| { | |
| "best_metric": 1.0281891822814941, | |
| "best_model_checkpoint": "miner_id_24/checkpoint-10", | |
| "epoch": 0.004227882887644013, | |
| "eval_steps": 5, | |
| "global_step": 10, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00042278828876440123, | |
| "eval_loss": 1.1933761835098267, | |
| "eval_runtime": 135.4339, | |
| "eval_samples_per_second": 7.354, | |
| "eval_steps_per_second": 3.677, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0012683648662932037, | |
| "grad_norm": 11.284184455871582, | |
| "learning_rate": 6e-05, | |
| "loss": 4.4735, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0021139414438220063, | |
| "eval_loss": 1.0962536334991455, | |
| "eval_runtime": 136.7913, | |
| "eval_samples_per_second": 7.281, | |
| "eval_steps_per_second": 3.641, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0025367297325864075, | |
| "grad_norm": 10.291092872619629, | |
| "learning_rate": 0.00012, | |
| "loss": 4.5116, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.003805094598879611, | |
| "grad_norm": 12.492044448852539, | |
| "learning_rate": 0.00018, | |
| "loss": 4.0044, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.004227882887644013, | |
| "eval_loss": 1.0281891822814941, | |
| "eval_runtime": 136.8737, | |
| "eval_samples_per_second": 7.277, | |
| "eval_steps_per_second": 3.638, | |
| "step": 10 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 1, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3867216671932416.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |