| { | |
| "best_metric": 0.03876456990838051, | |
| "best_model_checkpoint": "../model/xlm-roberta-large_8_5e-06_512_train_danish_5_512/tmp/checkpoint-138971", | |
| "epoch": 8.0, | |
| "global_step": 158824, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.0719, | |
| "step": 19853 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8061753891099582, | |
| "eval_loss": 0.05261862650513649, | |
| "eval_runtime": 485.7414, | |
| "eval_samples_per_second": 40.872, | |
| "eval_steps_per_second": 5.11, | |
| "step": 19853 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.0459, | |
| "step": 39706 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8427441696469048, | |
| "eval_loss": 0.04516961798071861, | |
| "eval_runtime": 485.7767, | |
| "eval_samples_per_second": 40.869, | |
| "eval_steps_per_second": 5.109, | |
| "step": 39706 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.5e-06, | |
| "loss": 0.0354, | |
| "step": 59559 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8623885558857604, | |
| "eval_loss": 0.041948962956666946, | |
| "eval_runtime": 484.5309, | |
| "eval_samples_per_second": 40.974, | |
| "eval_steps_per_second": 5.122, | |
| "step": 59559 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0277, | |
| "step": 79412 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8793129501838514, | |
| "eval_loss": 0.04084821045398712, | |
| "eval_runtime": 484.6834, | |
| "eval_samples_per_second": 40.961, | |
| "eval_steps_per_second": 5.121, | |
| "step": 79412 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.0217, | |
| "step": 99265 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8918047650229185, | |
| "eval_loss": 0.03935844823718071, | |
| "eval_runtime": 485.8351, | |
| "eval_samples_per_second": 40.864, | |
| "eval_steps_per_second": 5.109, | |
| "step": 99265 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.0171, | |
| "step": 119118 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.899712889739586, | |
| "eval_loss": 0.03974699229001999, | |
| "eval_runtime": 483.2863, | |
| "eval_samples_per_second": 41.079, | |
| "eval_steps_per_second": 5.136, | |
| "step": 119118 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.0137, | |
| "step": 138971 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9080743464463809, | |
| "eval_loss": 0.03876456990838051, | |
| "eval_runtime": 485.012, | |
| "eval_samples_per_second": 40.933, | |
| "eval_steps_per_second": 5.117, | |
| "step": 138971 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.0106, | |
| "step": 158824 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9115498917040246, | |
| "eval_loss": 0.03986608609557152, | |
| "eval_runtime": 484.9173, | |
| "eval_samples_per_second": 40.941, | |
| "eval_steps_per_second": 5.118, | |
| "step": 158824 | |
| } | |
| ], | |
| "max_steps": 198530, | |
| "num_train_epochs": 10, | |
| "total_flos": 1.1841805507553526e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |