| { | |
| "best_metric": 0.8, | |
| "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-fgzc4rms/checkpoint-200", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 200, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 4.183725833892822, | |
| "learning_rate": 7.142193434848993e-05, | |
| "loss": 0.5543, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 4.510015964508057, | |
| "learning_rate": 6.766288517225361e-05, | |
| "loss": 0.553, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 4.04560661315918, | |
| "learning_rate": 6.39038359960173e-05, | |
| "loss": 0.4519, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 4.4942779541015625, | |
| "learning_rate": 6.0144786819781e-05, | |
| "loss": 0.42, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 7.81617546081543, | |
| "learning_rate": 5.6385737643544685e-05, | |
| "loss": 0.3636, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 8.769570350646973, | |
| "learning_rate": 5.2626688467308364e-05, | |
| "loss": 0.419, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 7.014673233032227, | |
| "learning_rate": 4.886763929107206e-05, | |
| "loss": 0.4041, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 4.402252674102783, | |
| "learning_rate": 4.510859011483574e-05, | |
| "loss": 0.3462, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 4.1924614906311035, | |
| "learning_rate": 4.1349540938599435e-05, | |
| "loss": 0.3413, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 8.328794479370117, | |
| "learning_rate": 3.759049176236312e-05, | |
| "loss": 0.3807, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8525, | |
| "eval_f1": 0.7064676616915423, | |
| "eval_loss": 0.3551560342311859, | |
| "eval_precision": 0.9102564102564102, | |
| "eval_recall": 0.5772357723577236, | |
| "eval_runtime": 1.5598, | |
| "eval_samples_per_second": 256.449, | |
| "eval_steps_per_second": 16.028, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 4.5000457763671875, | |
| "learning_rate": 3.383144258612681e-05, | |
| "loss": 0.2533, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 3.0154271125793457, | |
| "learning_rate": 3.00723934098905e-05, | |
| "loss": 0.2507, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 14.534327507019043, | |
| "learning_rate": 2.6313344233654182e-05, | |
| "loss": 0.2524, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 3.4240000247955322, | |
| "learning_rate": 2.255429505741787e-05, | |
| "loss": 0.2401, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 5.545469760894775, | |
| "learning_rate": 1.879524588118156e-05, | |
| "loss": 0.2283, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 9.40932559967041, | |
| "learning_rate": 1.503619670494525e-05, | |
| "loss": 0.3299, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 5.235571384429932, | |
| "learning_rate": 1.1277147528708936e-05, | |
| "loss": 0.2728, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 2.9379899501800537, | |
| "learning_rate": 7.518098352472625e-06, | |
| "loss": 0.2671, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 7.245594024658203, | |
| "learning_rate": 3.7590491762363125e-06, | |
| "loss": 0.2036, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 6.384284496307373, | |
| "learning_rate": 0.0, | |
| "loss": 0.2773, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8775, | |
| "eval_f1": 0.8, | |
| "eval_loss": 0.31534862518310547, | |
| "eval_precision": 0.8032786885245902, | |
| "eval_recall": 0.7967479674796748, | |
| "eval_runtime": 1.5751, | |
| "eval_samples_per_second": 253.947, | |
| "eval_steps_per_second": 15.872, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 423630740901888.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "learning_rate": 7.518098352472624e-05, | |
| "metric": "eval/loss", | |
| "num_train_epochs": 2, | |
| "per_device_train_batch_size": 16, | |
| "seed": 22 | |
| } | |
| } | |