| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0006399902477676531, | |
| "eval_steps": 5, | |
| "global_step": 21, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.0475726084173955e-05, | |
| "eval_loss": 1.0151351690292358, | |
| "eval_runtime": 1911.0259, | |
| "eval_samples_per_second": 7.23, | |
| "eval_steps_per_second": 3.615, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 9.142717825252186e-05, | |
| "grad_norm": 2.0595638751983643, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 3.6748, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00015237863042086977, | |
| "eval_loss": 1.0050420761108398, | |
| "eval_runtime": 1916.5778, | |
| "eval_samples_per_second": 7.209, | |
| "eval_steps_per_second": 3.604, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00018285435650504373, | |
| "grad_norm": 1.2734413146972656, | |
| "learning_rate": 5.714285714285714e-05, | |
| "loss": 4.541, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0002742815347575656, | |
| "grad_norm": 2.5725290775299072, | |
| "learning_rate": 8.571428571428571e-05, | |
| "loss": 3.8809, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00030475726084173955, | |
| "eval_loss": 0.9246754050254822, | |
| "eval_runtime": 1917.7451, | |
| "eval_samples_per_second": 7.204, | |
| "eval_steps_per_second": 3.602, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00036570871301008746, | |
| "grad_norm": 1.181660532951355, | |
| "learning_rate": 0.00011428571428571428, | |
| "loss": 3.6147, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.00045713589126260935, | |
| "grad_norm": 1.1298421621322632, | |
| "learning_rate": 0.00014285714285714287, | |
| "loss": 3.2705, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00045713589126260935, | |
| "eval_loss": 0.8345330357551575, | |
| "eval_runtime": 1916.8879, | |
| "eval_samples_per_second": 7.208, | |
| "eval_steps_per_second": 3.604, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0005485630695151312, | |
| "grad_norm": 1.11538827419281, | |
| "learning_rate": 0.00017142857142857143, | |
| "loss": 3.8981, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0006095145216834791, | |
| "eval_loss": 0.7881443500518799, | |
| "eval_runtime": 1916.3846, | |
| "eval_samples_per_second": 7.209, | |
| "eval_steps_per_second": 3.605, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0006399902477676531, | |
| "grad_norm": 1.4301753044128418, | |
| "learning_rate": 0.0002, | |
| "loss": 2.8858, | |
| "step": 21 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 39, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 21, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.57107795001344e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |