| { |
| "best_metric": 0.9028571428571428, |
| "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-humeda-DAV65/checkpoint-495", |
| "epoch": 45.0, |
| "eval_steps": 500, |
| "global_step": 675, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.6666666666666666, |
| "grad_norm": 4.2027106285095215, |
| "learning_rate": 8.823529411764707e-06, |
| "loss": 1.1096, |
| "step": 10 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.48, |
| "eval_loss": 1.020611047744751, |
| "eval_runtime": 1.9144, |
| "eval_samples_per_second": 91.411, |
| "eval_steps_per_second": 5.746, |
| "step": 15 |
| }, |
| { |
| "epoch": 1.3333333333333333, |
| "grad_norm": 4.204517841339111, |
| "learning_rate": 1.7647058823529414e-05, |
| "loss": 0.9978, |
| "step": 20 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 9.762907981872559, |
| "learning_rate": 2.647058823529412e-05, |
| "loss": 0.7949, |
| "step": 30 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8057142857142857, |
| "eval_loss": 0.5636562705039978, |
| "eval_runtime": 1.9435, |
| "eval_samples_per_second": 90.044, |
| "eval_steps_per_second": 5.66, |
| "step": 30 |
| }, |
| { |
| "epoch": 2.6666666666666665, |
| "grad_norm": 15.502463340759277, |
| "learning_rate": 3.529411764705883e-05, |
| "loss": 0.6305, |
| "step": 40 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8457142857142858, |
| "eval_loss": 0.3961445391178131, |
| "eval_runtime": 1.9278, |
| "eval_samples_per_second": 90.778, |
| "eval_steps_per_second": 5.706, |
| "step": 45 |
| }, |
| { |
| "epoch": 3.3333333333333335, |
| "grad_norm": 5.7075605392456055, |
| "learning_rate": 4.411764705882353e-05, |
| "loss": 0.454, |
| "step": 50 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 7.159061908721924, |
| "learning_rate": 5.294117647058824e-05, |
| "loss": 0.4475, |
| "step": 60 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8628571428571429, |
| "eval_loss": 0.3463701605796814, |
| "eval_runtime": 2.3504, |
| "eval_samples_per_second": 74.457, |
| "eval_steps_per_second": 4.68, |
| "step": 60 |
| }, |
| { |
| "epoch": 4.666666666666667, |
| "grad_norm": 32.75265884399414, |
| "learning_rate": 5.9802306425041185e-05, |
| "loss": 0.4342, |
| "step": 70 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8628571428571429, |
| "eval_loss": 0.3489004373550415, |
| "eval_runtime": 1.9048, |
| "eval_samples_per_second": 91.871, |
| "eval_steps_per_second": 5.775, |
| "step": 75 |
| }, |
| { |
| "epoch": 5.333333333333333, |
| "grad_norm": 9.125953674316406, |
| "learning_rate": 5.881383855024712e-05, |
| "loss": 0.373, |
| "step": 80 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 5.667314529418945, |
| "learning_rate": 5.782537067545305e-05, |
| "loss": 0.3843, |
| "step": 90 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.8457142857142858, |
| "eval_loss": 0.33044305443763733, |
| "eval_runtime": 2.9105, |
| "eval_samples_per_second": 60.126, |
| "eval_steps_per_second": 3.779, |
| "step": 90 |
| }, |
| { |
| "epoch": 6.666666666666667, |
| "grad_norm": 8.047316551208496, |
| "learning_rate": 5.683690280065898e-05, |
| "loss": 0.3059, |
| "step": 100 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.88, |
| "eval_loss": 0.3037264049053192, |
| "eval_runtime": 1.843, |
| "eval_samples_per_second": 94.953, |
| "eval_steps_per_second": 5.968, |
| "step": 105 |
| }, |
| { |
| "epoch": 7.333333333333333, |
| "grad_norm": 11.130711555480957, |
| "learning_rate": 5.5848434925864906e-05, |
| "loss": 0.2592, |
| "step": 110 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 11.404470443725586, |
| "learning_rate": 5.4859967051070845e-05, |
| "loss": 0.2993, |
| "step": 120 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.8685714285714285, |
| "eval_loss": 0.32861611247062683, |
| "eval_runtime": 1.8871, |
| "eval_samples_per_second": 92.734, |
| "eval_steps_per_second": 5.829, |
| "step": 120 |
| }, |
| { |
| "epoch": 8.666666666666666, |
| "grad_norm": 12.865260124206543, |
| "learning_rate": 5.387149917627677e-05, |
| "loss": 0.2799, |
| "step": 130 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.8457142857142858, |
| "eval_loss": 0.4499312937259674, |
| "eval_runtime": 2.3042, |
| "eval_samples_per_second": 75.949, |
| "eval_steps_per_second": 4.774, |
| "step": 135 |
| }, |
| { |
| "epoch": 9.333333333333334, |
| "grad_norm": 14.084300994873047, |
| "learning_rate": 5.28830313014827e-05, |
| "loss": 0.3012, |
| "step": 140 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 16.453907012939453, |
| "learning_rate": 5.1993410214168035e-05, |
| "loss": 0.302, |
| "step": 150 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_accuracy": 0.8857142857142857, |
| "eval_loss": 0.29645153880119324, |
| "eval_runtime": 1.925, |
| "eval_samples_per_second": 90.911, |
| "eval_steps_per_second": 5.714, |
| "step": 150 |
| }, |
| { |
| "epoch": 10.666666666666666, |
| "grad_norm": 6.858348369598389, |
| "learning_rate": 5.1004942339373974e-05, |
| "loss": 0.2445, |
| "step": 160 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_accuracy": 0.8857142857142857, |
| "eval_loss": 0.2568677067756653, |
| "eval_runtime": 1.96, |
| "eval_samples_per_second": 89.286, |
| "eval_steps_per_second": 5.612, |
| "step": 165 |
| }, |
| { |
| "epoch": 11.333333333333334, |
| "grad_norm": 4.542701721191406, |
| "learning_rate": 5.00164744645799e-05, |
| "loss": 0.2314, |
| "step": 170 |
| }, |
| { |
| "epoch": 12.0, |
| "grad_norm": 10.66151237487793, |
| "learning_rate": 4.902800658978583e-05, |
| "loss": 0.2317, |
| "step": 180 |
| }, |
| { |
| "epoch": 12.0, |
| "eval_accuracy": 0.8285714285714286, |
| "eval_loss": 0.4269125461578369, |
| "eval_runtime": 1.9306, |
| "eval_samples_per_second": 90.645, |
| "eval_steps_per_second": 5.698, |
| "step": 180 |
| }, |
| { |
| "epoch": 12.666666666666666, |
| "grad_norm": 11.812453269958496, |
| "learning_rate": 4.803953871499176e-05, |
| "loss": 0.2189, |
| "step": 190 |
| }, |
| { |
| "epoch": 13.0, |
| "eval_accuracy": 0.8514285714285714, |
| "eval_loss": 0.4250338673591614, |
| "eval_runtime": 2.6493, |
| "eval_samples_per_second": 66.056, |
| "eval_steps_per_second": 4.152, |
| "step": 195 |
| }, |
| { |
| "epoch": 13.333333333333334, |
| "grad_norm": 9.137828826904297, |
| "learning_rate": 4.7051070840197695e-05, |
| "loss": 0.2342, |
| "step": 200 |
| }, |
| { |
| "epoch": 14.0, |
| "grad_norm": 4.940479755401611, |
| "learning_rate": 4.606260296540363e-05, |
| "loss": 0.2001, |
| "step": 210 |
| }, |
| { |
| "epoch": 14.0, |
| "eval_accuracy": 0.8742857142857143, |
| "eval_loss": 0.37246543169021606, |
| "eval_runtime": 1.8761, |
| "eval_samples_per_second": 93.28, |
| "eval_steps_per_second": 5.863, |
| "step": 210 |
| }, |
| { |
| "epoch": 14.666666666666666, |
| "grad_norm": 11.54446029663086, |
| "learning_rate": 4.507413509060956e-05, |
| "loss": 0.1924, |
| "step": 220 |
| }, |
| { |
| "epoch": 15.0, |
| "eval_accuracy": 0.8571428571428571, |
| "eval_loss": 0.30415335297584534, |
| "eval_runtime": 1.8876, |
| "eval_samples_per_second": 92.709, |
| "eval_steps_per_second": 5.827, |
| "step": 225 |
| }, |
| { |
| "epoch": 15.333333333333334, |
| "grad_norm": 6.666581630706787, |
| "learning_rate": 4.4085667215815484e-05, |
| "loss": 0.1656, |
| "step": 230 |
| }, |
| { |
| "epoch": 16.0, |
| "grad_norm": 14.857701301574707, |
| "learning_rate": 4.309719934102142e-05, |
| "loss": 0.1617, |
| "step": 240 |
| }, |
| { |
| "epoch": 16.0, |
| "eval_accuracy": 0.8685714285714285, |
| "eval_loss": 0.4525071084499359, |
| "eval_runtime": 2.396, |
| "eval_samples_per_second": 73.037, |
| "eval_steps_per_second": 4.591, |
| "step": 240 |
| }, |
| { |
| "epoch": 16.666666666666668, |
| "grad_norm": 6.976684093475342, |
| "learning_rate": 4.210873146622735e-05, |
| "loss": 0.131, |
| "step": 250 |
| }, |
| { |
| "epoch": 17.0, |
| "eval_accuracy": 0.8457142857142858, |
| "eval_loss": 0.4021240174770355, |
| "eval_runtime": 1.9025, |
| "eval_samples_per_second": 91.982, |
| "eval_steps_per_second": 5.782, |
| "step": 255 |
| }, |
| { |
| "epoch": 17.333333333333332, |
| "grad_norm": 10.898083686828613, |
| "learning_rate": 4.112026359143328e-05, |
| "loss": 0.1347, |
| "step": 260 |
| }, |
| { |
| "epoch": 18.0, |
| "grad_norm": 5.943521976470947, |
| "learning_rate": 4.013179571663921e-05, |
| "loss": 0.1491, |
| "step": 270 |
| }, |
| { |
| "epoch": 18.0, |
| "eval_accuracy": 0.8571428571428571, |
| "eval_loss": 0.3315869867801666, |
| "eval_runtime": 1.9124, |
| "eval_samples_per_second": 91.51, |
| "eval_steps_per_second": 5.752, |
| "step": 270 |
| }, |
| { |
| "epoch": 18.666666666666668, |
| "grad_norm": 5.572836399078369, |
| "learning_rate": 3.914332784184514e-05, |
| "loss": 0.1576, |
| "step": 280 |
| }, |
| { |
| "epoch": 19.0, |
| "eval_accuracy": 0.8857142857142857, |
| "eval_loss": 0.3287747800350189, |
| "eval_runtime": 2.6482, |
| "eval_samples_per_second": 66.082, |
| "eval_steps_per_second": 4.154, |
| "step": 285 |
| }, |
| { |
| "epoch": 19.333333333333332, |
| "grad_norm": 16.414600372314453, |
| "learning_rate": 3.815485996705107e-05, |
| "loss": 0.1778, |
| "step": 290 |
| }, |
| { |
| "epoch": 20.0, |
| "grad_norm": 5.983540058135986, |
| "learning_rate": 3.7166392092257e-05, |
| "loss": 0.1189, |
| "step": 300 |
| }, |
| { |
| "epoch": 20.0, |
| "eval_accuracy": 0.88, |
| "eval_loss": 0.34641945362091064, |
| "eval_runtime": 2.4117, |
| "eval_samples_per_second": 72.562, |
| "eval_steps_per_second": 4.561, |
| "step": 300 |
| }, |
| { |
| "epoch": 20.666666666666668, |
| "grad_norm": 9.16804027557373, |
| "learning_rate": 3.617792421746293e-05, |
| "loss": 0.1384, |
| "step": 310 |
| }, |
| { |
| "epoch": 21.0, |
| "eval_accuracy": 0.8857142857142857, |
| "eval_loss": 0.36178210377693176, |
| "eval_runtime": 1.8991, |
| "eval_samples_per_second": 92.15, |
| "eval_steps_per_second": 5.792, |
| "step": 315 |
| }, |
| { |
| "epoch": 21.333333333333332, |
| "grad_norm": 7.604579448699951, |
| "learning_rate": 3.5189456342668865e-05, |
| "loss": 0.1085, |
| "step": 320 |
| }, |
| { |
| "epoch": 22.0, |
| "grad_norm": 1.40953528881073, |
| "learning_rate": 3.42009884678748e-05, |
| "loss": 0.1109, |
| "step": 330 |
| }, |
| { |
| "epoch": 22.0, |
| "eval_accuracy": 0.8914285714285715, |
| "eval_loss": 0.35477569699287415, |
| "eval_runtime": 1.9293, |
| "eval_samples_per_second": 90.706, |
| "eval_steps_per_second": 5.702, |
| "step": 330 |
| }, |
| { |
| "epoch": 22.666666666666668, |
| "grad_norm": 0.6196412444114685, |
| "learning_rate": 3.321252059308072e-05, |
| "loss": 0.0985, |
| "step": 340 |
| }, |
| { |
| "epoch": 23.0, |
| "eval_accuracy": 0.8857142857142857, |
| "eval_loss": 0.3907780647277832, |
| "eval_runtime": 1.9111, |
| "eval_samples_per_second": 91.569, |
| "eval_steps_per_second": 5.756, |
| "step": 345 |
| }, |
| { |
| "epoch": 23.333333333333332, |
| "grad_norm": 5.967416286468506, |
| "learning_rate": 3.222405271828666e-05, |
| "loss": 0.083, |
| "step": 350 |
| }, |
| { |
| "epoch": 24.0, |
| "grad_norm": 4.097744941711426, |
| "learning_rate": 3.1235584843492586e-05, |
| "loss": 0.1095, |
| "step": 360 |
| }, |
| { |
| "epoch": 24.0, |
| "eval_accuracy": 0.8742857142857143, |
| "eval_loss": 0.4517044425010681, |
| "eval_runtime": 2.0346, |
| "eval_samples_per_second": 86.01, |
| "eval_steps_per_second": 5.406, |
| "step": 360 |
| }, |
| { |
| "epoch": 24.666666666666668, |
| "grad_norm": 6.900848388671875, |
| "learning_rate": 3.024711696869852e-05, |
| "loss": 0.1007, |
| "step": 370 |
| }, |
| { |
| "epoch": 25.0, |
| "eval_accuracy": 0.8742857142857143, |
| "eval_loss": 0.5405580997467041, |
| "eval_runtime": 1.8799, |
| "eval_samples_per_second": 93.091, |
| "eval_steps_per_second": 5.851, |
| "step": 375 |
| }, |
| { |
| "epoch": 25.333333333333332, |
| "grad_norm": 10.140560150146484, |
| "learning_rate": 2.925864909390445e-05, |
| "loss": 0.0913, |
| "step": 380 |
| }, |
| { |
| "epoch": 26.0, |
| "grad_norm": 18.192825317382812, |
| "learning_rate": 2.8270181219110378e-05, |
| "loss": 0.192, |
| "step": 390 |
| }, |
| { |
| "epoch": 26.0, |
| "eval_accuracy": 0.8685714285714285, |
| "eval_loss": 0.5257406830787659, |
| "eval_runtime": 2.1028, |
| "eval_samples_per_second": 83.223, |
| "eval_steps_per_second": 5.231, |
| "step": 390 |
| }, |
| { |
| "epoch": 26.666666666666668, |
| "grad_norm": 5.915510177612305, |
| "learning_rate": 2.728171334431631e-05, |
| "loss": 0.093, |
| "step": 400 |
| }, |
| { |
| "epoch": 27.0, |
| "eval_accuracy": 0.8685714285714285, |
| "eval_loss": 0.4441986083984375, |
| "eval_runtime": 1.8853, |
| "eval_samples_per_second": 92.826, |
| "eval_steps_per_second": 5.835, |
| "step": 405 |
| }, |
| { |
| "epoch": 27.333333333333332, |
| "grad_norm": 4.001173973083496, |
| "learning_rate": 2.6293245469522242e-05, |
| "loss": 0.0857, |
| "step": 410 |
| }, |
| { |
| "epoch": 28.0, |
| "grad_norm": 9.286985397338867, |
| "learning_rate": 2.540362438220758e-05, |
| "loss": 0.1337, |
| "step": 420 |
| }, |
| { |
| "epoch": 28.0, |
| "eval_accuracy": 0.8628571428571429, |
| "eval_loss": 0.5375655293464661, |
| "eval_runtime": 1.9012, |
| "eval_samples_per_second": 92.046, |
| "eval_steps_per_second": 5.786, |
| "step": 420 |
| }, |
| { |
| "epoch": 28.666666666666668, |
| "grad_norm": 4.529911518096924, |
| "learning_rate": 2.4415156507413507e-05, |
| "loss": 0.0737, |
| "step": 430 |
| }, |
| { |
| "epoch": 29.0, |
| "eval_accuracy": 0.8685714285714285, |
| "eval_loss": 0.4627114534378052, |
| "eval_runtime": 1.9144, |
| "eval_samples_per_second": 91.412, |
| "eval_steps_per_second": 5.746, |
| "step": 435 |
| }, |
| { |
| "epoch": 29.333333333333332, |
| "grad_norm": 5.005265712738037, |
| "learning_rate": 2.342668863261944e-05, |
| "loss": 0.0776, |
| "step": 440 |
| }, |
| { |
| "epoch": 30.0, |
| "grad_norm": 2.6288208961486816, |
| "learning_rate": 2.243822075782537e-05, |
| "loss": 0.0932, |
| "step": 450 |
| }, |
| { |
| "epoch": 30.0, |
| "eval_accuracy": 0.8914285714285715, |
| "eval_loss": 0.4370952546596527, |
| "eval_runtime": 1.8915, |
| "eval_samples_per_second": 92.517, |
| "eval_steps_per_second": 5.815, |
| "step": 450 |
| }, |
| { |
| "epoch": 30.666666666666668, |
| "grad_norm": 3.95957088470459, |
| "learning_rate": 2.1449752883031303e-05, |
| "loss": 0.0638, |
| "step": 460 |
| }, |
| { |
| "epoch": 31.0, |
| "eval_accuracy": 0.8971428571428571, |
| "eval_loss": 0.47408464550971985, |
| "eval_runtime": 2.6861, |
| "eval_samples_per_second": 65.151, |
| "eval_steps_per_second": 4.095, |
| "step": 465 |
| }, |
| { |
| "epoch": 31.333333333333332, |
| "grad_norm": 19.800823211669922, |
| "learning_rate": 2.0461285008237232e-05, |
| "loss": 0.119, |
| "step": 470 |
| }, |
| { |
| "epoch": 32.0, |
| "grad_norm": 12.307100296020508, |
| "learning_rate": 1.9472817133443164e-05, |
| "loss": 0.0796, |
| "step": 480 |
| }, |
| { |
| "epoch": 32.0, |
| "eval_accuracy": 0.88, |
| "eval_loss": 0.4220117926597595, |
| "eval_runtime": 1.8732, |
| "eval_samples_per_second": 93.421, |
| "eval_steps_per_second": 5.872, |
| "step": 480 |
| }, |
| { |
| "epoch": 32.666666666666664, |
| "grad_norm": 7.8525004386901855, |
| "learning_rate": 1.8484349258649096e-05, |
| "loss": 0.0674, |
| "step": 490 |
| }, |
| { |
| "epoch": 33.0, |
| "eval_accuracy": 0.9028571428571428, |
| "eval_loss": 0.443218857049942, |
| "eval_runtime": 1.8867, |
| "eval_samples_per_second": 92.754, |
| "eval_steps_per_second": 5.83, |
| "step": 495 |
| }, |
| { |
| "epoch": 33.333333333333336, |
| "grad_norm": 7.019139766693115, |
| "learning_rate": 1.7495881383855024e-05, |
| "loss": 0.0566, |
| "step": 500 |
| }, |
| { |
| "epoch": 34.0, |
| "grad_norm": 10.563660621643066, |
| "learning_rate": 1.6507413509060956e-05, |
| "loss": 0.0466, |
| "step": 510 |
| }, |
| { |
| "epoch": 34.0, |
| "eval_accuracy": 0.8914285714285715, |
| "eval_loss": 0.438544362783432, |
| "eval_runtime": 1.9031, |
| "eval_samples_per_second": 91.956, |
| "eval_steps_per_second": 5.78, |
| "step": 510 |
| }, |
| { |
| "epoch": 34.666666666666664, |
| "grad_norm": 7.634331226348877, |
| "learning_rate": 1.5518945634266888e-05, |
| "loss": 0.0586, |
| "step": 520 |
| }, |
| { |
| "epoch": 35.0, |
| "eval_accuracy": 0.8971428571428571, |
| "eval_loss": 0.4613957107067108, |
| "eval_runtime": 1.8921, |
| "eval_samples_per_second": 92.489, |
| "eval_steps_per_second": 5.814, |
| "step": 525 |
| }, |
| { |
| "epoch": 35.333333333333336, |
| "grad_norm": 5.198871612548828, |
| "learning_rate": 1.4530477759472817e-05, |
| "loss": 0.0795, |
| "step": 530 |
| }, |
| { |
| "epoch": 36.0, |
| "grad_norm": 6.447375297546387, |
| "learning_rate": 1.3542009884678749e-05, |
| "loss": 0.0634, |
| "step": 540 |
| }, |
| { |
| "epoch": 36.0, |
| "eval_accuracy": 0.8857142857142857, |
| "eval_loss": 0.4855011999607086, |
| "eval_runtime": 1.898, |
| "eval_samples_per_second": 92.2, |
| "eval_steps_per_second": 5.795, |
| "step": 540 |
| }, |
| { |
| "epoch": 36.666666666666664, |
| "grad_norm": 9.234650611877441, |
| "learning_rate": 1.2553542009884679e-05, |
| "loss": 0.0867, |
| "step": 550 |
| }, |
| { |
| "epoch": 37.0, |
| "eval_accuracy": 0.88, |
| "eval_loss": 0.4715723991394043, |
| "eval_runtime": 2.0263, |
| "eval_samples_per_second": 86.365, |
| "eval_steps_per_second": 5.429, |
| "step": 555 |
| }, |
| { |
| "epoch": 37.333333333333336, |
| "grad_norm": 4.588763236999512, |
| "learning_rate": 1.156507413509061e-05, |
| "loss": 0.052, |
| "step": 560 |
| }, |
| { |
| "epoch": 38.0, |
| "grad_norm": 1.6751114130020142, |
| "learning_rate": 1.0576606260296541e-05, |
| "loss": 0.0721, |
| "step": 570 |
| }, |
| { |
| "epoch": 38.0, |
| "eval_accuracy": 0.8914285714285715, |
| "eval_loss": 0.435300350189209, |
| "eval_runtime": 2.0261, |
| "eval_samples_per_second": 86.371, |
| "eval_steps_per_second": 5.429, |
| "step": 570 |
| }, |
| { |
| "epoch": 38.666666666666664, |
| "grad_norm": 2.670736789703369, |
| "learning_rate": 9.588138385502471e-06, |
| "loss": 0.0572, |
| "step": 580 |
| }, |
| { |
| "epoch": 39.0, |
| "eval_accuracy": 0.8914285714285715, |
| "eval_loss": 0.44432708621025085, |
| "eval_runtime": 1.9183, |
| "eval_samples_per_second": 91.224, |
| "eval_steps_per_second": 5.734, |
| "step": 585 |
| }, |
| { |
| "epoch": 39.333333333333336, |
| "grad_norm": 6.209992408752441, |
| "learning_rate": 8.599670510708402e-06, |
| "loss": 0.0404, |
| "step": 590 |
| }, |
| { |
| "epoch": 40.0, |
| "grad_norm": 2.4059910774230957, |
| "learning_rate": 7.611202635914333e-06, |
| "loss": 0.0613, |
| "step": 600 |
| }, |
| { |
| "epoch": 40.0, |
| "eval_accuracy": 0.8857142857142857, |
| "eval_loss": 0.465512216091156, |
| "eval_runtime": 2.7259, |
| "eval_samples_per_second": 64.199, |
| "eval_steps_per_second": 4.035, |
| "step": 600 |
| }, |
| { |
| "epoch": 40.666666666666664, |
| "grad_norm": 4.62605094909668, |
| "learning_rate": 6.622734761120264e-06, |
| "loss": 0.0911, |
| "step": 610 |
| }, |
| { |
| "epoch": 41.0, |
| "eval_accuracy": 0.8971428571428571, |
| "eval_loss": 0.44622743129730225, |
| "eval_runtime": 1.8649, |
| "eval_samples_per_second": 93.84, |
| "eval_steps_per_second": 5.899, |
| "step": 615 |
| }, |
| { |
| "epoch": 41.333333333333336, |
| "grad_norm": 2.7355751991271973, |
| "learning_rate": 5.634266886326195e-06, |
| "loss": 0.0684, |
| "step": 620 |
| }, |
| { |
| "epoch": 42.0, |
| "grad_norm": 14.49720573425293, |
| "learning_rate": 4.645799011532125e-06, |
| "loss": 0.0691, |
| "step": 630 |
| }, |
| { |
| "epoch": 42.0, |
| "eval_accuracy": 0.88, |
| "eval_loss": 0.4921935200691223, |
| "eval_runtime": 1.8585, |
| "eval_samples_per_second": 94.163, |
| "eval_steps_per_second": 5.919, |
| "step": 630 |
| }, |
| { |
| "epoch": 42.666666666666664, |
| "grad_norm": 8.129971504211426, |
| "learning_rate": 3.6573311367380562e-06, |
| "loss": 0.0767, |
| "step": 640 |
| }, |
| { |
| "epoch": 43.0, |
| "eval_accuracy": 0.8742857142857143, |
| "eval_loss": 0.4701913893222809, |
| "eval_runtime": 2.1361, |
| "eval_samples_per_second": 81.924, |
| "eval_steps_per_second": 5.149, |
| "step": 645 |
| }, |
| { |
| "epoch": 43.333333333333336, |
| "grad_norm": 2.500708818435669, |
| "learning_rate": 2.668863261943987e-06, |
| "loss": 0.0453, |
| "step": 650 |
| }, |
| { |
| "epoch": 44.0, |
| "grad_norm": 13.184605598449707, |
| "learning_rate": 1.6803953871499178e-06, |
| "loss": 0.0763, |
| "step": 660 |
| }, |
| { |
| "epoch": 44.0, |
| "eval_accuracy": 0.8742857142857143, |
| "eval_loss": 0.46701833605766296, |
| "eval_runtime": 1.8632, |
| "eval_samples_per_second": 93.926, |
| "eval_steps_per_second": 5.904, |
| "step": 660 |
| }, |
| { |
| "epoch": 44.666666666666664, |
| "grad_norm": 10.216205596923828, |
| "learning_rate": 6.919275123558484e-07, |
| "loss": 0.055, |
| "step": 670 |
| }, |
| { |
| "epoch": 45.0, |
| "eval_accuracy": 0.8685714285714285, |
| "eval_loss": 0.4691820740699768, |
| "eval_runtime": 2.747, |
| "eval_samples_per_second": 63.705, |
| "eval_steps_per_second": 4.004, |
| "step": 675 |
| }, |
| { |
| "epoch": 45.0, |
| "step": 675, |
| "total_flos": 1.390881324367872e+18, |
| "train_loss": 0.1957703040043513, |
| "train_runtime": 1064.0603, |
| "train_samples_per_second": 40.176, |
| "train_steps_per_second": 0.634 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 675, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 45, |
| "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": 1.390881324367872e+18, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
