| { | |
| "best_metric": 0.04292627424001694, | |
| "best_model_checkpoint": "./beans_outputs/checkpoint-650", | |
| "epoch": 5.0, | |
| "global_step": 650, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0003076923076923077, | |
| "loss": 1.0843, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0006153846153846154, | |
| "loss": 1.074, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0009230769230769232, | |
| "loss": 1.0582, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0012307692307692308, | |
| "loss": 0.9178, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0015384615384615387, | |
| "loss": 0.8364, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0018461538461538463, | |
| "loss": 0.5077, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0019996395276708854, | |
| "loss": 0.4777, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00199675730813421, | |
| "loss": 0.6303, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0019910011792459085, | |
| "loss": 0.558, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0019823877374156645, | |
| "loss": 0.5047, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.001970941817426052, | |
| "loss": 0.5314, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.0019566964208274254, | |
| "loss": 0.6402, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0019396926207859084, | |
| "loss": 0.7703, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5263157894736842, | |
| "eval_loss": 1.22383451461792, | |
| "eval_runtime": 0.797, | |
| "eval_samples_per_second": 166.882, | |
| "eval_steps_per_second": 21.331, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.0019199794436588243, | |
| "loss": 0.5088, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 0.0018976137276390142, | |
| "loss": 0.45, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.0018726599588756144, | |
| "loss": 0.5657, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.001845190085543795, | |
| "loss": 0.4771, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0018152833103995444, | |
| "loss": 0.5755, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.0017830258624176223, | |
| "loss": 0.414, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.001748510748171101, | |
| "loss": 0.217, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.0017118374836693406, | |
| "loss": 0.4516, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.0016731118074275702, | |
| "loss": 0.533, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 0.0016324453755953772, | |
| "loss": 0.4964, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.001589955440023123, | |
| "loss": 0.3408, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.0015457645101945045, | |
| "loss": 0.3039, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0015, | |
| "loss": 0.4905, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8270676691729323, | |
| "eval_loss": 0.5192657709121704, | |
| "eval_runtime": 0.7485, | |
| "eval_samples_per_second": 177.694, | |
| "eval_steps_per_second": 22.713, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 0.0014527938603696375, | |
| "loss": 0.6346, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.0014042821988243049, | |
| "loss": 0.4142, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 0.0013546048870425357, | |
| "loss": 0.3863, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 0.0013039051575742468, | |
| "loss": 0.2966, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 0.0012523291908642217, | |
| "loss": 0.49, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 0.0012000256937760446, | |
| "loss": 0.2886, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 0.001147145470831716, | |
| "loss": 0.4434, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.0010938409894031794, | |
| "loss": 0.2125, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.0010402659401094153, | |
| "loss": 0.2523, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 0.0009865747936866028, | |
| "loss": 0.2735, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 0.0009329223556089975, | |
| "loss": 0.27, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 0.0008794633197446771, | |
| "loss": 0.2445, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.0008263518223330697, | |
| "loss": 0.4793, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9699248120300752, | |
| "eval_loss": 0.14211735129356384, | |
| "eval_runtime": 0.7484, | |
| "eval_samples_per_second": 177.701, | |
| "eval_steps_per_second": 22.714, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 0.0007737409975702781, | |
| "loss": 0.2375, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 0.0007217825360835474, | |
| "loss": 0.3971, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 0.0006706262475679205, | |
| "loss": 0.3654, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 0.0006204196288461038, | |
| "loss": 0.2186, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 0.0005713074385969457, | |
| "loss": 0.2977, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 0.0005234312799786921, | |
| "loss": 0.1177, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 0.00047692919235042255, | |
| "loss": 0.4466, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 0.0004319352532688443, | |
| "loss": 0.2118, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 0.0003885791919079877, | |
| "loss": 0.2298, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 0.0003469860150164151, | |
| "loss": 0.3758, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 0.00030727564649040063, | |
| "loss": 0.2384, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 0.00026956258160229697, | |
| "loss": 0.1422, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0002339555568810221, | |
| "loss": 0.2986, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9624060150375939, | |
| "eval_loss": 0.07603178173303604, | |
| "eval_runtime": 0.7494, | |
| "eval_samples_per_second": 177.482, | |
| "eval_steps_per_second": 22.686, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 0.00020055723659649904, | |
| "loss": 0.1728, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 0.00016946391675198836, | |
| "loss": 0.1329, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 0.00014076524743778318, | |
| "loss": 0.1473, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 0.0001145439743467902, | |
| "loss": 0.1782, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 9.08757001972762e-05, | |
| "loss": 0.1615, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 6.98286667506618e-05, | |
| "loss": 0.2641, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "learning_rate": 5.146355805285452e-05, | |
| "loss": 0.167, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 3.5833325466437696e-05, | |
| "loss": 0.1689, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 2.2983034998182995e-05, | |
| "loss": 0.1596, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 1.2949737362087154e-05, | |
| "loss": 0.0924, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "learning_rate": 5.7623611525721155e-06, | |
| "loss": 0.1395, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 1.4416294358582382e-06, | |
| "loss": 0.1392, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.1927, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.04292627424001694, | |
| "eval_runtime": 0.7439, | |
| "eval_samples_per_second": 178.791, | |
| "eval_steps_per_second": 22.853, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 650, | |
| "total_flos": 1.0979660327860224e+17, | |
| "train_loss": 0.3968321686524611, | |
| "train_runtime": 71.285, | |
| "train_samples_per_second": 72.526, | |
| "train_steps_per_second": 9.118 | |
| } | |
| ], | |
| "max_steps": 650, | |
| "num_train_epochs": 5, | |
| "total_flos": 1.0979660327860224e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |