "eval_train_samples_per_second": 347.6, | |
| "eval_train_steps_per_second": 10.897, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.8018018018018018, | |
| "grad_norm": 9.27906322479248, | |
| "learning_rate": 1.8224891558224895e-05, | |
| "loss": 2.1772, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.8018018018018018, | |
| "eval_train_loss": 2.389059543609619, | |
| "eval_train_runtime": 15.629, | |
| "eval_train_samples_per_second": 342.889, | |
| "eval_train_steps_per_second": 10.749, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.4024024024024024, | |
| "grad_norm": 8.641522407531738, | |
| "learning_rate": 1.6890223556890223e-05, | |
| "loss": 2.0635, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.4024024024024024, | |
| "eval_train_loss": 2.369140148162842, | |
| "eval_train_runtime": 15.6802, | |
| "eval_train_samples_per_second": 341.768, | |
| "eval_train_steps_per_second": 10.714, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.003003003003003, | |
| "grad_norm": 9.14696216583252, | |
| "learning_rate": 1.555555555555556e-05, | |
| "loss": 1.9915, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.003003003003003, | |
| "eval_train_loss": 2.360917806625366, | |
| "eval_train_runtime": 15.1166, | |
| "eval_train_samples_per_second": 354.511, | |
| "eval_train_steps_per_second": 11.114, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.6036036036036037, | |
| "grad_norm": 8.530783653259277, | |
| "learning_rate": 1.4220887554220888e-05, | |
| "loss": 1.9008, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.6036036036036037, | |
| "eval_train_loss": 2.3688642978668213, | |
| "eval_train_runtime": 15.4199, | |
| "eval_train_samples_per_second": 347.539, | |
| "eval_train_steps_per_second": 10.895, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.2042042042042045, | |
| "grad_norm": 10.91430377960205, | |
| "learning_rate": 1.288621955288622e-05, | |
| "loss": 1.8603, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.2042042042042045, | |
| "eval_train_loss": 2.3849704265594482, | |
| "eval_train_runtime": 15.4842, | |
| "eval_train_samples_per_second": 346.094, | |
| "eval_train_steps_per_second": 10.85, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.804804804804805, | |
| "grad_norm": 7.67103910446167, | |
| "learning_rate": 1.1551551551551552e-05, | |
| "loss": 1.8421, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.804804804804805, | |
| "eval_train_loss": 2.3467628955841064, | |
| "eval_train_runtime": 15.716, | |
| "eval_train_samples_per_second": 340.99, | |
| "eval_train_steps_per_second": 10.69, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.405405405405405, | |
| "grad_norm": 10.279878616333008, | |
| "learning_rate": 1.0216883550216886e-05, | |
| "loss": 1.785, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 5.405405405405405, | |
| "eval_train_loss": 2.3648931980133057, | |
| "eval_train_runtime": 15.8675, | |
| "eval_train_samples_per_second": 337.735, | |
| "eval_train_steps_per_second": 10.588, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 6.006006006006006, | |
| "grad_norm": 9.513252258300781, | |
| "learning_rate": 8.882215548882216e-06, | |
| "loss": 1.786, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.006006006006006, | |
| "eval_train_loss": 2.378257989883423, | |
| "eval_train_runtime": 16.2336, | |
| "eval_train_samples_per_second": 330.118, | |
| "eval_train_steps_per_second": 10.349, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.606606606606607, | |
| "grad_norm": 11.29470157623291, | |
| "learning_rate": 7.547547547547548e-06, | |
| "loss": 1.7331, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 6.606606606606607, | |
| "eval_train_loss": 2.3782169818878174, | |
| "eval_train_runtime": 15.9451, | |
| "eval_train_samples_per_second": 336.091, | |
| "eval_train_steps_per_second": 10.536, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 7.207207207207207, | |
| "grad_norm": 10.022029876708984, | |
| "learning_rate": 6.21287954621288e-06, | |
| "loss": 1.7062, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 7.207207207207207, | |
| "eval_train_loss": 2.3825714588165283, | |
| "eval_train_runtime": 15.9203, | |
| "eval_train_samples_per_second": 336.615, | |
| "eval_train_steps_per_second": 10.553, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 7.807807807807808, | |
| "grad_norm": 10.572230339050293, | |
| "learning_rate": 4.878211544878212e-06, | |
| "loss": 1.6929, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 7.807807807807808, | |
| "eval_train_loss": 2.3926050662994385, | |
| "eval_train_runtime": 15.9657, | |
| "eval_train_samples_per_second": 335.657, | |
| "eval_train_steps_per_second": 10.523, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 8.408408408408409, | |
| "grad_norm": 10.39648723602295, | |
| "learning_rate": 3.5435435435435437e-06, | |
| "loss": 1.6618, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 8.408408408408409, | |
| "eval_train_loss": 2.4069101810455322, | |
| "eval_train_runtime": 15.7461, | |
| "eval_train_samples_per_second": 340.337, | |
| "eval_train_steps_per_second": 10.669, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 9.00900900900901, | |
| "grad_norm": 12.46285343170166, | |
| "learning_rate": 2.2088755422088755e-06, | |
| "loss": 1.6348, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.00900900900901, | |
| "eval_train_loss": 2.415452480316162, | |
| "eval_train_runtime": 15.6461, | |
| "eval_train_samples_per_second": 342.513, | |
| "eval_train_steps_per_second": 10.737, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.60960960960961, | |
| "grad_norm": 11.425995826721191, | |
| "learning_rate": 8.742075408742076e-07, | |
| "loss": 1.6553, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 9.60960960960961, | |
| "eval_train_loss": 2.40596079826355, | |
| "eval_train_runtime": 15.9222, | |
| "eval_train_samples_per_second": 336.573, | |
| "eval_train_steps_per_second": 10.551, | |
| "step": 3200 | |
| } | |
| ], | |
| "logging_steps": 200, | |
| "max_steps": 3330, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 2000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |