| { | |
| "best_metric": 0.6274509803921569, | |
| "best_model_checkpoint": "/users/hailey/code/03_code/SBD_ai/videomae-base-finetuned-ucf101-subset-SBDtoy/checkpoint-492", | |
| "epoch": 29.009166666666665, | |
| "eval_steps": 500, | |
| "global_step": 1200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 8.483016014099121, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.8897, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 10.62829303741455, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.7501, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 10.224696159362793, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.6748, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 5.6519365310668945, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.7634, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_accuracy": 0.5882352941176471, | |
| "eval_loss": 0.6320443153381348, | |
| "eval_runtime": 137.732, | |
| "eval_samples_per_second": 0.37, | |
| "eval_steps_per_second": 0.189, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 8.263680458068848, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.7384, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 9.806657791137695, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.6558, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 4.578342437744141, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 0.569, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 1.7302507162094116, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.7445, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "eval_accuracy": 0.5686274509803921, | |
| "eval_loss": 1.4197746515274048, | |
| "eval_runtime": 132.2118, | |
| "eval_samples_per_second": 0.386, | |
| "eval_steps_per_second": 0.197, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 29.237693786621094, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.7204, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 3.5326147079467773, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.7924, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 10.499820709228516, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 0.6213, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 6.973377227783203, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4931, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "eval_accuracy": 0.5686274509803921, | |
| "eval_loss": 1.3688796758651733, | |
| "eval_runtime": 132.0446, | |
| "eval_samples_per_second": 0.386, | |
| "eval_steps_per_second": 0.197, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "grad_norm": 24.18391990661621, | |
| "learning_rate": 4.9537037037037035e-05, | |
| "loss": 0.6786, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "grad_norm": 19.578279495239258, | |
| "learning_rate": 4.9074074074074075e-05, | |
| "loss": 0.597, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 0.4028966724872589, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 0.7246, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "grad_norm": 29.119096755981445, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 0.7206, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "eval_accuracy": 0.5882352941176471, | |
| "eval_loss": 1.328170657157898, | |
| "eval_runtime": 134.48, | |
| "eval_samples_per_second": 0.379, | |
| "eval_steps_per_second": 0.193, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 4.413028717041016, | |
| "learning_rate": 4.768518518518519e-05, | |
| "loss": 0.6192, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "grad_norm": 0.44842419028282166, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.3903, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "grad_norm": 9.306065559387207, | |
| "learning_rate": 4.675925925925926e-05, | |
| "loss": 0.7218, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "grad_norm": 52.77680587768555, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.735, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "eval_accuracy": 0.5686274509803921, | |
| "eval_loss": 2.0849499702453613, | |
| "eval_runtime": 132.8084, | |
| "eval_samples_per_second": 0.384, | |
| "eval_steps_per_second": 0.196, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 20.23072624206543, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 0.9199, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "grad_norm": 3.353752613067627, | |
| "learning_rate": 4.5370370370370374e-05, | |
| "loss": 0.4171, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "grad_norm": 0.07062011957168579, | |
| "learning_rate": 4.490740740740741e-05, | |
| "loss": 0.1716, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "grad_norm": 17.275123596191406, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 1.769, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 1.0332462787628174, | |
| "eval_runtime": 141.0066, | |
| "eval_samples_per_second": 0.362, | |
| "eval_steps_per_second": 0.184, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.1296720504760742, | |
| "learning_rate": 4.3981481481481486e-05, | |
| "loss": 0.4991, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "grad_norm": 0.18362000584602356, | |
| "learning_rate": 4.351851851851852e-05, | |
| "loss": 0.4565, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "grad_norm": 9.544737815856934, | |
| "learning_rate": 4.305555555555556e-05, | |
| "loss": 1.5794, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "grad_norm": 9.105454444885254, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 0.9494, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 2.130749464035034, | |
| "eval_runtime": 147.5482, | |
| "eval_samples_per_second": 0.346, | |
| "eval_steps_per_second": 0.176, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.585872769355774, | |
| "learning_rate": 4.212962962962963e-05, | |
| "loss": 0.3491, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "grad_norm": 6.12731409072876, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.5312, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "grad_norm": 20.4097957611084, | |
| "learning_rate": 4.1203703703703705e-05, | |
| "loss": 0.6494, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "grad_norm": 1.075332522392273, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 0.3857, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 2.598536252975464, | |
| "eval_runtime": 141.9381, | |
| "eval_samples_per_second": 0.359, | |
| "eval_steps_per_second": 0.183, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.12138959765434265, | |
| "learning_rate": 4.027777777777778e-05, | |
| "loss": 0.8439, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "grad_norm": 68.02975463867188, | |
| "learning_rate": 3.981481481481482e-05, | |
| "loss": 0.8624, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "grad_norm": 0.4764052927494049, | |
| "learning_rate": 3.935185185185186e-05, | |
| "loss": 0.3225, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "grad_norm": 1.0622520446777344, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.3639, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 2.2299933433532715, | |
| "eval_runtime": 147.3983, | |
| "eval_samples_per_second": 0.346, | |
| "eval_steps_per_second": 0.176, | |
| "step": 369 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 24.868986129760742, | |
| "learning_rate": 3.8425925925925924e-05, | |
| "loss": 0.6411, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "grad_norm": 0.08051346242427826, | |
| "learning_rate": 3.7962962962962964e-05, | |
| "loss": 0.2678, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "grad_norm": 14.524049758911133, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 1.1313, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "grad_norm": 5.631989479064941, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.7205, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "grad_norm": 0.37318071722984314, | |
| "learning_rate": 3.6574074074074076e-05, | |
| "loss": 0.6456, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 1.8992350101470947, | |
| "eval_runtime": 135.9679, | |
| "eval_samples_per_second": 0.375, | |
| "eval_steps_per_second": 0.191, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "grad_norm": 0.06262947618961334, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.331, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "grad_norm": 0.16807079315185547, | |
| "learning_rate": 3.564814814814815e-05, | |
| "loss": 0.4851, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "grad_norm": 0.23209631443023682, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.6103, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "grad_norm": 0.20488207042217255, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.9483, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 1.955637812614441, | |
| "eval_runtime": 130.9716, | |
| "eval_samples_per_second": 0.389, | |
| "eval_steps_per_second": 0.199, | |
| "step": 451 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "grad_norm": 42.45355987548828, | |
| "learning_rate": 3.425925925925926e-05, | |
| "loss": 0.3743, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "grad_norm": 0.16700521111488342, | |
| "learning_rate": 3.3796296296296295e-05, | |
| "loss": 0.368, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "grad_norm": 1.189526915550232, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.8548, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "grad_norm": 23.483234405517578, | |
| "learning_rate": 3.2870370370370375e-05, | |
| "loss": 0.4518, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "eval_accuracy": 0.6274509803921569, | |
| "eval_loss": 1.834560751914978, | |
| "eval_runtime": 139.8695, | |
| "eval_samples_per_second": 0.365, | |
| "eval_steps_per_second": 0.186, | |
| "step": 492 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "grad_norm": 14.332037925720215, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.3553, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "grad_norm": 0.2652936279773712, | |
| "learning_rate": 3.194444444444444e-05, | |
| "loss": 0.3726, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "grad_norm": 0.1738712340593338, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 0.4627, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 12.03, | |
| "grad_norm": 16.079055786132812, | |
| "learning_rate": 3.101851851851852e-05, | |
| "loss": 0.9109, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 12.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 1.993680477142334, | |
| "eval_runtime": 138.9405, | |
| "eval_samples_per_second": 0.367, | |
| "eval_steps_per_second": 0.187, | |
| "step": 533 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "grad_norm": 0.2999970018863678, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.1598, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "grad_norm": 0.08949922025203705, | |
| "learning_rate": 3.0092592592592593e-05, | |
| "loss": 0.0925, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "grad_norm": 13.087860107421875, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.8925, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "grad_norm": 12.626638412475586, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 0.5441, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 1.401322364807129, | |
| "eval_runtime": 140.7378, | |
| "eval_samples_per_second": 0.362, | |
| "eval_steps_per_second": 0.185, | |
| "step": 574 | |
| }, | |
| { | |
| "epoch": 14.01, | |
| "grad_norm": 0.09884772449731827, | |
| "learning_rate": 2.8703703703703706e-05, | |
| "loss": 0.2266, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 14.01, | |
| "grad_norm": 0.16622687876224518, | |
| "learning_rate": 2.824074074074074e-05, | |
| "loss": 0.0233, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "grad_norm": 0.08121990412473679, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.4697, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "grad_norm": 0.04321848973631859, | |
| "learning_rate": 2.7314814814814816e-05, | |
| "loss": 0.4035, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 1.9077707529067993, | |
| "eval_runtime": 140.6526, | |
| "eval_samples_per_second": 0.363, | |
| "eval_steps_per_second": 0.185, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.08672894537448883, | |
| "learning_rate": 2.6851851851851855e-05, | |
| "loss": 0.7602, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "grad_norm": 0.03848231956362724, | |
| "learning_rate": 2.6388888888888892e-05, | |
| "loss": 0.0041, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 15.02, | |
| "grad_norm": 0.09817267954349518, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.7232, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "grad_norm": 2.0446386337280273, | |
| "learning_rate": 2.5462962962962965e-05, | |
| "loss": 0.1713, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "eval_accuracy": 0.5882352941176471, | |
| "eval_loss": 2.080268383026123, | |
| "eval_runtime": 140.5712, | |
| "eval_samples_per_second": 0.363, | |
| "eval_steps_per_second": 0.185, | |
| "step": 656 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 13.678278923034668, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.4517, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 16.01, | |
| "grad_norm": 6.980419635772705, | |
| "learning_rate": 2.4537037037037038e-05, | |
| "loss": 0.0059, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 16.02, | |
| "grad_norm": 0.3049313724040985, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.0051, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 16.03, | |
| "grad_norm": 0.06310291588306427, | |
| "learning_rate": 2.361111111111111e-05, | |
| "loss": 0.0542, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 16.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 2.543215036392212, | |
| "eval_runtime": 139.4428, | |
| "eval_samples_per_second": 0.366, | |
| "eval_steps_per_second": 0.186, | |
| "step": 697 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.047225985676050186, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.4816, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 17.01, | |
| "grad_norm": 0.025472501292824745, | |
| "learning_rate": 2.2685185185185187e-05, | |
| "loss": 0.2573, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 17.02, | |
| "grad_norm": 0.1224554181098938, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0025, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 17.03, | |
| "grad_norm": 0.01003193762153387, | |
| "learning_rate": 2.175925925925926e-05, | |
| "loss": 0.3084, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 17.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 2.57525372505188, | |
| "eval_runtime": 138.9495, | |
| "eval_samples_per_second": 0.367, | |
| "eval_steps_per_second": 0.187, | |
| "step": 738 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 1.1518467664718628, | |
| "learning_rate": 2.1296296296296296e-05, | |
| "loss": 0.242, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 18.01, | |
| "grad_norm": 0.033695243299007416, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.2333, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 18.02, | |
| "grad_norm": 0.06799116730690002, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 0.0012, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 18.03, | |
| "grad_norm": 0.00848899595439434, | |
| "learning_rate": 1.990740740740741e-05, | |
| "loss": 0.2476, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 18.03, | |
| "eval_accuracy": 0.6274509803921569, | |
| "eval_loss": 2.4252748489379883, | |
| "eval_runtime": 140.2598, | |
| "eval_samples_per_second": 0.364, | |
| "eval_steps_per_second": 0.185, | |
| "step": 779 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.015138844028115273, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.3527, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 19.01, | |
| "grad_norm": 0.008504153229296207, | |
| "learning_rate": 1.8981481481481482e-05, | |
| "loss": 0.0087, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 19.02, | |
| "grad_norm": 140.87295532226562, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.2821, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 19.03, | |
| "grad_norm": 0.027083896100521088, | |
| "learning_rate": 1.8055555555555555e-05, | |
| "loss": 0.404, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 19.03, | |
| "grad_norm": 0.15064358711242676, | |
| "learning_rate": 1.7592592592592595e-05, | |
| "loss": 0.006, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 19.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 2.4320342540740967, | |
| "eval_runtime": 136.4673, | |
| "eval_samples_per_second": 0.374, | |
| "eval_steps_per_second": 0.191, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 20.01, | |
| "grad_norm": 0.05090714618563652, | |
| "learning_rate": 1.712962962962963e-05, | |
| "loss": 0.1214, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 20.02, | |
| "grad_norm": 0.05721502751111984, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.3519, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 20.02, | |
| "grad_norm": 0.025916827842593193, | |
| "learning_rate": 1.6203703703703704e-05, | |
| "loss": 0.0014, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 20.03, | |
| "grad_norm": 0.04987427219748497, | |
| "learning_rate": 1.574074074074074e-05, | |
| "loss": 0.0021, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 20.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 2.718170166015625, | |
| "eval_runtime": 143.3116, | |
| "eval_samples_per_second": 0.356, | |
| "eval_steps_per_second": 0.181, | |
| "step": 861 | |
| }, | |
| { | |
| "epoch": 21.01, | |
| "grad_norm": 0.009684134274721146, | |
| "learning_rate": 1.527777777777778e-05, | |
| "loss": 0.22, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 21.02, | |
| "grad_norm": 20.616426467895508, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.2033, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 21.02, | |
| "grad_norm": 0.028017116710543633, | |
| "learning_rate": 1.4351851851851853e-05, | |
| "loss": 0.4944, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 21.03, | |
| "grad_norm": 0.04853179305791855, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.008, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 21.03, | |
| "eval_accuracy": 0.5882352941176471, | |
| "eval_loss": 2.771120548248291, | |
| "eval_runtime": 141.504, | |
| "eval_samples_per_second": 0.36, | |
| "eval_steps_per_second": 0.184, | |
| "step": 902 | |
| }, | |
| { | |
| "epoch": 22.01, | |
| "grad_norm": 0.010195921175181866, | |
| "learning_rate": 1.3425925925925928e-05, | |
| "loss": 0.2491, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 22.02, | |
| "grad_norm": 0.058783795684576035, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.1002, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 22.02, | |
| "grad_norm": 0.009438421577215195, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2846, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 22.03, | |
| "grad_norm": 0.017627988010644913, | |
| "learning_rate": 1.2037037037037037e-05, | |
| "loss": 0.0011, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 22.03, | |
| "eval_accuracy": 0.5882352941176471, | |
| "eval_loss": 2.808924674987793, | |
| "eval_runtime": 140.1443, | |
| "eval_samples_per_second": 0.364, | |
| "eval_steps_per_second": 0.186, | |
| "step": 943 | |
| }, | |
| { | |
| "epoch": 23.01, | |
| "grad_norm": 0.009988226927816868, | |
| "learning_rate": 1.1574074074074075e-05, | |
| "loss": 0.0157, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 23.01, | |
| "grad_norm": 0.02875751443207264, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0036, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 23.02, | |
| "grad_norm": 0.5501790046691895, | |
| "learning_rate": 1.0648148148148148e-05, | |
| "loss": 0.0029, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 23.03, | |
| "grad_norm": 0.013142755255103111, | |
| "learning_rate": 1.0185185185185185e-05, | |
| "loss": 0.7546, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 23.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 2.8142335414886475, | |
| "eval_runtime": 147.8203, | |
| "eval_samples_per_second": 0.345, | |
| "eval_steps_per_second": 0.176, | |
| "step": 984 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 1.282440423965454, | |
| "learning_rate": 9.722222222222223e-06, | |
| "loss": 0.0821, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 24.01, | |
| "grad_norm": 0.011100966483354568, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.0014, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 24.02, | |
| "grad_norm": 0.020793117582798004, | |
| "learning_rate": 8.796296296296297e-06, | |
| "loss": 0.1571, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 24.03, | |
| "grad_norm": 0.02288602478802204, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.0006, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 24.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 2.8796093463897705, | |
| "eval_runtime": 142.281, | |
| "eval_samples_per_second": 0.358, | |
| "eval_steps_per_second": 0.183, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 0.031101850792765617, | |
| "learning_rate": 7.87037037037037e-06, | |
| "loss": 0.0051, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 25.01, | |
| "grad_norm": 0.017316602170467377, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.0006, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 25.02, | |
| "grad_norm": 0.015025320462882519, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.2444, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 25.03, | |
| "grad_norm": 0.014637992717325687, | |
| "learning_rate": 6.481481481481481e-06, | |
| "loss": 0.0008, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 25.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 2.8485913276672363, | |
| "eval_runtime": 140.4219, | |
| "eval_samples_per_second": 0.363, | |
| "eval_steps_per_second": 0.185, | |
| "step": 1066 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "grad_norm": 0.1402996927499771, | |
| "learning_rate": 6.0185185185185185e-06, | |
| "loss": 0.1973, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 26.01, | |
| "grad_norm": 0.006944969296455383, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0005, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 26.02, | |
| "grad_norm": 18.494768142700195, | |
| "learning_rate": 5.092592592592592e-06, | |
| "loss": 0.1727, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 26.03, | |
| "grad_norm": 0.010701355524361134, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.1745, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 26.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 2.8475284576416016, | |
| "eval_runtime": 143.4719, | |
| "eval_samples_per_second": 0.355, | |
| "eval_steps_per_second": 0.181, | |
| "step": 1107 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "grad_norm": 0.012416614219546318, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.0074, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 27.01, | |
| "grad_norm": 0.06672808527946472, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.0006, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 27.02, | |
| "grad_norm": 0.008200524374842644, | |
| "learning_rate": 3.2407407407407406e-06, | |
| "loss": 0.0006, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 27.03, | |
| "grad_norm": 14.067534446716309, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.5421, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 27.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 2.8462321758270264, | |
| "eval_runtime": 137.4454, | |
| "eval_samples_per_second": 0.371, | |
| "eval_steps_per_second": 0.189, | |
| "step": 1148 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 0.009883771650493145, | |
| "learning_rate": 2.3148148148148148e-06, | |
| "loss": 0.0889, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 28.01, | |
| "grad_norm": 0.007996085099875927, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.1251, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 28.02, | |
| "grad_norm": 0.008312738500535488, | |
| "learning_rate": 1.388888888888889e-06, | |
| "loss": 0.339, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 28.03, | |
| "grad_norm": 0.07012984901666641, | |
| "learning_rate": 9.259259259259259e-07, | |
| "loss": 0.1233, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 28.03, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 2.8324010372161865, | |
| "eval_runtime": 132.7093, | |
| "eval_samples_per_second": 0.384, | |
| "eval_steps_per_second": 0.196, | |
| "step": 1189 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "grad_norm": 0.029733408242464066, | |
| "learning_rate": 4.6296296296296297e-07, | |
| "loss": 0.0054, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 29.01, | |
| "grad_norm": 0.033483583480119705, | |
| "learning_rate": 0.0, | |
| "loss": 0.1298, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 29.01, | |
| "eval_accuracy": 0.6078431372549019, | |
| "eval_loss": 2.833205461502075, | |
| "eval_runtime": 130.982, | |
| "eval_samples_per_second": 0.389, | |
| "eval_steps_per_second": 0.199, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 29.01, | |
| "step": 1200, | |
| "total_flos": 2.954419141637505e+18, | |
| "train_loss": 0.40525790957190716, | |
| "train_runtime": 10653.7383, | |
| "train_samples_per_second": 0.225, | |
| "train_steps_per_second": 0.113 | |
| }, | |
| { | |
| "epoch": 29.01, | |
| "eval_accuracy": 0.6274509803921569, | |
| "eval_loss": 1.834560751914978, | |
| "eval_runtime": 132.9753, | |
| "eval_samples_per_second": 0.384, | |
| "eval_steps_per_second": 0.196, | |
| "step": 1200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 2.954419141637505e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |