| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 1000, | |
| "global_step": 3122, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00032030749519538755, | |
| "grad_norm": 6.058427333831787, | |
| "learning_rate": 0.0, | |
| "loss": 2.3526, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.3203074951953876, | |
| "grad_norm": 3.9029946327209473, | |
| "learning_rate": 2.977933325424131e-05, | |
| "loss": 1.6965, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3203074951953876, | |
| "eval_cosine_accuracy": 0.9661373496055603, | |
| "eval_loss": 0.7820664644241333, | |
| "eval_runtime": 28.1656, | |
| "eval_samples_per_second": 337.61, | |
| "eval_steps_per_second": 1.349, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6406149903907752, | |
| "grad_norm": 3.5058798789978027, | |
| "learning_rate": 2.6220192193617276e-05, | |
| "loss": 1.1157, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.6406149903907752, | |
| "eval_cosine_accuracy": 0.9707645177841187, | |
| "eval_loss": 0.7116907835006714, | |
| "eval_runtime": 27.4286, | |
| "eval_samples_per_second": 346.681, | |
| "eval_steps_per_second": 1.385, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9609224855861627, | |
| "grad_norm": 3.5930891036987305, | |
| "learning_rate": 2.2661051132993238e-05, | |
| "loss": 1.5779, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.9609224855861627, | |
| "eval_cosine_accuracy": 0.9699232578277588, | |
| "eval_loss": 0.7110825777053833, | |
| "eval_runtime": 27.7293, | |
| "eval_samples_per_second": 342.922, | |
| "eval_steps_per_second": 1.37, | |
| "step": 3000 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 9366, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |