| { | |
| "best_metric": 0.8692817116658176, | |
| "best_model_checkpoint": "./outputs/xlnet-base-cased/fine_tuned_models/checkpoint-61360", | |
| "epoch": 5.0, | |
| "global_step": 61360, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.7388593694651082e-05, | |
| "loss": 0.5389, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8557310239429444, | |
| "eval_loss": 0.38328248262405396, | |
| "eval_runtime": 36.0916, | |
| "eval_samples_per_second": 271.947, | |
| "eval_steps_per_second": 33.997, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.3041445270988312e-05, | |
| "loss": 0.3481, | |
| "step": 24544 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8628629648497198, | |
| "eval_loss": 0.38958263397216797, | |
| "eval_runtime": 36.2238, | |
| "eval_samples_per_second": 270.955, | |
| "eval_steps_per_second": 33.873, | |
| "step": 24544 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.694296847325541e-06, | |
| "loss": 0.2489, | |
| "step": 36816 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8630667345899133, | |
| "eval_loss": 0.3986847400665283, | |
| "eval_runtime": 36.0981, | |
| "eval_samples_per_second": 271.898, | |
| "eval_steps_per_second": 33.991, | |
| "step": 36816 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.3471484236627705e-06, | |
| "loss": 0.1777, | |
| "step": 49088 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8691798267957208, | |
| "eval_loss": 0.46929463744163513, | |
| "eval_runtime": 36.2027, | |
| "eval_samples_per_second": 271.112, | |
| "eval_steps_per_second": 33.892, | |
| "step": 49088 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.1303, | |
| "step": 61360 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8692817116658176, | |
| "eval_loss": 0.5496333837509155, | |
| "eval_runtime": 36.1007, | |
| "eval_samples_per_second": 271.879, | |
| "eval_steps_per_second": 33.988, | |
| "step": 61360 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 61360, | |
| "total_flos": 1.769024893707648e+17, | |
| "train_loss": 0.28879537967846014, | |
| "train_runtime": 20254.0743, | |
| "train_samples_per_second": 96.944, | |
| "train_steps_per_second": 3.03 | |
| } | |
| ], | |
| "max_steps": 61360, | |
| "num_train_epochs": 5, | |
| "total_flos": 1.769024893707648e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |