| { | |
| "best_metric": 0.0171243567019701, | |
| "best_model_checkpoint": "D:\\models/outputsstar/checkpoint-1350", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 1350, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.037037037037037035, | |
| "grad_norm": 1.6629393100738525, | |
| "learning_rate": 1.9851851851851855e-05, | |
| "loss": 0.5627, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07407407407407407, | |
| "grad_norm": 1.3468966484069824, | |
| "learning_rate": 1.9703703703703704e-05, | |
| "loss": 0.3602, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1111111111111111, | |
| "grad_norm": 1.9818885326385498, | |
| "learning_rate": 1.9555555555555557e-05, | |
| "loss": 0.2409, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.14814814814814814, | |
| "grad_norm": 0.6068008542060852, | |
| "learning_rate": 1.9407407407407407e-05, | |
| "loss": 0.1136, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.18518518518518517, | |
| "grad_norm": 0.5792471170425415, | |
| "learning_rate": 1.925925925925926e-05, | |
| "loss": 0.0912, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 0.36238569021224976, | |
| "learning_rate": 1.9111111111111113e-05, | |
| "loss": 0.0793, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.25925925925925924, | |
| "grad_norm": 0.40632012486457825, | |
| "learning_rate": 1.8962962962962966e-05, | |
| "loss": 0.0783, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2962962962962963, | |
| "grad_norm": 0.3358970582485199, | |
| "learning_rate": 1.8814814814814816e-05, | |
| "loss": 0.0976, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 0.22383123636245728, | |
| "learning_rate": 1.866666666666667e-05, | |
| "loss": 0.0453, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.37037037037037035, | |
| "grad_norm": 0.17972876131534576, | |
| "learning_rate": 1.851851851851852e-05, | |
| "loss": 0.0435, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4074074074074074, | |
| "grad_norm": 0.17635494470596313, | |
| "learning_rate": 1.837037037037037e-05, | |
| "loss": 0.0255, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 2.2137696743011475, | |
| "learning_rate": 1.8222222222222224e-05, | |
| "loss": 0.0251, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.48148148148148145, | |
| "grad_norm": 0.1810089647769928, | |
| "learning_rate": 1.8074074074074074e-05, | |
| "loss": 0.0588, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.5185185185185185, | |
| "grad_norm": 0.1574196070432663, | |
| "learning_rate": 1.7925925925925927e-05, | |
| "loss": 0.043, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 0.13450497388839722, | |
| "learning_rate": 1.7777777777777777e-05, | |
| "loss": 0.0647, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5925925925925926, | |
| "grad_norm": 0.14055782556533813, | |
| "learning_rate": 1.7629629629629633e-05, | |
| "loss": 0.025, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.6296296296296297, | |
| "grad_norm": 2.997166395187378, | |
| "learning_rate": 1.7481481481481483e-05, | |
| "loss": 0.0926, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 0.12140078842639923, | |
| "learning_rate": 1.7333333333333336e-05, | |
| "loss": 0.015, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.7037037037037037, | |
| "grad_norm": 0.11859618127346039, | |
| "learning_rate": 1.7185185185185185e-05, | |
| "loss": 0.058, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.7407407407407407, | |
| "grad_norm": 0.09671270847320557, | |
| "learning_rate": 1.7037037037037038e-05, | |
| "loss": 0.0127, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7777777777777778, | |
| "grad_norm": 0.100140281021595, | |
| "learning_rate": 1.688888888888889e-05, | |
| "loss": 0.0384, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.8148148148148148, | |
| "grad_norm": 0.10083166509866714, | |
| "learning_rate": 1.674074074074074e-05, | |
| "loss": 0.0121, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.8518518518518519, | |
| "grad_norm": 1.3739676475524902, | |
| "learning_rate": 1.6592592592592594e-05, | |
| "loss": 0.089, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 0.10838634520769119, | |
| "learning_rate": 1.6444444444444444e-05, | |
| "loss": 0.1098, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.9259259259259259, | |
| "grad_norm": 0.1810838133096695, | |
| "learning_rate": 1.6296296296296297e-05, | |
| "loss": 0.0287, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.9629629629629629, | |
| "grad_norm": 0.11047308892011642, | |
| "learning_rate": 1.614814814814815e-05, | |
| "loss": 0.0266, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.2907960414886475, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.0394, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.989501312335958, | |
| "eval_loss": 0.04065331816673279, | |
| "eval_runtime": 52.3322, | |
| "eval_samples_per_second": 7.28, | |
| "eval_steps_per_second": 0.917, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.037037037037037, | |
| "grad_norm": 0.08815015852451324, | |
| "learning_rate": 1.5851851851851852e-05, | |
| "loss": 0.0255, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.074074074074074, | |
| "grad_norm": 0.09346853941679001, | |
| "learning_rate": 1.5703703703703705e-05, | |
| "loss": 0.0107, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.1111111111111112, | |
| "grad_norm": 0.07555428147315979, | |
| "learning_rate": 1.555555555555556e-05, | |
| "loss": 0.0188, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.1481481481481481, | |
| "grad_norm": 0.08554250746965408, | |
| "learning_rate": 1.5407407407407408e-05, | |
| "loss": 0.056, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "grad_norm": 0.07133428752422333, | |
| "learning_rate": 1.525925925925926e-05, | |
| "loss": 0.0376, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.2222222222222223, | |
| "grad_norm": 0.07578225433826447, | |
| "learning_rate": 1.5111111111111112e-05, | |
| "loss": 0.0115, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.2592592592592593, | |
| "grad_norm": 0.07077538967132568, | |
| "learning_rate": 1.4962962962962964e-05, | |
| "loss": 0.0091, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.2962962962962963, | |
| "grad_norm": 0.06836975365877151, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.0089, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 23.20952606201172, | |
| "learning_rate": 1.4666666666666666e-05, | |
| "loss": 0.0995, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.3703703703703702, | |
| "grad_norm": 0.07452990114688873, | |
| "learning_rate": 1.4518518518518521e-05, | |
| "loss": 0.0089, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.4074074074074074, | |
| "grad_norm": 0.07127617299556732, | |
| "learning_rate": 1.4370370370370372e-05, | |
| "loss": 0.0104, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.4444444444444444, | |
| "grad_norm": 0.06184624508023262, | |
| "learning_rate": 1.4222222222222224e-05, | |
| "loss": 0.0081, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.4814814814814814, | |
| "grad_norm": 0.08141285926103592, | |
| "learning_rate": 1.4074074074074075e-05, | |
| "loss": 0.0104, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.5185185185185186, | |
| "grad_norm": 0.06248854473233223, | |
| "learning_rate": 1.3925925925925928e-05, | |
| "loss": 0.095, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.5555555555555556, | |
| "grad_norm": 0.07375632971525192, | |
| "learning_rate": 1.377777777777778e-05, | |
| "loss": 0.0918, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.5925925925925926, | |
| "grad_norm": 0.06345756351947784, | |
| "learning_rate": 1.362962962962963e-05, | |
| "loss": 0.0077, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.6296296296296298, | |
| "grad_norm": 0.06055911257863045, | |
| "learning_rate": 1.3481481481481482e-05, | |
| "loss": 0.0237, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 0.05871434509754181, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.0072, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.7037037037037037, | |
| "grad_norm": 0.0945742130279541, | |
| "learning_rate": 1.3185185185185185e-05, | |
| "loss": 0.0674, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.7407407407407407, | |
| "grad_norm": 0.0636238381266594, | |
| "learning_rate": 1.303703703703704e-05, | |
| "loss": 0.0623, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 0.08100485056638718, | |
| "learning_rate": 1.288888888888889e-05, | |
| "loss": 0.0103, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.8148148148148149, | |
| "grad_norm": 0.06018243730068207, | |
| "learning_rate": 1.2740740740740742e-05, | |
| "loss": 0.0072, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.8518518518518519, | |
| "grad_norm": 0.06805320084095001, | |
| "learning_rate": 1.2592592592592593e-05, | |
| "loss": 0.0602, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.8888888888888888, | |
| "grad_norm": 9.21893310546875, | |
| "learning_rate": 1.2444444444444446e-05, | |
| "loss": 0.0842, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.925925925925926, | |
| "grad_norm": 0.07045555114746094, | |
| "learning_rate": 1.2296296296296298e-05, | |
| "loss": 0.1148, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.9629629629629628, | |
| "grad_norm": 0.07674137502908707, | |
| "learning_rate": 1.2148148148148149e-05, | |
| "loss": 0.0094, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 62.091854095458984, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.0973, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.984251968503937, | |
| "eval_loss": 0.07090622931718826, | |
| "eval_runtime": 52.2703, | |
| "eval_samples_per_second": 7.289, | |
| "eval_steps_per_second": 0.918, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.037037037037037, | |
| "grad_norm": 0.05459068343043327, | |
| "learning_rate": 1.1851851851851852e-05, | |
| "loss": 0.0663, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.074074074074074, | |
| "grad_norm": 0.06096257269382477, | |
| "learning_rate": 1.1703703703703703e-05, | |
| "loss": 0.0113, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.111111111111111, | |
| "grad_norm": 0.05534420534968376, | |
| "learning_rate": 1.1555555555555556e-05, | |
| "loss": 0.0072, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.148148148148148, | |
| "grad_norm": 3.7594122886657715, | |
| "learning_rate": 1.1407407407407409e-05, | |
| "loss": 0.0592, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.185185185185185, | |
| "grad_norm": 0.05278675630688667, | |
| "learning_rate": 1.125925925925926e-05, | |
| "loss": 0.007, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "grad_norm": 0.050314392894506454, | |
| "learning_rate": 1.1111111111111113e-05, | |
| "loss": 0.0081, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.259259259259259, | |
| "grad_norm": 0.0523914210498333, | |
| "learning_rate": 1.0962962962962965e-05, | |
| "loss": 0.0104, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.2962962962962963, | |
| "grad_norm": 0.06894738972187042, | |
| "learning_rate": 1.0814814814814816e-05, | |
| "loss": 0.0071, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.3333333333333335, | |
| "grad_norm": 0.05035197734832764, | |
| "learning_rate": 1.0666666666666667e-05, | |
| "loss": 0.0062, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "grad_norm": 0.04814285784959793, | |
| "learning_rate": 1.0518518518518519e-05, | |
| "loss": 0.1157, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.4074074074074074, | |
| "grad_norm": 2.7678489685058594, | |
| "learning_rate": 1.037037037037037e-05, | |
| "loss": 0.0565, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.4444444444444446, | |
| "grad_norm": 3.185380220413208, | |
| "learning_rate": 1.0222222222222223e-05, | |
| "loss": 0.0508, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.4814814814814814, | |
| "grad_norm": 0.05677470192313194, | |
| "learning_rate": 1.0074074074074074e-05, | |
| "loss": 0.0074, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.5185185185185186, | |
| "grad_norm": 0.3886967599391937, | |
| "learning_rate": 9.925925925925927e-06, | |
| "loss": 0.0129, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.5555555555555554, | |
| "grad_norm": 0.0454292856156826, | |
| "learning_rate": 9.777777777777779e-06, | |
| "loss": 0.0958, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.5925925925925926, | |
| "grad_norm": 0.04909510165452957, | |
| "learning_rate": 9.62962962962963e-06, | |
| "loss": 0.0064, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.6296296296296298, | |
| "grad_norm": 0.04817191883921623, | |
| "learning_rate": 9.481481481481483e-06, | |
| "loss": 0.0192, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 3.5723814964294434, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 0.2254, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.7037037037037037, | |
| "grad_norm": 0.8698053359985352, | |
| "learning_rate": 9.185185185185186e-06, | |
| "loss": 0.0099, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.7407407407407405, | |
| "grad_norm": 3.2689051628112793, | |
| "learning_rate": 9.037037037037037e-06, | |
| "loss": 0.0534, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.7777777777777777, | |
| "grad_norm": 0.1893489956855774, | |
| "learning_rate": 8.888888888888888e-06, | |
| "loss": 0.0465, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.814814814814815, | |
| "grad_norm": 0.05454478785395622, | |
| "learning_rate": 8.740740740740741e-06, | |
| "loss": 0.0106, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.851851851851852, | |
| "grad_norm": 0.0487472228705883, | |
| "learning_rate": 8.592592592592593e-06, | |
| "loss": 0.0914, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.888888888888889, | |
| "grad_norm": 0.05404801294207573, | |
| "learning_rate": 8.444444444444446e-06, | |
| "loss": 0.0168, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.925925925925926, | |
| "grad_norm": 0.043936073780059814, | |
| "learning_rate": 8.296296296296297e-06, | |
| "loss": 0.0079, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.962962962962963, | |
| "grad_norm": 0.05625942721962929, | |
| "learning_rate": 8.148148148148148e-06, | |
| "loss": 0.0148, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.04617544263601303, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.0057, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9868766404199475, | |
| "eval_loss": 0.04252076894044876, | |
| "eval_runtime": 52.5906, | |
| "eval_samples_per_second": 7.245, | |
| "eval_steps_per_second": 0.913, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 3.037037037037037, | |
| "grad_norm": 0.5136130452156067, | |
| "learning_rate": 7.851851851851853e-06, | |
| "loss": 0.0257, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.074074074074074, | |
| "grad_norm": 2.9422848224639893, | |
| "learning_rate": 7.703703703703704e-06, | |
| "loss": 0.0544, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 3.111111111111111, | |
| "grad_norm": 0.0419733040034771, | |
| "learning_rate": 7.555555555555556e-06, | |
| "loss": 0.0073, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 3.148148148148148, | |
| "grad_norm": 0.052699312567710876, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.0068, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 3.185185185185185, | |
| "grad_norm": 0.04446011409163475, | |
| "learning_rate": 7.2592592592592605e-06, | |
| "loss": 0.0086, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 3.2222222222222223, | |
| "grad_norm": 0.04285755380988121, | |
| "learning_rate": 7.111111111111112e-06, | |
| "loss": 0.0063, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 3.259259259259259, | |
| "grad_norm": 0.04666002467274666, | |
| "learning_rate": 6.962962962962964e-06, | |
| "loss": 0.0586, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 3.2962962962962963, | |
| "grad_norm": 0.04253947734832764, | |
| "learning_rate": 6.814814814814815e-06, | |
| "loss": 0.0561, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.04394235461950302, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.0059, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.3703703703703702, | |
| "grad_norm": 0.04241425171494484, | |
| "learning_rate": 6.51851851851852e-06, | |
| "loss": 0.0056, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 3.4074074074074074, | |
| "grad_norm": 0.08582015335559845, | |
| "learning_rate": 6.370370370370371e-06, | |
| "loss": 0.0067, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 3.4444444444444446, | |
| "grad_norm": 0.03971382975578308, | |
| "learning_rate": 6.222222222222223e-06, | |
| "loss": 0.0393, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 3.4814814814814814, | |
| "grad_norm": 0.04343000799417496, | |
| "learning_rate": 6.0740740740740745e-06, | |
| "loss": 0.0056, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 3.5185185185185186, | |
| "grad_norm": 0.04006602242588997, | |
| "learning_rate": 5.925925925925926e-06, | |
| "loss": 0.0533, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "grad_norm": 0.040533244609832764, | |
| "learning_rate": 5.777777777777778e-06, | |
| "loss": 0.0277, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 3.5925925925925926, | |
| "grad_norm": 0.0381636768579483, | |
| "learning_rate": 5.62962962962963e-06, | |
| "loss": 0.022, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 3.6296296296296298, | |
| "grad_norm": 0.0465327687561512, | |
| "learning_rate": 5.481481481481482e-06, | |
| "loss": 0.0086, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 3.6666666666666665, | |
| "grad_norm": 0.039545219391584396, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 0.0053, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.7037037037037037, | |
| "grad_norm": 0.038673460483551025, | |
| "learning_rate": 5.185185185185185e-06, | |
| "loss": 0.0787, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.7407407407407405, | |
| "grad_norm": 0.04055389016866684, | |
| "learning_rate": 5.037037037037037e-06, | |
| "loss": 0.0062, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 3.7777777777777777, | |
| "grad_norm": 9.543475151062012, | |
| "learning_rate": 4.888888888888889e-06, | |
| "loss": 0.044, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 3.814814814814815, | |
| "grad_norm": 0.050516992807388306, | |
| "learning_rate": 4.7407407407407415e-06, | |
| "loss": 0.0531, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 3.851851851851852, | |
| "grad_norm": 0.040959686040878296, | |
| "learning_rate": 4.592592592592593e-06, | |
| "loss": 0.0054, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 3.888888888888889, | |
| "grad_norm": 0.2083148956298828, | |
| "learning_rate": 4.444444444444444e-06, | |
| "loss": 0.0055, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.925925925925926, | |
| "grad_norm": 0.04079532250761986, | |
| "learning_rate": 4.296296296296296e-06, | |
| "loss": 0.0758, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 3.962962962962963, | |
| "grad_norm": 3.1428794860839844, | |
| "learning_rate": 4.1481481481481485e-06, | |
| "loss": 0.1065, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.04060327261686325, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.0403, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9868766404199475, | |
| "eval_loss": 0.04992913454771042, | |
| "eval_runtime": 52.4443, | |
| "eval_samples_per_second": 7.265, | |
| "eval_steps_per_second": 0.915, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 4.037037037037037, | |
| "grad_norm": 0.037297219038009644, | |
| "learning_rate": 3.851851851851852e-06, | |
| "loss": 0.005, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 4.074074074074074, | |
| "grad_norm": 0.040778644382953644, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.005, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.111111111111111, | |
| "grad_norm": 0.03788766264915466, | |
| "learning_rate": 3.555555555555556e-06, | |
| "loss": 0.005, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 4.148148148148148, | |
| "grad_norm": 0.04125545546412468, | |
| "learning_rate": 3.4074074074074077e-06, | |
| "loss": 0.0047, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 4.185185185185185, | |
| "grad_norm": 0.03932119160890579, | |
| "learning_rate": 3.25925925925926e-06, | |
| "loss": 0.0209, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 4.222222222222222, | |
| "grad_norm": 0.040045417845249176, | |
| "learning_rate": 3.1111111111111116e-06, | |
| "loss": 0.005, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 4.2592592592592595, | |
| "grad_norm": 0.04132600873708725, | |
| "learning_rate": 2.962962962962963e-06, | |
| "loss": 0.0567, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 4.296296296296296, | |
| "grad_norm": 0.16079649329185486, | |
| "learning_rate": 2.814814814814815e-06, | |
| "loss": 0.0058, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 4.333333333333333, | |
| "grad_norm": 0.03831535950303078, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 0.0473, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 4.37037037037037, | |
| "grad_norm": 0.03685208782553673, | |
| "learning_rate": 2.5185185185185186e-06, | |
| "loss": 0.0051, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 4.407407407407407, | |
| "grad_norm": 0.08284825831651688, | |
| "learning_rate": 2.3703703703703707e-06, | |
| "loss": 0.0053, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 0.037671931087970734, | |
| "learning_rate": 2.222222222222222e-06, | |
| "loss": 0.0048, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.481481481481482, | |
| "grad_norm": 0.048201784491539, | |
| "learning_rate": 2.0740740740740742e-06, | |
| "loss": 0.0046, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 4.518518518518518, | |
| "grad_norm": 0.036273613572120667, | |
| "learning_rate": 1.925925925925926e-06, | |
| "loss": 0.008, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 4.555555555555555, | |
| "grad_norm": 0.03639749437570572, | |
| "learning_rate": 1.777777777777778e-06, | |
| "loss": 0.0268, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 4.592592592592593, | |
| "grad_norm": 0.03778412565588951, | |
| "learning_rate": 1.62962962962963e-06, | |
| "loss": 0.0047, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 4.62962962962963, | |
| "grad_norm": 0.037292417138814926, | |
| "learning_rate": 1.4814814814814815e-06, | |
| "loss": 0.0522, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 4.666666666666667, | |
| "grad_norm": 0.03608781844377518, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 0.0156, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 4.703703703703704, | |
| "grad_norm": 0.04082287847995758, | |
| "learning_rate": 1.1851851851851854e-06, | |
| "loss": 0.0072, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "grad_norm": 0.03990180045366287, | |
| "learning_rate": 1.0370370370370371e-06, | |
| "loss": 0.027, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 4.777777777777778, | |
| "grad_norm": 0.2720341086387634, | |
| "learning_rate": 8.88888888888889e-07, | |
| "loss": 0.0384, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 4.814814814814815, | |
| "grad_norm": 0.21008038520812988, | |
| "learning_rate": 7.407407407407407e-07, | |
| "loss": 0.0495, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.851851851851852, | |
| "grad_norm": 0.038784921169281006, | |
| "learning_rate": 5.925925925925927e-07, | |
| "loss": 0.0605, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 4.888888888888889, | |
| "grad_norm": 0.03550861403346062, | |
| "learning_rate": 4.444444444444445e-07, | |
| "loss": 0.007, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 4.925925925925926, | |
| "grad_norm": 0.03672279790043831, | |
| "learning_rate": 2.9629629629629634e-07, | |
| "loss": 0.005, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 4.962962962962963, | |
| "grad_norm": 0.04202316328883171, | |
| "learning_rate": 1.4814814814814817e-07, | |
| "loss": 0.0475, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.0387168787419796, | |
| "learning_rate": 0.0, | |
| "loss": 0.0608, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.994750656167979, | |
| "eval_loss": 0.0171243567019701, | |
| "eval_runtime": 52.6363, | |
| "eval_samples_per_second": 7.238, | |
| "eval_steps_per_second": 0.912, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 1350, | |
| "total_flos": 8.345887281491558e+17, | |
| "train_loss": 0.04418781167379132, | |
| "train_runtime": 3285.5702, | |
| "train_samples_per_second": 3.278, | |
| "train_steps_per_second": 0.411 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1350, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.345887281491558e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |