| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 2228, | |
| "global_step": 4455, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6734006734006734, | |
| "grad_norm": 1.0852323426946855, | |
| "learning_rate": 4.768086838764333e-06, | |
| "loss": 0.348, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.282153457403183, | |
| "eval_runtime": 16.9042, | |
| "eval_samples_per_second": 28.395, | |
| "eval_steps_per_second": 1.775, | |
| "step": 1485 | |
| }, | |
| { | |
| "epoch": 1.3468013468013469, | |
| "grad_norm": 0.9716987928802957, | |
| "learning_rate": 3.3643505985087776e-06, | |
| "loss": 0.2655, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.2749512493610382, | |
| "eval_runtime": 16.2016, | |
| "eval_samples_per_second": 29.627, | |
| "eval_steps_per_second": 1.852, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 2.0202020202020203, | |
| "grad_norm": 0.9125202085812286, | |
| "learning_rate": 1.456442867057975e-06, | |
| "loss": 0.2184, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.6936026936026938, | |
| "grad_norm": 0.9699172328121177, | |
| "learning_rate": 1.5723692487014912e-07, | |
| "loss": 0.1382, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.29699257016181946, | |
| "eval_runtime": 16.1778, | |
| "eval_samples_per_second": 29.67, | |
| "eval_steps_per_second": 1.854, | |
| "step": 4455 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 4455, | |
| "tflops": 0.07104939945192446, | |
| "token/s": 964.1313936883596, | |
| "total_flos": 940586567729152.0, | |
| "train_loss": 0.23212839591115145, | |
| "train_runtime": 12724.6535, | |
| "train_samples_per_second": 11.203, | |
| "train_steps_per_second": 0.35 | |
| } | |
| ], | |
| "log_save_evaluate_time": 684.3207356929779, | |
| "logging_steps": 1000, | |
| "max_steps": 4455, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 1, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 940586567729152.0, | |
| "total_tokens": 11608464.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |