| { | |
| "best_metric": 0.9399143550352754, | |
| "best_model_checkpoint": "model_saves/xlnet-large_spell_5k_3_p3/checkpoint-268", | |
| "epoch": 4.0, | |
| "global_step": 1072, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9399143550352754, | |
| "eval_loss": 0.4382997751235962, | |
| "eval_runtime": 4.7784, | |
| "eval_samples_per_second": 917.037, | |
| "eval_steps_per_second": 7.325, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9397801853777434, | |
| "eval_loss": 0.4542422294616699, | |
| "eval_runtime": 4.7485, | |
| "eval_samples_per_second": 922.816, | |
| "eval_steps_per_second": 7.371, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9386732857031049, | |
| "eval_loss": 0.48965439200401306, | |
| "eval_runtime": 4.746, | |
| "eval_samples_per_second": 923.313, | |
| "eval_steps_per_second": 7.375, | |
| "step": 804 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2986, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9388521785798142, | |
| "eval_loss": 0.5420284271240234, | |
| "eval_runtime": 4.7408, | |
| "eval_samples_per_second": 924.321, | |
| "eval_steps_per_second": 7.383, | |
| "step": 1072 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 1072, | |
| "total_flos": 1.6970958350843904e+16, | |
| "train_loss": 0.2934352132811475, | |
| "train_runtime": 450.9419, | |
| "train_samples_per_second": 1141.078, | |
| "train_steps_per_second": 8.915 | |
| } | |
| ], | |
| "max_steps": 4020, | |
| "num_train_epochs": 15, | |
| "total_flos": 1.6970958350843904e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |