| { | |
| "best_metric": 0.9975288303130149, | |
| "best_model_checkpoint": "ai_detector/checkpoint-843", | |
| "epoch": 2.997333333333333, | |
| "eval_steps": 500, | |
| "global_step": 843, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 5.882352941176471e-06, | |
| "loss": 0.6876, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.1764705882352942e-05, | |
| "loss": 0.4009, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.7647058823529414e-05, | |
| "loss": 0.2111, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 2.3529411764705884e-05, | |
| "loss": 0.0845, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 2.9411764705882354e-05, | |
| "loss": 0.052, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.529411764705883e-05, | |
| "loss": 0.0831, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.11764705882353e-05, | |
| "loss": 0.0719, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 4.705882352941177e-05, | |
| "loss": 0.0493, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.967018469656992e-05, | |
| "loss": 0.0278, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.901055408970976e-05, | |
| "loss": 0.0447, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 4.835092348284961e-05, | |
| "loss": 0.0597, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 4.7691292875989446e-05, | |
| "loss": 0.0194, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.703166226912929e-05, | |
| "loss": 0.0308, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.6372031662269136e-05, | |
| "loss": 0.0328, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.5712401055408974e-05, | |
| "loss": 0.0415, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.505277044854881e-05, | |
| "loss": 0.0075, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.439313984168866e-05, | |
| "loss": 0.0358, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 4.3733509234828496e-05, | |
| "loss": 0.0684, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 4.307387862796834e-05, | |
| "loss": 0.0382, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.2414248021108186e-05, | |
| "loss": 0.0252, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.1754617414248024e-05, | |
| "loss": 0.0356, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.109498680738786e-05, | |
| "loss": 0.0214, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.043535620052771e-05, | |
| "loss": 0.0137, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3.9775725593667545e-05, | |
| "loss": 0.091, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 3.911609498680739e-05, | |
| "loss": 0.0103, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 3.8456464379947235e-05, | |
| "loss": 0.0418, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 3.779683377308707e-05, | |
| "loss": 0.0089, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.713720316622691e-05, | |
| "loss": 0.0347, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9964804552942939, | |
| "eval_loss": 0.013468941673636436, | |
| "eval_runtime": 252.3562, | |
| "eval_samples_per_second": 52.917, | |
| "eval_steps_per_second": 3.309, | |
| "step": 281 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 3.6477572559366756e-05, | |
| "loss": 0.0258, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 3.58179419525066e-05, | |
| "loss": 0.0297, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 3.515831134564644e-05, | |
| "loss": 0.029, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 3.4498680738786285e-05, | |
| "loss": 0.0023, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 3.383905013192612e-05, | |
| "loss": 0.013, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 3.317941952506596e-05, | |
| "loss": 0.0203, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 3.2519788918205806e-05, | |
| "loss": 0.0286, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 3.186015831134565e-05, | |
| "loss": 0.0236, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 3.120052770448549e-05, | |
| "loss": 0.0126, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 3.0540897097625334e-05, | |
| "loss": 0.02, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 2.9881266490765176e-05, | |
| "loss": 0.0215, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 2.9221635883905014e-05, | |
| "loss": 0.0143, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 2.8562005277044855e-05, | |
| "loss": 0.0054, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 2.79023746701847e-05, | |
| "loss": 0.0072, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 2.724274406332454e-05, | |
| "loss": 0.005, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 2.658311345646438e-05, | |
| "loss": 0.0142, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 2.5923482849604225e-05, | |
| "loss": 0.0155, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 2.5263852242744063e-05, | |
| "loss": 0.0109, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 2.4604221635883905e-05, | |
| "loss": 0.0073, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 2.3944591029023746e-05, | |
| "loss": 0.0126, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 2.328496042216359e-05, | |
| "loss": 0.0154, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 2.262532981530343e-05, | |
| "loss": 0.0065, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 2.196569920844327e-05, | |
| "loss": 0.0275, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 2.1306068601583116e-05, | |
| "loss": 0.0296, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 2.0646437994722954e-05, | |
| "loss": 0.0043, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 1.9986807387862796e-05, | |
| "loss": 0.0012, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 1.932717678100264e-05, | |
| "loss": 0.0044, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 1.866754617414248e-05, | |
| "loss": 0.0009, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9942339373970346, | |
| "eval_loss": 0.02113017998635769, | |
| "eval_runtime": 251.3615, | |
| "eval_samples_per_second": 53.127, | |
| "eval_steps_per_second": 3.322, | |
| "step": 562 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 1.8007915567282324e-05, | |
| "loss": 0.002, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 1.7348284960422166e-05, | |
| "loss": 0.0154, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 1.6688654353562007e-05, | |
| "loss": 0.0058, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 1.602902374670185e-05, | |
| "loss": 0.0066, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 1.536939313984169e-05, | |
| "loss": 0.0007, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 1.470976253298153e-05, | |
| "loss": 0.0008, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 1.4050131926121373e-05, | |
| "loss": 0.0037, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 1.3390501319261215e-05, | |
| "loss": 0.0057, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 1.2730870712401055e-05, | |
| "loss": 0.0002, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 1.2071240105540896e-05, | |
| "loss": 0.0241, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 1.141160949868074e-05, | |
| "loss": 0.0007, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 1.0751978891820581e-05, | |
| "loss": 0.0058, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 1.0092348284960421e-05, | |
| "loss": 0.0003, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 9.432717678100264e-06, | |
| "loss": 0.0075, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 8.773087071240106e-06, | |
| "loss": 0.001, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 8.113456464379948e-06, | |
| "loss": 0.002, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 7.453825857519789e-06, | |
| "loss": 0.0125, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 6.794195250659631e-06, | |
| "loss": 0.0001, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 6.134564643799472e-06, | |
| "loss": 0.0003, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 5.474934036939315e-06, | |
| "loss": 0.001, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 4.8153034300791555e-06, | |
| "loss": 0.0001, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 4.155672823218998e-06, | |
| "loss": 0.0078, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 3.496042216358839e-06, | |
| "loss": 0.0031, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 2.836411609498681e-06, | |
| "loss": 0.0264, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 2.1767810026385226e-06, | |
| "loss": 0.0124, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 1.5171503957783642e-06, | |
| "loss": 0.0195, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 8.575197889182058e-07, | |
| "loss": 0.005, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 1.9788918205804752e-07, | |
| "loss": 0.0241, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9975288303130149, | |
| "eval_loss": 0.010735484771430492, | |
| "eval_runtime": 284.9371, | |
| "eval_samples_per_second": 46.866, | |
| "eval_steps_per_second": 2.93, | |
| "step": 843 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 843, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 4.2268994172435825e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |