| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.01011804384485666, | |
| "eval_steps": 9, | |
| "global_step": 18, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0005621135469364812, | |
| "eval_loss": 0.9150053262710571, | |
| "eval_runtime": 163.2125, | |
| "eval_samples_per_second": 18.363, | |
| "eval_steps_per_second": 2.298, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0016863406408094434, | |
| "grad_norm": 0.2956816852092743, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9448, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.003372681281618887, | |
| "grad_norm": 0.2726631164550781, | |
| "learning_rate": 6e-05, | |
| "loss": 0.9084, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00505902192242833, | |
| "grad_norm": 0.21325236558914185, | |
| "learning_rate": 9e-05, | |
| "loss": 0.8568, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00505902192242833, | |
| "eval_loss": 0.8506203293800354, | |
| "eval_runtime": 164.5381, | |
| "eval_samples_per_second": 18.215, | |
| "eval_steps_per_second": 2.279, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.006745362563237774, | |
| "grad_norm": 0.2024015486240387, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.841, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.008431703204047217, | |
| "grad_norm": 0.21213184297084808, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.8612, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.01011804384485666, | |
| "grad_norm": 0.18286366760730743, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.7799, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.01011804384485666, | |
| "eval_loss": 0.7897757887840271, | |
| "eval_runtime": 164.6058, | |
| "eval_samples_per_second": 18.207, | |
| "eval_steps_per_second": 2.278, | |
| "step": 18 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9405135625125888.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |