| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.11284046692607, | |
| "eval_steps": 50, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1556420233463035, | |
| "grad_norm": 5.363790512084961, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 2.5426, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.311284046692607, | |
| "grad_norm": 5.297606468200684, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 2.4551, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.4669260700389105, | |
| "grad_norm": 1.6734049320220947, | |
| "learning_rate": 9.999999999999999e-06, | |
| "loss": 1.5765, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.622568093385214, | |
| "grad_norm": 0.5996205806732178, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.9482, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.7782101167315175, | |
| "grad_norm": 0.5326632857322693, | |
| "learning_rate": 1.4994303528285384e-05, | |
| "loss": 0.7767, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.7782101167315175, | |
| "eval_loss": 0.7420370578765869, | |
| "eval_runtime": 75.0351, | |
| "eval_samples_per_second": 3.052, | |
| "eval_steps_per_second": 1.533, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.933852140077821, | |
| "grad_norm": 0.47296807169914246, | |
| "learning_rate": 1.4948783661087592e-05, | |
| "loss": 0.6917, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.0894941634241244, | |
| "grad_norm": 0.5496929883956909, | |
| "learning_rate": 1.485802041113141e-05, | |
| "loss": 0.5968, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.245136186770428, | |
| "grad_norm": 0.5656551122665405, | |
| "learning_rate": 1.4722565067948796e-05, | |
| "loss": 0.5096, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.4007782101167314, | |
| "grad_norm": 0.7151289582252502, | |
| "learning_rate": 1.454324037767081e-05, | |
| "loss": 0.5053, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.556420233463035, | |
| "grad_norm": 0.7926196455955505, | |
| "learning_rate": 1.4321135545726069e-05, | |
| "loss": 0.4193, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.556420233463035, | |
| "eval_loss": 0.4160374701023102, | |
| "eval_runtime": 75.0144, | |
| "eval_samples_per_second": 3.053, | |
| "eval_steps_per_second": 1.533, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.7120622568093387, | |
| "grad_norm": 0.7219254970550537, | |
| "learning_rate": 1.4057599621084365e-05, | |
| "loss": 0.3695, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.867704280155642, | |
| "grad_norm": 0.7828359007835388, | |
| "learning_rate": 1.3754233302229055e-05, | |
| "loss": 0.3554, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.0233463035019454, | |
| "grad_norm": 0.6474707126617432, | |
| "learning_rate": 1.3412879214628194e-05, | |
| "loss": 0.3115, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.178988326848249, | |
| "grad_norm": 0.8858464360237122, | |
| "learning_rate": 1.3035610718758362e-05, | |
| "loss": 0.2754, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.3346303501945527, | |
| "grad_norm": 0.6484001278877258, | |
| "learning_rate": 1.2624719316660416e-05, | |
| "loss": 0.2815, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.3346303501945527, | |
| "eval_loss": 0.3077065050601959, | |
| "eval_runtime": 74.9901, | |
| "eval_samples_per_second": 3.054, | |
| "eval_steps_per_second": 1.534, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.490272373540856, | |
| "grad_norm": 0.9611061811447144, | |
| "learning_rate": 1.218270073351891e-05, | |
| "loss": 0.2641, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.6459143968871595, | |
| "grad_norm": 1.0954933166503906, | |
| "learning_rate": 1.1712239758804626e-05, | |
| "loss": 0.2675, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.801556420233463, | |
| "grad_norm": 0.7623236775398254, | |
| "learning_rate": 1.1216193939054056e-05, | |
| "loss": 0.26, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.9571984435797667, | |
| "grad_norm": 0.8693030476570129, | |
| "learning_rate": 1.0697576221334781e-05, | |
| "loss": 0.2494, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.11284046692607, | |
| "grad_norm": 0.9107272028923035, | |
| "learning_rate": 1.0159536652819018e-05, | |
| "loss": 0.2131, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.11284046692607, | |
| "eval_loss": 0.26981863379478455, | |
| "eval_runtime": 75.23, | |
| "eval_samples_per_second": 3.044, | |
| "eval_steps_per_second": 1.529, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 448, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.9392870840113562e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |