| { | |
| "best_metric": 0.8672727272727273, | |
| "best_model_checkpoint": "beit-base-patch16-224-RD\\checkpoint-764", | |
| "epoch": 39.75155279503105, | |
| "eval_steps": 500, | |
| "global_step": 1600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 3.125e-06, | |
| "loss": 1.6079, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 6.25e-06, | |
| "loss": 1.594, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 9.375000000000001e-06, | |
| "loss": 1.5602, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 1.25e-05, | |
| "loss": 1.4986, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.49454545454545457, | |
| "eval_loss": 1.4512125253677368, | |
| "eval_runtime": 3.6819, | |
| "eval_samples_per_second": 149.381, | |
| "eval_steps_per_second": 9.506, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.5625e-05, | |
| "loss": 1.4191, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 1.3138, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 2.1875e-05, | |
| "loss": 1.1635, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.0553, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_accuracy": 0.7472727272727273, | |
| "eval_loss": 0.9355355501174927, | |
| "eval_runtime": 3.5838, | |
| "eval_samples_per_second": 153.468, | |
| "eval_steps_per_second": 9.766, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 2.8125000000000003e-05, | |
| "loss": 0.9336, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 3.125e-05, | |
| "loss": 0.8113, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 3.4375e-05, | |
| "loss": 0.771, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.7972, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "eval_accuracy": 0.7436363636363637, | |
| "eval_loss": 0.7250072956085205, | |
| "eval_runtime": 3.805, | |
| "eval_samples_per_second": 144.547, | |
| "eval_steps_per_second": 9.198, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 4.0625000000000005e-05, | |
| "loss": 0.7561, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 4.375e-05, | |
| "loss": 0.7764, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 4.6875e-05, | |
| "loss": 0.7277, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7156, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7581818181818182, | |
| "eval_loss": 0.5845356583595276, | |
| "eval_runtime": 3.7344, | |
| "eval_samples_per_second": 147.279, | |
| "eval_steps_per_second": 9.372, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 4.965277777777778e-05, | |
| "loss": 0.7195, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 4.930555555555556e-05, | |
| "loss": 0.6915, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 4.8958333333333335e-05, | |
| "loss": 0.6135, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 0.6723, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.8036363636363636, | |
| "eval_loss": 0.5509195327758789, | |
| "eval_runtime": 3.7029, | |
| "eval_samples_per_second": 148.533, | |
| "eval_steps_per_second": 9.452, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 4.8263888888888895e-05, | |
| "loss": 0.6887, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.47, | |
| "learning_rate": 4.791666666666667e-05, | |
| "loss": 0.684, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "learning_rate": 4.756944444444444e-05, | |
| "loss": 0.6179, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.5942, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "eval_accuracy": 0.8218181818181818, | |
| "eval_loss": 0.5018081068992615, | |
| "eval_runtime": 3.6674, | |
| "eval_samples_per_second": 149.971, | |
| "eval_steps_per_second": 9.544, | |
| "step": 241 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 4.6875e-05, | |
| "loss": 0.5778, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.46, | |
| "learning_rate": 4.652777777777778e-05, | |
| "loss": 0.5984, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.71, | |
| "learning_rate": 4.618055555555556e-05, | |
| "loss": 0.6025, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 0.6223, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "eval_accuracy": 0.8218181818181818, | |
| "eval_loss": 0.49931806325912476, | |
| "eval_runtime": 3.5744, | |
| "eval_samples_per_second": 153.874, | |
| "eval_steps_per_second": 9.792, | |
| "step": 281 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 4.5486111111111114e-05, | |
| "loss": 0.5995, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.45, | |
| "learning_rate": 4.5138888888888894e-05, | |
| "loss": 0.5855, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.7, | |
| "learning_rate": 4.4791666666666673e-05, | |
| "loss": 0.5888, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 7.95, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.5731, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8290909090909091, | |
| "eval_loss": 0.4590268135070801, | |
| "eval_runtime": 3.5704, | |
| "eval_samples_per_second": 154.046, | |
| "eval_steps_per_second": 9.803, | |
| "step": 322 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "learning_rate": 4.4097222222222226e-05, | |
| "loss": 0.5685, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.45, | |
| "learning_rate": 4.375e-05, | |
| "loss": 0.614, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 4.340277777777778e-05, | |
| "loss": 0.5958, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "learning_rate": 4.305555555555556e-05, | |
| "loss": 0.5583, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.4877796471118927, | |
| "eval_runtime": 3.6449, | |
| "eval_samples_per_second": 150.896, | |
| "eval_steps_per_second": 9.603, | |
| "step": 362 | |
| }, | |
| { | |
| "epoch": 9.19, | |
| "learning_rate": 4.270833333333333e-05, | |
| "loss": 0.5185, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "learning_rate": 4.236111111111111e-05, | |
| "loss": 0.5777, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.69, | |
| "learning_rate": 4.201388888888889e-05, | |
| "loss": 0.572, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.5784, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "eval_accuracy": 0.8454545454545455, | |
| "eval_loss": 0.4484546482563019, | |
| "eval_runtime": 3.6484, | |
| "eval_samples_per_second": 150.753, | |
| "eval_steps_per_second": 9.593, | |
| "step": 402 | |
| }, | |
| { | |
| "epoch": 10.19, | |
| "learning_rate": 4.1319444444444445e-05, | |
| "loss": 0.5207, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 10.43, | |
| "learning_rate": 4.0972222222222225e-05, | |
| "loss": 0.4891, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 10.68, | |
| "learning_rate": 4.0625000000000005e-05, | |
| "loss": 0.568, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 10.93, | |
| "learning_rate": 4.027777777777778e-05, | |
| "loss": 0.4968, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 10.98, | |
| "eval_accuracy": 0.8345454545454546, | |
| "eval_loss": 0.4305056631565094, | |
| "eval_runtime": 3.6569, | |
| "eval_samples_per_second": 150.401, | |
| "eval_steps_per_second": 9.571, | |
| "step": 442 | |
| }, | |
| { | |
| "epoch": 11.18, | |
| "learning_rate": 3.993055555555556e-05, | |
| "loss": 0.4998, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 11.43, | |
| "learning_rate": 3.958333333333333e-05, | |
| "loss": 0.4738, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 11.68, | |
| "learning_rate": 3.923611111111111e-05, | |
| "loss": 0.5471, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.5324, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8345454545454546, | |
| "eval_loss": 0.47365644574165344, | |
| "eval_runtime": 3.8879, | |
| "eval_samples_per_second": 141.464, | |
| "eval_steps_per_second": 9.002, | |
| "step": 483 | |
| }, | |
| { | |
| "epoch": 12.17, | |
| "learning_rate": 3.854166666666667e-05, | |
| "loss": 0.5133, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 12.42, | |
| "learning_rate": 3.8194444444444444e-05, | |
| "loss": 0.486, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 12.67, | |
| "learning_rate": 3.7847222222222224e-05, | |
| "loss": 0.4803, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 12.92, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.4629, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "eval_accuracy": 0.8436363636363636, | |
| "eval_loss": 0.4253169000148773, | |
| "eval_runtime": 3.7694, | |
| "eval_samples_per_second": 145.912, | |
| "eval_steps_per_second": 9.285, | |
| "step": 523 | |
| }, | |
| { | |
| "epoch": 13.17, | |
| "learning_rate": 3.715277777777778e-05, | |
| "loss": 0.5659, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 13.42, | |
| "learning_rate": 3.6805555555555556e-05, | |
| "loss": 0.4986, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "learning_rate": 3.6458333333333336e-05, | |
| "loss": 0.4903, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 13.91, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.4398, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "eval_accuracy": 0.8472727272727273, | |
| "eval_loss": 0.41836607456207275, | |
| "eval_runtime": 3.9425, | |
| "eval_samples_per_second": 139.507, | |
| "eval_steps_per_second": 8.878, | |
| "step": 563 | |
| }, | |
| { | |
| "epoch": 14.16, | |
| "learning_rate": 3.576388888888889e-05, | |
| "loss": 0.4112, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 14.41, | |
| "learning_rate": 3.541666666666667e-05, | |
| "loss": 0.4664, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 14.66, | |
| "learning_rate": 3.506944444444444e-05, | |
| "loss": 0.4816, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 14.91, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.4575, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 14.98, | |
| "eval_accuracy": 0.8563636363636363, | |
| "eval_loss": 0.39294329285621643, | |
| "eval_runtime": 3.6178, | |
| "eval_samples_per_second": 152.024, | |
| "eval_steps_per_second": 9.674, | |
| "step": 603 | |
| }, | |
| { | |
| "epoch": 15.16, | |
| "learning_rate": 3.4375e-05, | |
| "loss": 0.4644, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 15.4, | |
| "learning_rate": 3.402777777777778e-05, | |
| "loss": 0.4587, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 15.65, | |
| "learning_rate": 3.368055555555556e-05, | |
| "loss": 0.4597, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 15.9, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.4554, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8490909090909091, | |
| "eval_loss": 0.4281716048717499, | |
| "eval_runtime": 3.5593, | |
| "eval_samples_per_second": 154.523, | |
| "eval_steps_per_second": 9.833, | |
| "step": 644 | |
| }, | |
| { | |
| "epoch": 16.15, | |
| "learning_rate": 3.2986111111111115e-05, | |
| "loss": 0.4964, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 16.4, | |
| "learning_rate": 3.263888888888889e-05, | |
| "loss": 0.3862, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 16.65, | |
| "learning_rate": 3.229166666666667e-05, | |
| "loss": 0.437, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 16.89, | |
| "learning_rate": 3.194444444444444e-05, | |
| "loss": 0.4646, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "eval_accuracy": 0.8236363636363636, | |
| "eval_loss": 0.4363492727279663, | |
| "eval_runtime": 3.5633, | |
| "eval_samples_per_second": 154.349, | |
| "eval_steps_per_second": 9.822, | |
| "step": 684 | |
| }, | |
| { | |
| "epoch": 17.14, | |
| "learning_rate": 3.159722222222222e-05, | |
| "loss": 0.5133, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 17.39, | |
| "learning_rate": 3.125e-05, | |
| "loss": 0.4389, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 17.64, | |
| "learning_rate": 3.090277777777778e-05, | |
| "loss": 0.4114, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 17.89, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.4535, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 17.99, | |
| "eval_accuracy": 0.8454545454545455, | |
| "eval_loss": 0.43369540572166443, | |
| "eval_runtime": 3.6178, | |
| "eval_samples_per_second": 152.024, | |
| "eval_steps_per_second": 9.674, | |
| "step": 724 | |
| }, | |
| { | |
| "epoch": 18.14, | |
| "learning_rate": 3.0208333333333334e-05, | |
| "loss": 0.4653, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 18.39, | |
| "learning_rate": 2.9861111111111113e-05, | |
| "loss": 0.4407, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 18.63, | |
| "learning_rate": 2.951388888888889e-05, | |
| "loss": 0.4223, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 18.88, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 0.3823, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "eval_accuracy": 0.8672727272727273, | |
| "eval_loss": 0.37708351016044617, | |
| "eval_runtime": 3.6744, | |
| "eval_samples_per_second": 149.685, | |
| "eval_steps_per_second": 9.525, | |
| "step": 764 | |
| }, | |
| { | |
| "epoch": 19.13, | |
| "learning_rate": 2.8819444444444443e-05, | |
| "loss": 0.3769, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 19.38, | |
| "learning_rate": 2.8472222222222223e-05, | |
| "loss": 0.3677, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 19.63, | |
| "learning_rate": 2.8125000000000003e-05, | |
| "loss": 0.4063, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 19.88, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.4584, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8563636363636363, | |
| "eval_loss": 0.39663317799568176, | |
| "eval_runtime": 3.6579, | |
| "eval_samples_per_second": 150.36, | |
| "eval_steps_per_second": 9.568, | |
| "step": 805 | |
| }, | |
| { | |
| "epoch": 20.12, | |
| "learning_rate": 2.743055555555556e-05, | |
| "loss": 0.4067, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 20.37, | |
| "learning_rate": 2.7083333333333332e-05, | |
| "loss": 0.4594, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 20.62, | |
| "learning_rate": 2.6736111111111112e-05, | |
| "loss": 0.4006, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 20.87, | |
| "learning_rate": 2.6388888888888892e-05, | |
| "loss": 0.4103, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 20.99, | |
| "eval_accuracy": 0.8490909090909091, | |
| "eval_loss": 0.4001007080078125, | |
| "eval_runtime": 3.5848, | |
| "eval_samples_per_second": 153.424, | |
| "eval_steps_per_second": 9.763, | |
| "step": 845 | |
| }, | |
| { | |
| "epoch": 21.12, | |
| "learning_rate": 2.604166666666667e-05, | |
| "loss": 0.3902, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 21.37, | |
| "learning_rate": 2.5694444444444445e-05, | |
| "loss": 0.3294, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 21.61, | |
| "learning_rate": 2.534722222222222e-05, | |
| "loss": 0.3812, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 21.86, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.3659, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 21.99, | |
| "eval_accuracy": 0.8581818181818182, | |
| "eval_loss": 0.3947709798812866, | |
| "eval_runtime": 3.6099, | |
| "eval_samples_per_second": 152.36, | |
| "eval_steps_per_second": 9.696, | |
| "step": 885 | |
| }, | |
| { | |
| "epoch": 22.11, | |
| "learning_rate": 2.465277777777778e-05, | |
| "loss": 0.3471, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 22.36, | |
| "learning_rate": 2.4305555555555558e-05, | |
| "loss": 0.3868, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 22.61, | |
| "learning_rate": 2.3958333333333334e-05, | |
| "loss": 0.3871, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 22.86, | |
| "learning_rate": 2.361111111111111e-05, | |
| "loss": 0.3241, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 22.98, | |
| "eval_accuracy": 0.8581818181818182, | |
| "eval_loss": 0.40068891644477844, | |
| "eval_runtime": 3.5493, | |
| "eval_samples_per_second": 154.959, | |
| "eval_steps_per_second": 9.861, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 23.11, | |
| "learning_rate": 2.326388888888889e-05, | |
| "loss": 0.3288, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 23.35, | |
| "learning_rate": 2.2916666666666667e-05, | |
| "loss": 0.3785, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 23.6, | |
| "learning_rate": 2.2569444444444447e-05, | |
| "loss": 0.3412, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 23.85, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.3575, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8327272727272728, | |
| "eval_loss": 0.43282008171081543, | |
| "eval_runtime": 3.7309, | |
| "eval_samples_per_second": 147.417, | |
| "eval_steps_per_second": 9.381, | |
| "step": 966 | |
| }, | |
| { | |
| "epoch": 24.1, | |
| "learning_rate": 2.1875e-05, | |
| "loss": 0.3713, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 24.35, | |
| "learning_rate": 2.152777777777778e-05, | |
| "loss": 0.3191, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 24.6, | |
| "learning_rate": 2.1180555555555556e-05, | |
| "loss": 0.3523, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 24.84, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.3411, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 24.99, | |
| "eval_accuracy": 0.8563636363636363, | |
| "eval_loss": 0.39898931980133057, | |
| "eval_runtime": 3.6919, | |
| "eval_samples_per_second": 148.976, | |
| "eval_steps_per_second": 9.48, | |
| "step": 1006 | |
| }, | |
| { | |
| "epoch": 25.09, | |
| "learning_rate": 2.0486111111111113e-05, | |
| "loss": 0.3233, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 25.34, | |
| "learning_rate": 2.013888888888889e-05, | |
| "loss": 0.288, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 25.59, | |
| "learning_rate": 1.9791666666666665e-05, | |
| "loss": 0.3424, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 25.84, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.3829, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 25.99, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.40113887190818787, | |
| "eval_runtime": 3.6378, | |
| "eval_samples_per_second": 151.189, | |
| "eval_steps_per_second": 9.621, | |
| "step": 1046 | |
| }, | |
| { | |
| "epoch": 26.09, | |
| "learning_rate": 1.9097222222222222e-05, | |
| "loss": 0.3457, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 26.34, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 0.341, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 26.58, | |
| "learning_rate": 1.8402777777777778e-05, | |
| "loss": 0.3439, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 26.83, | |
| "learning_rate": 1.8055555555555555e-05, | |
| "loss": 0.2855, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 26.98, | |
| "eval_accuracy": 0.8654545454545455, | |
| "eval_loss": 0.3858577013015747, | |
| "eval_runtime": 3.6854, | |
| "eval_samples_per_second": 149.238, | |
| "eval_steps_per_second": 9.497, | |
| "step": 1086 | |
| }, | |
| { | |
| "epoch": 27.08, | |
| "learning_rate": 1.7708333333333335e-05, | |
| "loss": 0.3007, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 27.33, | |
| "learning_rate": 1.736111111111111e-05, | |
| "loss": 0.3321, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 27.58, | |
| "learning_rate": 1.701388888888889e-05, | |
| "loss": 0.2806, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 27.83, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.254, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8672727272727273, | |
| "eval_loss": 0.4196436107158661, | |
| "eval_runtime": 3.6909, | |
| "eval_samples_per_second": 149.017, | |
| "eval_steps_per_second": 9.483, | |
| "step": 1127 | |
| }, | |
| { | |
| "epoch": 28.07, | |
| "learning_rate": 1.6319444444444444e-05, | |
| "loss": 0.3221, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 28.32, | |
| "learning_rate": 1.597222222222222e-05, | |
| "loss": 0.2784, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 28.57, | |
| "learning_rate": 1.5625e-05, | |
| "loss": 0.2717, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 28.82, | |
| "learning_rate": 1.527777777777778e-05, | |
| "loss": 0.2937, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 28.99, | |
| "eval_accuracy": 0.8618181818181818, | |
| "eval_loss": 0.43404603004455566, | |
| "eval_runtime": 3.7904, | |
| "eval_samples_per_second": 145.103, | |
| "eval_steps_per_second": 9.234, | |
| "step": 1167 | |
| }, | |
| { | |
| "epoch": 29.07, | |
| "learning_rate": 1.4930555555555557e-05, | |
| "loss": 0.2744, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 29.32, | |
| "learning_rate": 1.4583333333333335e-05, | |
| "loss": 0.247, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 29.57, | |
| "learning_rate": 1.4236111111111111e-05, | |
| "loss": 0.2614, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 29.81, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.258, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 29.99, | |
| "eval_accuracy": 0.850909090909091, | |
| "eval_loss": 0.43866536021232605, | |
| "eval_runtime": 3.6539, | |
| "eval_samples_per_second": 150.525, | |
| "eval_steps_per_second": 9.579, | |
| "step": 1207 | |
| }, | |
| { | |
| "epoch": 30.06, | |
| "learning_rate": 1.3541666666666666e-05, | |
| "loss": 0.3287, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 30.31, | |
| "learning_rate": 1.3194444444444446e-05, | |
| "loss": 0.2767, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 30.56, | |
| "learning_rate": 1.2847222222222222e-05, | |
| "loss": 0.2629, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 30.81, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2735, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 30.98, | |
| "eval_accuracy": 0.8654545454545455, | |
| "eval_loss": 0.40970146656036377, | |
| "eval_runtime": 3.8014, | |
| "eval_samples_per_second": 144.683, | |
| "eval_steps_per_second": 9.207, | |
| "step": 1247 | |
| }, | |
| { | |
| "epoch": 31.06, | |
| "learning_rate": 1.2152777777777779e-05, | |
| "loss": 0.3213, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 31.3, | |
| "learning_rate": 1.1805555555555555e-05, | |
| "loss": 0.2501, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 31.55, | |
| "learning_rate": 1.1458333333333333e-05, | |
| "loss": 0.2155, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 31.8, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.2674, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8527272727272728, | |
| "eval_loss": 0.41827914118766785, | |
| "eval_runtime": 3.6814, | |
| "eval_samples_per_second": 149.401, | |
| "eval_steps_per_second": 9.507, | |
| "step": 1288 | |
| }, | |
| { | |
| "epoch": 32.05, | |
| "learning_rate": 1.076388888888889e-05, | |
| "loss": 0.2601, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 32.3, | |
| "learning_rate": 1.0416666666666668e-05, | |
| "loss": 0.2215, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 32.55, | |
| "learning_rate": 1.0069444444444445e-05, | |
| "loss": 0.2523, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 32.8, | |
| "learning_rate": 9.722222222222223e-06, | |
| "loss": 0.2547, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 32.99, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.421658456325531, | |
| "eval_runtime": 3.9024, | |
| "eval_samples_per_second": 140.938, | |
| "eval_steps_per_second": 8.969, | |
| "step": 1328 | |
| }, | |
| { | |
| "epoch": 33.04, | |
| "learning_rate": 9.375000000000001e-06, | |
| "loss": 0.2755, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 33.29, | |
| "learning_rate": 9.027777777777777e-06, | |
| "loss": 0.2406, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 33.54, | |
| "learning_rate": 8.680555555555556e-06, | |
| "loss": 0.235, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 33.79, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.2109, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 33.99, | |
| "eval_accuracy": 0.8527272727272728, | |
| "eval_loss": 0.42401665449142456, | |
| "eval_runtime": 3.8429, | |
| "eval_samples_per_second": 143.12, | |
| "eval_steps_per_second": 9.108, | |
| "step": 1368 | |
| }, | |
| { | |
| "epoch": 34.04, | |
| "learning_rate": 7.98611111111111e-06, | |
| "loss": 0.278, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 34.29, | |
| "learning_rate": 7.63888888888889e-06, | |
| "loss": 0.2164, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 34.53, | |
| "learning_rate": 7.2916666666666674e-06, | |
| "loss": 0.2403, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 34.78, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.2248, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 34.98, | |
| "eval_accuracy": 0.86, | |
| "eval_loss": 0.4249715805053711, | |
| "eval_runtime": 3.9044, | |
| "eval_samples_per_second": 140.865, | |
| "eval_steps_per_second": 8.964, | |
| "step": 1408 | |
| }, | |
| { | |
| "epoch": 35.03, | |
| "learning_rate": 6.597222222222223e-06, | |
| "loss": 0.249, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 35.28, | |
| "learning_rate": 6.25e-06, | |
| "loss": 0.2012, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 35.53, | |
| "learning_rate": 5.902777777777778e-06, | |
| "loss": 0.2184, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 35.78, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.2397, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.8581818181818182, | |
| "eval_loss": 0.44308120012283325, | |
| "eval_runtime": 3.7259, | |
| "eval_samples_per_second": 147.616, | |
| "eval_steps_per_second": 9.394, | |
| "step": 1449 | |
| }, | |
| { | |
| "epoch": 36.02, | |
| "learning_rate": 5.208333333333334e-06, | |
| "loss": 0.2341, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 36.27, | |
| "learning_rate": 4.861111111111111e-06, | |
| "loss": 0.2272, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 36.52, | |
| "learning_rate": 4.513888888888889e-06, | |
| "loss": 0.216, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 36.77, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.1823, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 36.99, | |
| "eval_accuracy": 0.8581818181818182, | |
| "eval_loss": 0.4442140758037567, | |
| "eval_runtime": 3.7874, | |
| "eval_samples_per_second": 145.218, | |
| "eval_steps_per_second": 9.241, | |
| "step": 1489 | |
| }, | |
| { | |
| "epoch": 37.02, | |
| "learning_rate": 3.819444444444445e-06, | |
| "loss": 0.1781, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 37.27, | |
| "learning_rate": 3.4722222222222224e-06, | |
| "loss": 0.201, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 37.52, | |
| "learning_rate": 3.125e-06, | |
| "loss": 0.1889, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 37.76, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.1834, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 37.99, | |
| "eval_accuracy": 0.8618181818181818, | |
| "eval_loss": 0.4361916780471802, | |
| "eval_runtime": 3.6724, | |
| "eval_samples_per_second": 149.767, | |
| "eval_steps_per_second": 9.531, | |
| "step": 1529 | |
| }, | |
| { | |
| "epoch": 38.01, | |
| "learning_rate": 2.4305555555555557e-06, | |
| "loss": 0.2874, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 38.26, | |
| "learning_rate": 2.0833333333333334e-06, | |
| "loss": 0.2249, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 38.51, | |
| "learning_rate": 1.7361111111111112e-06, | |
| "loss": 0.2189, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 38.76, | |
| "learning_rate": 1.388888888888889e-06, | |
| "loss": 0.1864, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 38.98, | |
| "eval_accuracy": 0.8545454545454545, | |
| "eval_loss": 0.43384620547294617, | |
| "eval_runtime": 3.7839, | |
| "eval_samples_per_second": 145.353, | |
| "eval_steps_per_second": 9.25, | |
| "step": 1569 | |
| }, | |
| { | |
| "epoch": 39.01, | |
| "learning_rate": 1.0416666666666667e-06, | |
| "loss": 0.1708, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 39.25, | |
| "learning_rate": 6.944444444444445e-07, | |
| "loss": 0.2142, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 39.5, | |
| "learning_rate": 3.4722222222222224e-07, | |
| "loss": 0.1828, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 39.75, | |
| "learning_rate": 0.0, | |
| "loss": 0.1779, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 39.75, | |
| "eval_accuracy": 0.8581818181818182, | |
| "eval_loss": 0.4332370162010193, | |
| "eval_runtime": 3.7169, | |
| "eval_samples_per_second": 147.975, | |
| "eval_steps_per_second": 9.417, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 39.75, | |
| "step": 1600, | |
| "total_flos": 7.892728807229393e+18, | |
| "train_loss": 0.45905578792095186, | |
| "train_runtime": 1870.7092, | |
| "train_samples_per_second": 54.803, | |
| "train_steps_per_second": 0.855 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1600, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 7.892728807229393e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |