| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0, |
| "eval_steps": 1000, |
| "global_step": 7093, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.00042319085907744394, |
| "grad_norm": 4.317407131195068, |
| "learning_rate": 0.0, |
| "loss": 1.466, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.4231908590774439, |
| "grad_norm": 4.519246578216553, |
| "learning_rate": 2.535532994923858e-05, |
| "loss": 1.4125, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.4231908590774439, |
| "eval_cosine_accuracy": 0.9784414768218994, |
| "eval_loss": 1.2133984565734863, |
| "eval_runtime": 22.3964, |
| "eval_samples_per_second": 424.577, |
| "eval_steps_per_second": 1.697, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.8463817181548878, |
| "grad_norm": 3.9520606994628906, |
| "learning_rate": 2.7697733471268692e-05, |
| "loss": 1.3526, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.8463817181548878, |
| "eval_cosine_accuracy": 0.9788621068000793, |
| "eval_loss": 1.1938709020614624, |
| "eval_runtime": 22.5907, |
| "eval_samples_per_second": 420.926, |
| "eval_steps_per_second": 1.682, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.2693446088794926, |
| "grad_norm": 4.076956748962402, |
| "learning_rate": 2.487632841154895e-05, |
| "loss": 1.2544, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.2693446088794926, |
| "eval_cosine_accuracy": 0.9801241159439087, |
| "eval_loss": 1.1857126951217651, |
| "eval_runtime": 22.6784, |
| "eval_samples_per_second": 419.298, |
| "eval_steps_per_second": 1.676, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.6921775898520086, |
| "grad_norm": 4.19527006149292, |
| "learning_rate": 2.2054923351829213e-05, |
| "loss": 1.2074, |
| "step": 4000 |
| }, |
| { |
| "epoch": 1.6921775898520086, |
| "eval_cosine_accuracy": 0.9784414768218994, |
| "eval_loss": 1.16515052318573, |
| "eval_runtime": 22.4384, |
| "eval_samples_per_second": 423.782, |
| "eval_steps_per_second": 1.694, |
| "step": 4000 |
| }, |
| { |
| "epoch": 2.1150105708245244, |
| "grad_norm": 3.546943187713623, |
| "learning_rate": 1.923351829210947e-05, |
| "loss": 1.1504, |
| "step": 5000 |
| }, |
| { |
| "epoch": 2.1150105708245244, |
| "eval_cosine_accuracy": 0.9799137711524963, |
| "eval_loss": 1.1423498392105103, |
| "eval_runtime": 22.7973, |
| "eval_samples_per_second": 417.11, |
| "eval_steps_per_second": 1.667, |
| "step": 5000 |
| }, |
| { |
| "epoch": 2.53784355179704, |
| "grad_norm": 3.9248595237731934, |
| "learning_rate": 1.641211323238973e-05, |
| "loss": 1.1051, |
| "step": 6000 |
| }, |
| { |
| "epoch": 2.53784355179704, |
| "eval_cosine_accuracy": 0.9798086285591125, |
| "eval_loss": 1.1364668607711792, |
| "eval_runtime": 23.9066, |
| "eval_samples_per_second": 397.756, |
| "eval_steps_per_second": 1.59, |
| "step": 6000 |
| }, |
| { |
| "epoch": 2.960676532769556, |
| "grad_norm": 3.992157459259033, |
| "learning_rate": 1.359070817266999e-05, |
| "loss": 1.0794, |
| "step": 7000 |
| }, |
| { |
| "epoch": 2.960676532769556, |
| "eval_cosine_accuracy": 0.9799137711524963, |
| "eval_loss": 1.1306134462356567, |
| "eval_runtime": 22.6082, |
| "eval_samples_per_second": 420.6, |
| "eval_steps_per_second": 1.681, |
| "step": 7000 |
| } |
| ], |
| "logging_steps": 1000, |
| "max_steps": 11815, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 0.0, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
