{ "best_metric": 0.7607843137254902, "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-0i7sidmz/checkpoint-100", "epoch": 1.0, "eval_steps": 500, "global_step": 100, "is_hyper_param_search": true, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.1, "grad_norm": 4.02569055557251, "learning_rate": 5.021793617667588e-05, "loss": 0.5846, "step": 10 }, { "epoch": 0.2, "grad_norm": 2.9225821495056152, "learning_rate": 4.4638165490378564e-05, "loss": 0.4403, "step": 20 }, { "epoch": 0.3, "grad_norm": 3.3444550037384033, "learning_rate": 3.905839480408124e-05, "loss": 0.6524, "step": 30 }, { "epoch": 0.4, "grad_norm": 3.390305280685425, "learning_rate": 3.3478624117783916e-05, "loss": 0.477, "step": 40 }, { "epoch": 0.5, "grad_norm": 4.890294551849365, "learning_rate": 2.78988534314866e-05, "loss": 0.4469, "step": 50 }, { "epoch": 0.6, "grad_norm": 3.794447183609009, "learning_rate": 2.2319082745189282e-05, "loss": 0.4012, "step": 60 }, { "epoch": 0.7, "grad_norm": 5.297450065612793, "learning_rate": 1.6739312058891958e-05, "loss": 0.396, "step": 70 }, { "epoch": 0.8, "grad_norm": 3.4959356784820557, "learning_rate": 1.1159541372594641e-05, "loss": 0.4017, "step": 80 }, { "epoch": 0.9, "grad_norm": 4.851554870605469, "learning_rate": 5.5797706862973205e-06, "loss": 0.4157, "step": 90 }, { "epoch": 1.0, "grad_norm": 4.022997856140137, "learning_rate": 0.0, "loss": 0.3782, "step": 100 }, { "epoch": 1.0, "eval_accuracy": 0.8475, "eval_f1": 0.7607843137254902, "eval_loss": 0.3398902416229248, "eval_precision": 0.7348484848484849, "eval_recall": 0.7886178861788617, "eval_runtime": 1.5686, "eval_samples_per_second": 255.004, "eval_steps_per_second": 15.938, "step": 100 } ], "logging_steps": 10, "max_steps": 100, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 211815370450944.0, "train_batch_size": 16, "trial_name": null, "trial_params": { "_wandb": {}, "assignments": {}, "learning_rate": 5.57977068629732e-05, "metric": "eval/loss", "num_train_epochs": 1, "per_device_train_batch_size": 16, "seed": 28 } }