| { | |
| "best_metric": 0.8678611516952515, | |
| "best_model_checkpoint": "./xlm-roberta-base/fine_tuned_models_en/checkpoint-6215", | |
| "epoch": 5.0, | |
| "global_step": 6215, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.777936706597533e-05, | |
| "loss": 0.5217, | |
| "step": 1243 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.828219473361969, | |
| "eval_loss": 0.3914763927459717, | |
| "eval_runtime": 20.0966, | |
| "eval_samples_per_second": 222.177, | |
| "eval_steps_per_second": 27.816, | |
| "step": 1243 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.3334525299481496e-05, | |
| "loss": 0.3298, | |
| "step": 2486 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8600224256515503, | |
| "eval_loss": 0.3453393578529358, | |
| "eval_runtime": 20.0863, | |
| "eval_samples_per_second": 222.291, | |
| "eval_steps_per_second": 27.83, | |
| "step": 2486 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.889683532987665e-06, | |
| "loss": 0.2401, | |
| "step": 3729 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8629339337348938, | |
| "eval_loss": 0.36967143416404724, | |
| "eval_runtime": 20.0753, | |
| "eval_samples_per_second": 222.413, | |
| "eval_steps_per_second": 27.845, | |
| "step": 3729 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.444841766493832e-06, | |
| "loss": 0.1766, | |
| "step": 4972 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8662933707237244, | |
| "eval_loss": 0.4054161012172699, | |
| "eval_runtime": 20.0933, | |
| "eval_samples_per_second": 222.213, | |
| "eval_steps_per_second": 27.82, | |
| "step": 4972 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.1353, | |
| "step": 6215 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8678611516952515, | |
| "eval_loss": 0.4647606313228607, | |
| "eval_runtime": 20.1327, | |
| "eval_samples_per_second": 221.779, | |
| "eval_steps_per_second": 27.766, | |
| "step": 6215 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 6215, | |
| "total_flos": 1.30795794507648e+16, | |
| "train_loss": 0.28068858704448035, | |
| "train_runtime": 2813.7427, | |
| "train_samples_per_second": 70.669, | |
| "train_steps_per_second": 2.209 | |
| } | |
| ], | |
| "max_steps": 6215, | |
| "num_train_epochs": 5, | |
| "total_flos": 1.30795794507648e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |