| { | |
| "best_metric": 1.1361786127090454, | |
| "best_model_checkpoint": "./vit-base-beans/checkpoint-480", | |
| "epoch": 3.6923076923076925, | |
| "global_step": 480, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00019897435897435898, | |
| "loss": 3.2519, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00019794871794871796, | |
| "loss": 3.1095, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.00019692307692307696, | |
| "loss": 2.9002, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0001958974358974359, | |
| "loss": 2.7917, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_accuracy": 0.35233160621761656, | |
| "eval_loss": 2.6724953651428223, | |
| "eval_runtime": 20.1054, | |
| "eval_samples_per_second": 57.596, | |
| "eval_steps_per_second": 7.212, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00019487179487179487, | |
| "loss": 2.6291, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00019384615384615385, | |
| "loss": 2.492, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00019282051282051282, | |
| "loss": 2.4302, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00019179487179487182, | |
| "loss": 2.3207, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_accuracy": 0.4481865284974093, | |
| "eval_loss": 2.263643980026245, | |
| "eval_runtime": 20.6875, | |
| "eval_samples_per_second": 55.976, | |
| "eval_steps_per_second": 7.009, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0001907692307692308, | |
| "loss": 2.1778, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00018974358974358974, | |
| "loss": 2.1215, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.0001887179487179487, | |
| "loss": 2.0333, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.0001876923076923077, | |
| "loss": 1.9258, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_accuracy": 0.5535405872193437, | |
| "eval_loss": 1.8725202083587646, | |
| "eval_runtime": 19.974, | |
| "eval_samples_per_second": 57.975, | |
| "eval_steps_per_second": 7.259, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0001866666666666667, | |
| "loss": 1.981, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00018564102564102566, | |
| "loss": 1.5576, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 0.00018461538461538463, | |
| "loss": 1.4327, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.00018358974358974358, | |
| "loss": 1.4339, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "eval_accuracy": 0.6260794473229706, | |
| "eval_loss": 1.5738734006881714, | |
| "eval_runtime": 20.2011, | |
| "eval_samples_per_second": 57.323, | |
| "eval_steps_per_second": 7.178, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.00018256410256410258, | |
| "loss": 1.2945, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.00018153846153846155, | |
| "loss": 1.3482, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.00018051282051282052, | |
| "loss": 1.2537, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.0001794871794871795, | |
| "loss": 1.3328, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "eval_accuracy": 0.6226252158894646, | |
| "eval_loss": 1.484718918800354, | |
| "eval_runtime": 20.1617, | |
| "eval_samples_per_second": 57.436, | |
| "eval_steps_per_second": 7.192, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.00017846153846153847, | |
| "loss": 1.1274, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00017743589743589744, | |
| "loss": 1.169, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 0.00017641025641025642, | |
| "loss": 1.1404, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.0001753846153846154, | |
| "loss": 1.0914, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_accuracy": 0.6442141623488774, | |
| "eval_loss": 1.3511847257614136, | |
| "eval_runtime": 19.8926, | |
| "eval_samples_per_second": 58.213, | |
| "eval_steps_per_second": 7.289, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.00017435897435897436, | |
| "loss": 1.1632, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00017333333333333334, | |
| "loss": 1.029, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 0.00017230769230769234, | |
| "loss": 0.7509, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.00017128205128205128, | |
| "loss": 0.6594, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "eval_accuracy": 0.6450777202072538, | |
| "eval_loss": 1.2636480331420898, | |
| "eval_runtime": 20.434, | |
| "eval_samples_per_second": 56.67, | |
| "eval_steps_per_second": 7.096, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 0.00017025641025641026, | |
| "loss": 0.557, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 0.00016923076923076923, | |
| "loss": 0.5061, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 0.00016820512820512823, | |
| "loss": 0.53, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 0.0001671794871794872, | |
| "loss": 0.5651, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "eval_accuracy": 0.6649395509499136, | |
| "eval_loss": 1.2330708503723145, | |
| "eval_runtime": 20.6893, | |
| "eval_samples_per_second": 55.971, | |
| "eval_steps_per_second": 7.008, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 0.00016615384615384617, | |
| "loss": 0.5417, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.00016512820512820512, | |
| "loss": 0.5584, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.0001641025641025641, | |
| "loss": 0.6261, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 0.0001630769230769231, | |
| "loss": 0.6847, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_accuracy": 0.6381692573402418, | |
| "eval_loss": 1.3150428533554077, | |
| "eval_runtime": 20.8049, | |
| "eval_samples_per_second": 55.66, | |
| "eval_steps_per_second": 6.97, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 0.00016205128205128207, | |
| "loss": 0.6079, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 0.00016102564102564104, | |
| "loss": 0.6087, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.00016, | |
| "loss": 0.6049, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 0.00015897435897435896, | |
| "loss": 0.2965, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "eval_accuracy": 0.6735751295336787, | |
| "eval_loss": 1.1951243877410889, | |
| "eval_runtime": 20.3022, | |
| "eval_samples_per_second": 57.038, | |
| "eval_steps_per_second": 7.142, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 0.00015794871794871796, | |
| "loss": 0.3409, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 0.00015692307692307693, | |
| "loss": 0.2707, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 0.0001558974358974359, | |
| "loss": 0.2439, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 0.00015487179487179488, | |
| "loss": 0.3052, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "eval_accuracy": 0.5915371329879102, | |
| "eval_loss": 1.4183249473571777, | |
| "eval_runtime": 19.9664, | |
| "eval_samples_per_second": 57.997, | |
| "eval_steps_per_second": 7.262, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 0.00015384615384615385, | |
| "loss": 0.2819, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 0.00015282051282051282, | |
| "loss": 0.2755, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 0.0001517948717948718, | |
| "loss": 0.2574, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 0.00015076923076923077, | |
| "loss": 0.2592, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "eval_accuracy": 0.697754749568221, | |
| "eval_loss": 1.1361786127090454, | |
| "eval_runtime": 19.9742, | |
| "eval_samples_per_second": 57.975, | |
| "eval_steps_per_second": 7.259, | |
| "step": 480 | |
| } | |
| ], | |
| "max_steps": 1950, | |
| "num_train_epochs": 15, | |
| "total_flos": 5.948067750708142e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |