| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.03568242640499554, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007136485280999108, | |
| "eval_loss": 3.7690577507019043, | |
| "eval_runtime": 4.3864, | |
| "eval_samples_per_second": 134.505, | |
| "eval_steps_per_second": 67.253, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.007136485280999108, | |
| "grad_norm": 6.6461286544799805, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 13.5698, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.009277430865298841, | |
| "eval_loss": 2.0543437004089355, | |
| "eval_runtime": 4.3852, | |
| "eval_samples_per_second": 134.544, | |
| "eval_steps_per_second": 67.272, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.014272970561998216, | |
| "grad_norm": 10.23231029510498, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 7.7515, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.018554861730597682, | |
| "eval_loss": 1.4419488906860352, | |
| "eval_runtime": 4.4312, | |
| "eval_samples_per_second": 133.145, | |
| "eval_steps_per_second": 66.573, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.021409455842997322, | |
| "grad_norm": 8.210744857788086, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 6.1164, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.02783229259589652, | |
| "eval_loss": 1.2978790998458862, | |
| "eval_runtime": 4.4556, | |
| "eval_samples_per_second": 132.417, | |
| "eval_steps_per_second": 66.208, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.028545941123996433, | |
| "grad_norm": 7.938344478607178, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 5.2656, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03568242640499554, | |
| "grad_norm": 8.023124694824219, | |
| "learning_rate": 0.0, | |
| "loss": 5.2916, | |
| "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": 1121837069107200.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |