| { | |
| "best_metric": 0.90932, | |
| "best_model_checkpoint": "../../checkpoint/imdb/electra-base/checkpoint-14076", | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 15640, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.89252, | |
| "eval_loss": 0.2527729868888855, | |
| "eval_runtime": 33.8056, | |
| "eval_samples_per_second": 739.522, | |
| "eval_steps_per_second": 2.899, | |
| "step": 782 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 4.680306905370844e-05, | |
| "loss": 0.272, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.90448, | |
| "eval_loss": 0.2465250939130783, | |
| "eval_runtime": 33.8629, | |
| "eval_samples_per_second": 738.272, | |
| "eval_steps_per_second": 2.894, | |
| "step": 1564 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 4.360613810741688e-05, | |
| "loss": 0.1451, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9034, | |
| "eval_loss": 0.39296430349349976, | |
| "eval_runtime": 33.8123, | |
| "eval_samples_per_second": 739.377, | |
| "eval_steps_per_second": 2.898, | |
| "step": 2346 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 4.040920716112532e-05, | |
| "loss": 0.0804, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.898, | |
| "eval_loss": 0.45841386914253235, | |
| "eval_runtime": 33.8469, | |
| "eval_samples_per_second": 738.621, | |
| "eval_steps_per_second": 2.895, | |
| "step": 3128 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9056, | |
| "eval_loss": 0.5110316872596741, | |
| "eval_runtime": 33.885, | |
| "eval_samples_per_second": 737.789, | |
| "eval_steps_per_second": 2.892, | |
| "step": 3910 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 3.721227621483376e-05, | |
| "loss": 0.0499, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.89784, | |
| "eval_loss": 0.5765578746795654, | |
| "eval_runtime": 33.9081, | |
| "eval_samples_per_second": 737.288, | |
| "eval_steps_per_second": 2.89, | |
| "step": 4692 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 3.40153452685422e-05, | |
| "loss": 0.0334, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.89992, | |
| "eval_loss": 0.5621291995048523, | |
| "eval_runtime": 34.0076, | |
| "eval_samples_per_second": 735.13, | |
| "eval_steps_per_second": 2.882, | |
| "step": 5474 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 3.081841432225064e-05, | |
| "loss": 0.0243, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9016, | |
| "eval_loss": 0.6355283856391907, | |
| "eval_runtime": 33.9931, | |
| "eval_samples_per_second": 735.443, | |
| "eval_steps_per_second": 2.883, | |
| "step": 6256 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "learning_rate": 2.7621483375959077e-05, | |
| "loss": 0.0195, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.90184, | |
| "eval_loss": 0.6819501519203186, | |
| "eval_runtime": 33.7784, | |
| "eval_samples_per_second": 740.119, | |
| "eval_steps_per_second": 2.901, | |
| "step": 7038 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.90152, | |
| "eval_loss": 0.6723979115486145, | |
| "eval_runtime": 33.9379, | |
| "eval_samples_per_second": 736.639, | |
| "eval_steps_per_second": 2.888, | |
| "step": 7820 | |
| }, | |
| { | |
| "epoch": 10.23, | |
| "learning_rate": 2.442455242966752e-05, | |
| "loss": 0.0132, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.90308, | |
| "eval_loss": 0.7561084628105164, | |
| "eval_runtime": 33.739, | |
| "eval_samples_per_second": 740.982, | |
| "eval_steps_per_second": 2.905, | |
| "step": 8602 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "learning_rate": 2.122762148337596e-05, | |
| "loss": 0.0095, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.90576, | |
| "eval_loss": 0.7514671087265015, | |
| "eval_runtime": 33.7517, | |
| "eval_samples_per_second": 740.704, | |
| "eval_steps_per_second": 2.904, | |
| "step": 9384 | |
| }, | |
| { | |
| "epoch": 12.79, | |
| "learning_rate": 1.80306905370844e-05, | |
| "loss": 0.0072, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.90604, | |
| "eval_loss": 0.7862638831138611, | |
| "eval_runtime": 33.8938, | |
| "eval_samples_per_second": 737.597, | |
| "eval_steps_per_second": 2.891, | |
| "step": 10166 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.90584, | |
| "eval_loss": 0.7231158018112183, | |
| "eval_runtime": 33.9086, | |
| "eval_samples_per_second": 737.276, | |
| "eval_steps_per_second": 2.89, | |
| "step": 10948 | |
| }, | |
| { | |
| "epoch": 14.07, | |
| "learning_rate": 1.483375959079284e-05, | |
| "loss": 0.0067, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.90152, | |
| "eval_loss": 0.8694790005683899, | |
| "eval_runtime": 33.8199, | |
| "eval_samples_per_second": 739.211, | |
| "eval_steps_per_second": 2.898, | |
| "step": 11730 | |
| }, | |
| { | |
| "epoch": 15.35, | |
| "learning_rate": 1.163682864450128e-05, | |
| "loss": 0.0037, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.90492, | |
| "eval_loss": 0.8443425297737122, | |
| "eval_runtime": 33.8917, | |
| "eval_samples_per_second": 737.644, | |
| "eval_steps_per_second": 2.892, | |
| "step": 12512 | |
| }, | |
| { | |
| "epoch": 16.62, | |
| "learning_rate": 8.439897698209718e-06, | |
| "loss": 0.0033, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.90808, | |
| "eval_loss": 0.8232758641242981, | |
| "eval_runtime": 33.8581, | |
| "eval_samples_per_second": 738.376, | |
| "eval_steps_per_second": 2.894, | |
| "step": 13294 | |
| }, | |
| { | |
| "epoch": 17.9, | |
| "learning_rate": 5.242966751918159e-06, | |
| "loss": 0.0019, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.90932, | |
| "eval_loss": 0.8650785088539124, | |
| "eval_runtime": 33.9491, | |
| "eval_samples_per_second": 736.398, | |
| "eval_steps_per_second": 2.887, | |
| "step": 14076 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.90856, | |
| "eval_loss": 0.8838527202606201, | |
| "eval_runtime": 33.9568, | |
| "eval_samples_per_second": 736.23, | |
| "eval_steps_per_second": 2.886, | |
| "step": 14858 | |
| }, | |
| { | |
| "epoch": 19.18, | |
| "learning_rate": 2.0460358056265987e-06, | |
| "loss": 0.0003, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.908, | |
| "eval_loss": 0.8944087624549866, | |
| "eval_runtime": 33.9267, | |
| "eval_samples_per_second": 736.882, | |
| "eval_steps_per_second": 2.889, | |
| "step": 15640 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 15640, | |
| "total_flos": 3.288888192e+16, | |
| "train_loss": 0.0428766398688259, | |
| "train_runtime": 2610.1004, | |
| "train_samples_per_second": 191.564, | |
| "train_steps_per_second": 5.992 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 15640, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 3.288888192e+16, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |