| { | |
| "best_metric": 0.019963504746556282, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-Kontur-competition-52K/checkpoint-496", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 1240, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 3.4303202629089355, | |
| "learning_rate": 4.032258064516129e-06, | |
| "loss": 0.7196, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 2.8987796306610107, | |
| "learning_rate": 8.064516129032258e-06, | |
| "loss": 0.6457, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 2.8838603496551514, | |
| "learning_rate": 1.2096774193548388e-05, | |
| "loss": 0.5219, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 2.1233789920806885, | |
| "learning_rate": 1.6129032258064517e-05, | |
| "loss": 0.4099, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 2.453838348388672, | |
| "learning_rate": 2.0161290322580645e-05, | |
| "loss": 0.2775, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 5.15359354019165, | |
| "learning_rate": 2.4193548387096777e-05, | |
| "loss": 0.1655, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 5.333171367645264, | |
| "learning_rate": 2.822580645161291e-05, | |
| "loss": 0.112, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 3.0966763496398926, | |
| "learning_rate": 3.2258064516129034e-05, | |
| "loss": 0.0818, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 13.956466674804688, | |
| "learning_rate": 3.6290322580645165e-05, | |
| "loss": 0.0884, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 3.109863758087158, | |
| "learning_rate": 4.032258064516129e-05, | |
| "loss": 0.0755, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 2.274998664855957, | |
| "learning_rate": 4.435483870967742e-05, | |
| "loss": 0.0562, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 5.711202621459961, | |
| "learning_rate": 4.8387096774193554e-05, | |
| "loss": 0.0622, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 9.838350296020508, | |
| "learning_rate": 4.973118279569893e-05, | |
| "loss": 0.0646, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 8.744378089904785, | |
| "learning_rate": 4.92831541218638e-05, | |
| "loss": 0.0631, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 13.287856101989746, | |
| "learning_rate": 4.8835125448028677e-05, | |
| "loss": 0.0732, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 4.635748863220215, | |
| "learning_rate": 4.8387096774193554e-05, | |
| "loss": 0.049, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 5.990776062011719, | |
| "learning_rate": 4.7939068100358424e-05, | |
| "loss": 0.0709, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 6.079687595367432, | |
| "learning_rate": 4.74910394265233e-05, | |
| "loss": 0.0492, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 1.6072505712509155, | |
| "learning_rate": 4.704301075268818e-05, | |
| "loss": 0.0527, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 1.9706329107284546, | |
| "learning_rate": 4.659498207885305e-05, | |
| "loss": 0.0609, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 5.284497261047363, | |
| "learning_rate": 4.614695340501792e-05, | |
| "loss": 0.0431, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 3.325084924697876, | |
| "learning_rate": 4.56989247311828e-05, | |
| "loss": 0.0321, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 9.077991485595703, | |
| "learning_rate": 4.5250896057347674e-05, | |
| "loss": 0.0375, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 4.708994388580322, | |
| "learning_rate": 4.4802867383512545e-05, | |
| "loss": 0.0376, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.06623569875955582, | |
| "eval_runtime": 113.2602, | |
| "eval_samples_per_second": 46.689, | |
| "eval_steps_per_second": 0.98, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 4.394409656524658, | |
| "learning_rate": 4.435483870967742e-05, | |
| "loss": 0.0348, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 20.027359008789062, | |
| "learning_rate": 4.390681003584229e-05, | |
| "loss": 0.0685, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "grad_norm": 2.6204428672790527, | |
| "learning_rate": 4.345878136200717e-05, | |
| "loss": 0.0445, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "grad_norm": 6.732935905456543, | |
| "learning_rate": 4.301075268817205e-05, | |
| "loss": 0.0357, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 9.537686347961426, | |
| "learning_rate": 4.256272401433692e-05, | |
| "loss": 0.0367, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 4.934083938598633, | |
| "learning_rate": 4.2114695340501795e-05, | |
| "loss": 0.0249, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 2.7931134700775146, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.0312, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "grad_norm": 1.5536555051803589, | |
| "learning_rate": 4.121863799283154e-05, | |
| "loss": 0.0329, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "grad_norm": 4.139862060546875, | |
| "learning_rate": 4.077060931899642e-05, | |
| "loss": 0.0306, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 2.881141185760498, | |
| "learning_rate": 4.032258064516129e-05, | |
| "loss": 0.0272, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "grad_norm": 4.688608646392822, | |
| "learning_rate": 3.987455197132617e-05, | |
| "loss": 0.0332, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 3.6917848587036133, | |
| "learning_rate": 3.9426523297491045e-05, | |
| "loss": 0.0365, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "grad_norm": 3.620156764984131, | |
| "learning_rate": 3.8978494623655915e-05, | |
| "loss": 0.0232, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "grad_norm": 5.679781913757324, | |
| "learning_rate": 3.8530465949820786e-05, | |
| "loss": 0.0259, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "grad_norm": 3.670245885848999, | |
| "learning_rate": 3.808243727598566e-05, | |
| "loss": 0.0296, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "grad_norm": 5.259506702423096, | |
| "learning_rate": 3.763440860215054e-05, | |
| "loss": 0.0155, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 2.921619176864624, | |
| "learning_rate": 3.718637992831541e-05, | |
| "loss": 0.0228, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 1.523906946182251, | |
| "learning_rate": 3.673835125448029e-05, | |
| "loss": 0.0197, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 4.975411891937256, | |
| "learning_rate": 3.6290322580645165e-05, | |
| "loss": 0.0296, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 3.5619239807128906, | |
| "learning_rate": 3.5842293906810036e-05, | |
| "loss": 0.0265, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "grad_norm": 2.4044506549835205, | |
| "learning_rate": 3.539426523297491e-05, | |
| "loss": 0.0217, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 8.599495887756348, | |
| "learning_rate": 3.494623655913979e-05, | |
| "loss": 0.034, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 4.817391395568848, | |
| "learning_rate": 3.449820788530466e-05, | |
| "loss": 0.0244, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 0.9919489622116089, | |
| "learning_rate": 3.405017921146954e-05, | |
| "loss": 0.0285, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 1.3462491035461426, | |
| "learning_rate": 3.360215053763441e-05, | |
| "loss": 0.0213, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.019963504746556282, | |
| "eval_runtime": 83.0975, | |
| "eval_samples_per_second": 63.636, | |
| "eval_steps_per_second": 1.336, | |
| "step": 496 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 2.6078598499298096, | |
| "learning_rate": 3.3154121863799286e-05, | |
| "loss": 0.029, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 0.9603502750396729, | |
| "learning_rate": 3.270609318996416e-05, | |
| "loss": 0.0263, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 11.797562599182129, | |
| "learning_rate": 3.2258064516129034e-05, | |
| "loss": 0.0256, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "grad_norm": 6.010237693786621, | |
| "learning_rate": 3.1810035842293904e-05, | |
| "loss": 0.0259, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "grad_norm": 11.067808151245117, | |
| "learning_rate": 3.136200716845878e-05, | |
| "loss": 0.0244, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "grad_norm": 8.755197525024414, | |
| "learning_rate": 3.091397849462366e-05, | |
| "loss": 0.0172, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "grad_norm": 5.274145603179932, | |
| "learning_rate": 3.046594982078853e-05, | |
| "loss": 0.0219, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 3.9025063514709473, | |
| "learning_rate": 3.0017921146953403e-05, | |
| "loss": 0.0121, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "grad_norm": 0.16941958665847778, | |
| "learning_rate": 2.9569892473118284e-05, | |
| "loss": 0.0151, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "grad_norm": 5.620195388793945, | |
| "learning_rate": 2.9121863799283154e-05, | |
| "loss": 0.02, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "grad_norm": 4.685577869415283, | |
| "learning_rate": 2.8673835125448028e-05, | |
| "loss": 0.0212, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 0.9184539914131165, | |
| "learning_rate": 2.822580645161291e-05, | |
| "loss": 0.0243, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 4.498030662536621, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.0199, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "grad_norm": 0.5963709354400635, | |
| "learning_rate": 2.7329749103942653e-05, | |
| "loss": 0.0189, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "grad_norm": 0.7427639961242676, | |
| "learning_rate": 2.6881720430107527e-05, | |
| "loss": 0.01, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "grad_norm": 1.3474091291427612, | |
| "learning_rate": 2.6433691756272404e-05, | |
| "loss": 0.0121, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "grad_norm": 2.8091518878936768, | |
| "learning_rate": 2.5985663082437278e-05, | |
| "loss": 0.0252, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 1.0539556741714478, | |
| "learning_rate": 2.5537634408602152e-05, | |
| "loss": 0.0192, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "grad_norm": 0.38558557629585266, | |
| "learning_rate": 2.5089605734767026e-05, | |
| "loss": 0.0115, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "grad_norm": 2.709695339202881, | |
| "learning_rate": 2.46415770609319e-05, | |
| "loss": 0.0178, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "grad_norm": 4.6164751052856445, | |
| "learning_rate": 2.4193548387096777e-05, | |
| "loss": 0.0118, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "grad_norm": 2.706479549407959, | |
| "learning_rate": 2.374551971326165e-05, | |
| "loss": 0.0165, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 2.981581211090088, | |
| "learning_rate": 2.3297491039426525e-05, | |
| "loss": 0.0206, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "grad_norm": 1.4225656986236572, | |
| "learning_rate": 2.28494623655914e-05, | |
| "loss": 0.0092, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "grad_norm": 1.1643812656402588, | |
| "learning_rate": 2.2401433691756272e-05, | |
| "loss": 0.0094, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.13013091683387756, | |
| "eval_runtime": 83.9945, | |
| "eval_samples_per_second": 62.957, | |
| "eval_steps_per_second": 1.322, | |
| "step": 744 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 2.4861252307891846, | |
| "learning_rate": 2.1953405017921146e-05, | |
| "loss": 0.0198, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "grad_norm": 2.1413233280181885, | |
| "learning_rate": 2.1505376344086024e-05, | |
| "loss": 0.0142, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 2.2505719661712646, | |
| "learning_rate": 2.1057347670250897e-05, | |
| "loss": 0.0151, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "grad_norm": 2.7683489322662354, | |
| "learning_rate": 2.060931899641577e-05, | |
| "loss": 0.0202, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "grad_norm": 1.9615308046340942, | |
| "learning_rate": 2.0161290322580645e-05, | |
| "loss": 0.0076, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "grad_norm": 2.115872621536255, | |
| "learning_rate": 1.9713261648745522e-05, | |
| "loss": 0.0133, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "grad_norm": 0.3355218768119812, | |
| "learning_rate": 1.9265232974910393e-05, | |
| "loss": 0.0149, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "grad_norm": 1.101788878440857, | |
| "learning_rate": 1.881720430107527e-05, | |
| "loss": 0.0168, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "grad_norm": 4.441117763519287, | |
| "learning_rate": 1.8369175627240144e-05, | |
| "loss": 0.0101, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "grad_norm": 0.4963151216506958, | |
| "learning_rate": 1.7921146953405018e-05, | |
| "loss": 0.0079, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "grad_norm": 0.14219801127910614, | |
| "learning_rate": 1.7473118279569895e-05, | |
| "loss": 0.0117, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "grad_norm": 0.04577568918466568, | |
| "learning_rate": 1.702508960573477e-05, | |
| "loss": 0.0065, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "grad_norm": 3.448432207107544, | |
| "learning_rate": 1.6577060931899643e-05, | |
| "loss": 0.0141, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "grad_norm": 0.43813198804855347, | |
| "learning_rate": 1.6129032258064517e-05, | |
| "loss": 0.0121, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "grad_norm": 1.8220750093460083, | |
| "learning_rate": 1.568100358422939e-05, | |
| "loss": 0.0057, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "grad_norm": 4.196650505065918, | |
| "learning_rate": 1.5232974910394265e-05, | |
| "loss": 0.0082, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "grad_norm": 1.0396842956542969, | |
| "learning_rate": 1.4784946236559142e-05, | |
| "loss": 0.0113, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "grad_norm": 3.6332926750183105, | |
| "learning_rate": 1.4336917562724014e-05, | |
| "loss": 0.0056, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 6.7421369552612305, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0085, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "grad_norm": 1.0675561428070068, | |
| "learning_rate": 1.3440860215053763e-05, | |
| "loss": 0.0061, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "grad_norm": 3.8124799728393555, | |
| "learning_rate": 1.2992831541218639e-05, | |
| "loss": 0.0087, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "grad_norm": 4.4822540283203125, | |
| "learning_rate": 1.2544802867383513e-05, | |
| "loss": 0.0048, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "grad_norm": 1.812424898147583, | |
| "learning_rate": 1.2096774193548388e-05, | |
| "loss": 0.0074, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "grad_norm": 6.007579803466797, | |
| "learning_rate": 1.1648745519713262e-05, | |
| "loss": 0.0116, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "grad_norm": 0.2030833214521408, | |
| "learning_rate": 1.1200716845878136e-05, | |
| "loss": 0.0103, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.060427818447351456, | |
| "eval_runtime": 84.6044, | |
| "eval_samples_per_second": 62.503, | |
| "eval_steps_per_second": 1.312, | |
| "step": 992 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "grad_norm": 0.05641289800405502, | |
| "learning_rate": 1.0752688172043012e-05, | |
| "loss": 0.0085, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "grad_norm": 0.805799126625061, | |
| "learning_rate": 1.0304659498207886e-05, | |
| "loss": 0.0077, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "grad_norm": 4.972249984741211, | |
| "learning_rate": 9.856630824372761e-06, | |
| "loss": 0.0113, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "grad_norm": 0.03549932688474655, | |
| "learning_rate": 9.408602150537635e-06, | |
| "loss": 0.0052, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "grad_norm": 1.279232144355774, | |
| "learning_rate": 8.960573476702509e-06, | |
| "loss": 0.0077, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "grad_norm": 0.22537721693515778, | |
| "learning_rate": 8.512544802867385e-06, | |
| "loss": 0.0082, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "grad_norm": 1.7011988162994385, | |
| "learning_rate": 8.064516129032258e-06, | |
| "loss": 0.0036, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "grad_norm": 4.37709379196167, | |
| "learning_rate": 7.616487455197132e-06, | |
| "loss": 0.0041, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "grad_norm": 2.2503268718719482, | |
| "learning_rate": 7.168458781362007e-06, | |
| "loss": 0.0152, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 3.610405683517456, | |
| "learning_rate": 6.720430107526882e-06, | |
| "loss": 0.0107, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "grad_norm": 0.22053413093090057, | |
| "learning_rate": 6.2724014336917564e-06, | |
| "loss": 0.0021, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "grad_norm": 8.507258415222168, | |
| "learning_rate": 5.824372759856631e-06, | |
| "loss": 0.0141, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "grad_norm": 5.034768581390381, | |
| "learning_rate": 5.376344086021506e-06, | |
| "loss": 0.0142, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "grad_norm": 0.22189994156360626, | |
| "learning_rate": 4.928315412186381e-06, | |
| "loss": 0.0094, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "grad_norm": 2.6591525077819824, | |
| "learning_rate": 4.4802867383512545e-06, | |
| "loss": 0.0039, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "grad_norm": 4.526949405670166, | |
| "learning_rate": 4.032258064516129e-06, | |
| "loss": 0.0032, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "grad_norm": 3.290435791015625, | |
| "learning_rate": 3.5842293906810035e-06, | |
| "loss": 0.0031, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "grad_norm": 0.6715773344039917, | |
| "learning_rate": 3.1362007168458782e-06, | |
| "loss": 0.0028, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "grad_norm": 0.7959145903587341, | |
| "learning_rate": 2.688172043010753e-06, | |
| "loss": 0.0063, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.10818018019199371, | |
| "learning_rate": 2.2401433691756272e-06, | |
| "loss": 0.0045, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "grad_norm": 0.03141075000166893, | |
| "learning_rate": 1.7921146953405017e-06, | |
| "loss": 0.0068, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "grad_norm": 5.118274688720703, | |
| "learning_rate": 1.3440860215053765e-06, | |
| "loss": 0.0071, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "grad_norm": 1.3246444463729858, | |
| "learning_rate": 8.960573476702509e-07, | |
| "loss": 0.0026, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "grad_norm": 2.687753915786743, | |
| "learning_rate": 4.4802867383512544e-07, | |
| "loss": 0.0046, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.166173055768013, | |
| "learning_rate": 0.0, | |
| "loss": 0.0073, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.07113554328680038, | |
| "eval_runtime": 85.5998, | |
| "eval_samples_per_second": 61.776, | |
| "eval_steps_per_second": 1.297, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 1240, | |
| "total_flos": 5.914482579184435e+18, | |
| "train_loss": 0.04452576920570385, | |
| "train_runtime": 5366.5614, | |
| "train_samples_per_second": 44.339, | |
| "train_steps_per_second": 0.231 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1240, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 5.914482579184435e+18, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |