| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 2.902575587905935, |
| "eval_steps": 500, |
| "global_step": 81, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.3583426651735722, |
| "grad_norm": 5.531366142635329, |
| "learning_rate": 1.9605181116313725e-05, |
| "loss": 1.6818, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.7166853303471444, |
| "grad_norm": 2.789059650297039, |
| "learning_rate": 1.7746049618276545e-05, |
| "loss": 0.6865, |
| "step": 20 |
| }, |
| { |
| "epoch": 1.0750279955207167, |
| "grad_norm": 2.5223167803496107, |
| "learning_rate": 1.4647231720437687e-05, |
| "loss": 0.5421, |
| "step": 30 |
| }, |
| { |
| "epoch": 1.4333706606942889, |
| "grad_norm": 2.1115491811570513, |
| "learning_rate": 1.0804665687167262e-05, |
| "loss": 0.387, |
| "step": 40 |
| }, |
| { |
| "epoch": 1.7917133258678613, |
| "grad_norm": 2.0127850710503514, |
| "learning_rate": 6.833320061985278e-06, |
| "loss": 0.3502, |
| "step": 50 |
| }, |
| { |
| "epoch": 2.1500559910414334, |
| "grad_norm": 1.7724633805166048, |
| "learning_rate": 3.3687734175920505e-06, |
| "loss": 0.2666, |
| "step": 60 |
| }, |
| { |
| "epoch": 2.5083986562150056, |
| "grad_norm": 1.7750457016789054, |
| "learning_rate": 9.65495653896179e-07, |
| "loss": 0.1604, |
| "step": 70 |
| }, |
| { |
| "epoch": 2.8667413213885777, |
| "grad_norm": 1.5147452122067369, |
| "learning_rate": 8.110018284304132e-09, |
| "loss": 0.1471, |
| "step": 80 |
| }, |
| { |
| "epoch": 2.902575587905935, |
| "step": 81, |
| "total_flos": 8460578979840.0, |
| "train_loss": 0.5228683275205118, |
| "train_runtime": 9122.3291, |
| "train_samples_per_second": 1.174, |
| "train_steps_per_second": 0.009 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 81, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 8460578979840.0, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
