| { | |
| "best_metric": 0.7068273092369478, | |
| "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-n64penuy/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": 1.4925928115844727, | |
| "learning_rate": 6.477989356097638e-06, | |
| "loss": 0.6661, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 2.3556463718414307, | |
| "learning_rate": 6.254610412783927e-06, | |
| "loss": 0.6128, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 2.2110280990600586, | |
| "learning_rate": 6.031231469470215e-06, | |
| "loss": 0.5869, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 1.7765781879425049, | |
| "learning_rate": 5.807852526156504e-06, | |
| "loss": 0.5578, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.7340786457061768, | |
| "learning_rate": 5.584473582842792e-06, | |
| "loss": 0.511, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 4.574438095092773, | |
| "learning_rate": 5.3610946395290804e-06, | |
| "loss": 0.4485, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 4.395756244659424, | |
| "learning_rate": 5.1377156962153684e-06, | |
| "loss": 0.5153, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.8996490240097046, | |
| "learning_rate": 4.9143367529016564e-06, | |
| "loss": 0.5166, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 3.7111058235168457, | |
| "learning_rate": 4.6909578095879444e-06, | |
| "loss": 0.4625, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.7154110670089722, | |
| "learning_rate": 4.467578866274233e-06, | |
| "loss": 0.5053, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7475, | |
| "eval_f1": 0.5809128630705395, | |
| "eval_loss": 0.48702481389045715, | |
| "eval_precision": 0.5932203389830508, | |
| "eval_recall": 0.5691056910569106, | |
| "eval_runtime": 1.5237, | |
| "eval_samples_per_second": 262.517, | |
| "eval_steps_per_second": 16.407, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 3.091020107269287, | |
| "learning_rate": 4.244199922960521e-06, | |
| "loss": 0.4949, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 2.2689990997314453, | |
| "learning_rate": 4.02082097964681e-06, | |
| "loss": 0.4326, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 4.819199085235596, | |
| "learning_rate": 3.797442036333098e-06, | |
| "loss": 0.4239, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 2.2359201908111572, | |
| "learning_rate": 3.5740630930193865e-06, | |
| "loss": 0.3589, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 3.3298685550689697, | |
| "learning_rate": 3.350684149705675e-06, | |
| "loss": 0.4728, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 7.007627010345459, | |
| "learning_rate": 3.1273052063919634e-06, | |
| "loss": 0.3661, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 5.745215892791748, | |
| "learning_rate": 2.903926263078252e-06, | |
| "loss": 0.3744, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 6.657752990722656, | |
| "learning_rate": 2.6805473197645402e-06, | |
| "loss": 0.3764, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 5.390446662902832, | |
| "learning_rate": 2.4571683764508282e-06, | |
| "loss": 0.374, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.511307716369629, | |
| "learning_rate": 2.2337894331371166e-06, | |
| "loss": 0.4045, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8175, | |
| "eval_f1": 0.7068273092369478, | |
| "eval_loss": 0.3974764943122864, | |
| "eval_precision": 0.6984126984126984, | |
| "eval_recall": 0.7154471544715447, | |
| "eval_runtime": 1.5366, | |
| "eval_samples_per_second": 260.318, | |
| "eval_steps_per_second": 16.27, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 423630740901888.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "learning_rate": 6.70136829941135e-06, | |
| "metric": "eval/loss", | |
| "num_train_epochs": 3, | |
| "per_device_train_batch_size": 16, | |
| "seed": 5 | |
| } | |
| } | |