| { | |
| "best_metric": 0.03965451195836067, | |
| "best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-500", | |
| "epoch": 1.0, | |
| "eval_steps": 100, | |
| "global_step": 596, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 9.83221476510067e-07, | |
| "loss": 0.0019, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 9.664429530201342e-07, | |
| "loss": 0.002, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.496644295302014e-07, | |
| "loss": 0.0037, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.328859060402684e-07, | |
| "loss": 0.0039, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.161073825503355e-07, | |
| "loss": 0.0204, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 8.993288590604026e-07, | |
| "loss": 0.0304, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 8.825503355704698e-07, | |
| "loss": 0.0142, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 8.657718120805368e-07, | |
| "loss": 0.038, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 8.48993288590604e-07, | |
| "loss": 0.0019, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.322147651006712e-07, | |
| "loss": 0.004, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_accuracy": 0.9903442485306465, | |
| "eval_loss": 0.04405502602458, | |
| "eval_runtime": 57.4827, | |
| "eval_samples_per_second": 41.439, | |
| "eval_steps_per_second": 5.184, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 8.154362416107382e-07, | |
| "loss": 0.0017, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 7.986577181208053e-07, | |
| "loss": 0.0018, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 7.818791946308725e-07, | |
| "loss": 0.0017, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 7.651006711409396e-07, | |
| "loss": 0.0182, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 7.483221476510066e-07, | |
| "loss": 0.0017, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 7.315436241610739e-07, | |
| "loss": 0.011, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7.147651006711409e-07, | |
| "loss": 0.0018, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 6.97986577181208e-07, | |
| "loss": 0.0065, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 6.81208053691275e-07, | |
| "loss": 0.0019, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 6.644295302013423e-07, | |
| "loss": 0.0025, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_accuracy": 0.9903442485306465, | |
| "eval_loss": 0.04235763102769852, | |
| "eval_runtime": 58.8296, | |
| "eval_samples_per_second": 40.49, | |
| "eval_steps_per_second": 5.065, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 6.476510067114094e-07, | |
| "loss": 0.0017, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 6.308724832214764e-07, | |
| "loss": 0.0016, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 6.140939597315436e-07, | |
| "loss": 0.0069, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 5.973154362416107e-07, | |
| "loss": 0.005, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 5.805369127516778e-07, | |
| "loss": 0.0021, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 5.637583892617449e-07, | |
| "loss": 0.0015, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 5.469798657718121e-07, | |
| "loss": 0.0114, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 5.302013422818791e-07, | |
| "loss": 0.0139, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 5.134228187919463e-07, | |
| "loss": 0.0046, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.966442953020134e-07, | |
| "loss": 0.0019, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_accuracy": 0.9907640638119227, | |
| "eval_loss": 0.04056321829557419, | |
| "eval_runtime": 58.6218, | |
| "eval_samples_per_second": 40.633, | |
| "eval_steps_per_second": 5.083, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4.798657718120805e-07, | |
| "loss": 0.002, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.630872483221476e-07, | |
| "loss": 0.0034, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.4630872483221475e-07, | |
| "loss": 0.0092, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.2953020134228183e-07, | |
| "loss": 0.0021, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 4.12751677852349e-07, | |
| "loss": 0.0015, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 3.959731543624161e-07, | |
| "loss": 0.0383, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 3.791946308724832e-07, | |
| "loss": 0.0292, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 3.624161073825503e-07, | |
| "loss": 0.0072, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 3.4563758389261744e-07, | |
| "loss": 0.0278, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 3.2885906040268457e-07, | |
| "loss": 0.0015, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_accuracy": 0.9916036943744753, | |
| "eval_loss": 0.039836905896663666, | |
| "eval_runtime": 58.6815, | |
| "eval_samples_per_second": 40.592, | |
| "eval_steps_per_second": 5.078, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 3.1208053691275165e-07, | |
| "loss": 0.0025, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 2.953020134228188e-07, | |
| "loss": 0.0036, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 2.7852348993288586e-07, | |
| "loss": 0.0014, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 2.61744966442953e-07, | |
| "loss": 0.0042, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 2.4496644295302013e-07, | |
| "loss": 0.0013, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 2.2818791946308723e-07, | |
| "loss": 0.0043, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 2.1140939597315434e-07, | |
| "loss": 0.0034, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 1.9463087248322147e-07, | |
| "loss": 0.0022, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.7785234899328858e-07, | |
| "loss": 0.0035, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 1.6107382550335569e-07, | |
| "loss": 0.0019, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_accuracy": 0.9916036943744753, | |
| "eval_loss": 0.03965451195836067, | |
| "eval_runtime": 59.706, | |
| "eval_samples_per_second": 39.895, | |
| "eval_steps_per_second": 4.991, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1.4429530201342282e-07, | |
| "loss": 0.0018, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 1.2751677852348992e-07, | |
| "loss": 0.0085, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.1073825503355704e-07, | |
| "loss": 0.0015, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 9.395973154362416e-08, | |
| "loss": 0.0016, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 7.718120805369127e-08, | |
| "loss": 0.0019, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 6.040268456375839e-08, | |
| "loss": 0.0036, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 4.36241610738255e-08, | |
| "loss": 0.0084, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 2.6845637583892614e-08, | |
| "loss": 0.0015, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 1.006711409395973e-08, | |
| "loss": 0.0017, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 596, | |
| "total_flos": 2.610159532500861e+18, | |
| "train_loss": 0.006745032241080431, | |
| "train_runtime": 1079.8985, | |
| "train_samples_per_second": 8.823, | |
| "train_steps_per_second": 0.552 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 596, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 2.610159532500861e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |