| { | |
| "best_metric": 1.8420542478561401, | |
| "best_model_checkpoint": "./vit-base-beans/checkpoint-640", | |
| "epoch": 3.248730964467005, | |
| "global_step": 640, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00019932318104906937, | |
| "loss": 3.8192, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00019864636209813877, | |
| "loss": 3.7847, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00019796954314720813, | |
| "loss": 3.6849, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00019729272419627753, | |
| "loss": 3.6122, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.10967741935483871, | |
| "eval_loss": 3.561206340789795, | |
| "eval_runtime": 31.2002, | |
| "eval_samples_per_second": 54.647, | |
| "eval_steps_per_second": 6.859, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00019661590524534686, | |
| "loss": 3.5748, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00019593908629441626, | |
| "loss": 3.491, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00019526226734348562, | |
| "loss": 3.3556, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.000194585448392555, | |
| "loss": 3.4008, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_accuracy": 0.2475073313782991, | |
| "eval_loss": 3.2414700984954834, | |
| "eval_runtime": 30.544, | |
| "eval_samples_per_second": 55.821, | |
| "eval_steps_per_second": 7.006, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00019390862944162438, | |
| "loss": 3.2255, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00019323181049069374, | |
| "loss": 3.1171, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00019255499153976314, | |
| "loss": 3.1394, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.0001918781725888325, | |
| "loss": 3.0872, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_accuracy": 0.2809384164222874, | |
| "eval_loss": 2.9989874362945557, | |
| "eval_runtime": 30.8814, | |
| "eval_samples_per_second": 55.211, | |
| "eval_steps_per_second": 6.93, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00019120135363790187, | |
| "loss": 2.9773, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00019052453468697123, | |
| "loss": 2.7993, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.0001898477157360406, | |
| "loss": 2.8672, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.00018917089678511, | |
| "loss": 2.8733, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "eval_accuracy": 0.34252199413489737, | |
| "eval_loss": 2.7833735942840576, | |
| "eval_runtime": 30.3566, | |
| "eval_samples_per_second": 56.166, | |
| "eval_steps_per_second": 7.05, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.00018849407783417936, | |
| "loss": 2.7912, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.00018781725888324875, | |
| "loss": 2.7577, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.00018714043993231811, | |
| "loss": 2.7865, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.00018646362098138748, | |
| "loss": 2.591, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_accuracy": 0.37360703812316715, | |
| "eval_loss": 2.6250743865966797, | |
| "eval_runtime": 30.4336, | |
| "eval_samples_per_second": 56.024, | |
| "eval_steps_per_second": 7.032, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.00018578680203045687, | |
| "loss": 2.3103, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.00018510998307952624, | |
| "loss": 2.2942, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.0001844331641285956, | |
| "loss": 2.3883, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.00018375634517766497, | |
| "loss": 2.1643, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "eval_accuracy": 0.40527859237536656, | |
| "eval_loss": 2.423466920852661, | |
| "eval_runtime": 30.3039, | |
| "eval_samples_per_second": 56.263, | |
| "eval_steps_per_second": 7.062, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 0.00018307952622673436, | |
| "loss": 2.2077, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 0.00018240270727580373, | |
| "loss": 2.1227, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 0.0001817258883248731, | |
| "loss": 2.0657, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.00018104906937394248, | |
| "loss": 2.0669, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_accuracy": 0.3882697947214076, | |
| "eval_loss": 2.3226494789123535, | |
| "eval_runtime": 30.2544, | |
| "eval_samples_per_second": 56.355, | |
| "eval_steps_per_second": 7.073, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.00018037225042301185, | |
| "loss": 2.1189, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.00017969543147208124, | |
| "loss": 2.1857, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 0.0001790186125211506, | |
| "loss": 2.272, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.00017834179357021997, | |
| "loss": 1.9717, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "eval_accuracy": 0.40821114369501466, | |
| "eval_loss": 2.2509706020355225, | |
| "eval_runtime": 30.2595, | |
| "eval_samples_per_second": 56.346, | |
| "eval_steps_per_second": 7.072, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.00017766497461928934, | |
| "loss": 2.0974, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.0001769881556683587, | |
| "loss": 2.1316, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.0001763113367174281, | |
| "loss": 2.0845, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 0.00017563451776649746, | |
| "loss": 2.1714, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_accuracy": 0.4351906158357771, | |
| "eval_loss": 2.203674077987671, | |
| "eval_runtime": 30.4099, | |
| "eval_samples_per_second": 56.067, | |
| "eval_steps_per_second": 7.037, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.00017495769881556685, | |
| "loss": 2.0881, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 0.00017428087986463622, | |
| "loss": 1.9428, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 0.0001736040609137056, | |
| "loss": 1.8203, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 0.00017292724196277498, | |
| "loss": 1.6183, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "eval_accuracy": 0.4533724340175953, | |
| "eval_loss": 2.0731186866760254, | |
| "eval_runtime": 30.3675, | |
| "eval_samples_per_second": 56.146, | |
| "eval_steps_per_second": 7.047, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 0.00017225042301184434, | |
| "loss": 1.3395, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 0.0001715736040609137, | |
| "loss": 1.4052, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 0.00017089678510998307, | |
| "loss": 1.3804, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 0.00017021996615905247, | |
| "loss": 1.2771, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "eval_accuracy": 0.42052785923753666, | |
| "eval_loss": 2.0928714275360107, | |
| "eval_runtime": 30.2484, | |
| "eval_samples_per_second": 56.367, | |
| "eval_steps_per_second": 7.075, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 0.00016954314720812183, | |
| "loss": 1.431, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 0.00016886632825719122, | |
| "loss": 1.353, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 0.0001681895093062606, | |
| "loss": 1.307, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 0.00016751269035532995, | |
| "loss": 1.2116, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "eval_accuracy": 0.4504398826979472, | |
| "eval_loss": 2.0171947479248047, | |
| "eval_runtime": 30.2989, | |
| "eval_samples_per_second": 56.273, | |
| "eval_steps_per_second": 7.063, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 0.00016683587140439935, | |
| "loss": 1.2762, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 0.0001661590524534687, | |
| "loss": 1.2253, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 0.00016548223350253808, | |
| "loss": 1.238, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 0.00016480541455160744, | |
| "loss": 1.158, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "eval_accuracy": 0.45689149560117304, | |
| "eval_loss": 1.9670987129211426, | |
| "eval_runtime": 30.2767, | |
| "eval_samples_per_second": 56.314, | |
| "eval_steps_per_second": 7.068, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.00016412859560067684, | |
| "loss": 1.1488, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 0.0001634517766497462, | |
| "loss": 1.294, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 0.00016277495769881557, | |
| "loss": 1.0431, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.00016209813874788496, | |
| "loss": 1.1864, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_accuracy": 0.4563049853372434, | |
| "eval_loss": 1.9570399522781372, | |
| "eval_runtime": 30.2455, | |
| "eval_samples_per_second": 56.372, | |
| "eval_steps_per_second": 7.075, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 0.00016142131979695432, | |
| "loss": 1.1271, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 0.00016074450084602372, | |
| "loss": 1.1006, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 0.00016006768189509308, | |
| "loss": 1.1867, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 0.00015939086294416242, | |
| "loss": 0.7778, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "eval_accuracy": 0.49560117302052786, | |
| "eval_loss": 1.8706681728363037, | |
| "eval_runtime": 30.3606, | |
| "eval_samples_per_second": 56.158, | |
| "eval_steps_per_second": 7.049, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 0.0001587140439932318, | |
| "loss": 0.6935, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 0.00015803722504230118, | |
| "loss": 0.5692, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.00015736040609137057, | |
| "loss": 0.5985, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 0.00015668358714043994, | |
| "loss": 0.6021, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "eval_accuracy": 0.4950146627565982, | |
| "eval_loss": 1.8420542478561401, | |
| "eval_runtime": 30.3002, | |
| "eval_samples_per_second": 56.27, | |
| "eval_steps_per_second": 7.063, | |
| "step": 640 | |
| } | |
| ], | |
| "max_steps": 2955, | |
| "num_train_epochs": 15, | |
| "total_flos": 7.9058203969348e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |