| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0008827449838016295, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.765489967603259e-05, | |
| "eval_loss": 6.259857177734375, | |
| "eval_runtime": 299.3259, | |
| "eval_samples_per_second": 39.84, | |
| "eval_steps_per_second": 19.921, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00017654899676032592, | |
| "grad_norm": 12.482221603393555, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 5.0307, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00022951369578842369, | |
| "eval_loss": 3.8883798122406006, | |
| "eval_runtime": 297.4626, | |
| "eval_samples_per_second": 40.089, | |
| "eval_steps_per_second": 20.046, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.00035309799352065184, | |
| "grad_norm": 10.5984468460083, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 3.8188, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00045902739157684737, | |
| "eval_loss": 3.6610496044158936, | |
| "eval_runtime": 298.9288, | |
| "eval_samples_per_second": 39.892, | |
| "eval_steps_per_second": 19.948, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0005296469902809777, | |
| "grad_norm": 17.2176513671875, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 3.7025, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.000688541087365271, | |
| "eval_loss": 3.5662386417388916, | |
| "eval_runtime": 297.522, | |
| "eval_samples_per_second": 40.081, | |
| "eval_steps_per_second": 20.042, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0007061959870413037, | |
| "grad_norm": 9.540621757507324, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 3.4454, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0008827449838016295, | |
| "grad_norm": 8.860605239868164, | |
| "learning_rate": 0.0, | |
| "loss": 3.6023, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1294900081459200.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |