| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.4011026878015162, | |
| "eval_steps": 291, | |
| "global_step": 582, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06891798759476224, | |
| "grad_norm": 4.408621311187744, | |
| "learning_rate": 9.92330596776479e-06, | |
| "loss": 0.6326, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13783597518952448, | |
| "grad_norm": 5.029026985168457, | |
| "learning_rate": 9.62094968320582e-06, | |
| "loss": 0.5365, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2005513439007581, | |
| "eval_accuracy": 0.7271186845182692, | |
| "eval_loss": 0.5419921875, | |
| "eval_runtime": 641.9062, | |
| "eval_samples_per_second": 130.552, | |
| "eval_steps_per_second": 8.16, | |
| "step": 291 | |
| }, | |
| { | |
| "epoch": 0.2067539627842867, | |
| "grad_norm": 5.613715648651123, | |
| "learning_rate": 9.102819483545054e-06, | |
| "loss": 0.4928, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.27567195037904896, | |
| "grad_norm": 5.0579833984375, | |
| "learning_rate": 8.39310930928775e-06, | |
| "loss": 0.4794, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.34458993797381116, | |
| "grad_norm": 6.232191562652588, | |
| "learning_rate": 7.524958872697738e-06, | |
| "loss": 0.4521, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4011026878015162, | |
| "eval_accuracy": 0.7484546908188349, | |
| "eval_loss": 0.50341796875, | |
| "eval_runtime": 627.5856, | |
| "eval_samples_per_second": 133.531, | |
| "eval_steps_per_second": 8.346, | |
| "step": 582 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1451, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 291, | |
| "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": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |