| { | |
| "best_metric": 0.9417692129092176, | |
| "best_model_checkpoint": "Crosswalk/dinov2/checkpoint-924", | |
| "epoch": 22.0, | |
| "eval_steps": 500, | |
| "global_step": 924, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.24242424242424243, | |
| "grad_norm": 1809.6781005859375, | |
| "learning_rate": 9.70873786407767e-07, | |
| "loss": 4.7087, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.48484848484848486, | |
| "grad_norm": 190.5601806640625, | |
| "learning_rate": 1.941747572815534e-06, | |
| "loss": 3.034, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 76.29146575927734, | |
| "learning_rate": 2.912621359223301e-06, | |
| "loss": 2.0024, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.9696969696969697, | |
| "grad_norm": 75.1009750366211, | |
| "learning_rate": 3.883495145631068e-06, | |
| "loss": 1.4019, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.2978227138519287, | |
| "eval_macro_f1": 0.8815037150933147, | |
| "eval_runtime": 7.5247, | |
| "eval_samples_per_second": 43.856, | |
| "eval_steps_per_second": 5.582, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.1939393939393939, | |
| "grad_norm": 86.90376281738281, | |
| "learning_rate": 4.854368932038836e-06, | |
| "loss": 0.7179, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.4363636363636363, | |
| "grad_norm": 118.144287109375, | |
| "learning_rate": 5.825242718446602e-06, | |
| "loss": 0.9737, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.6787878787878787, | |
| "grad_norm": 174.987548828125, | |
| "learning_rate": 6.79611650485437e-06, | |
| "loss": 1.3976, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.9212121212121214, | |
| "grad_norm": 35.86643981933594, | |
| "learning_rate": 7.766990291262136e-06, | |
| "loss": 0.967, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.22084859013557434, | |
| "eval_macro_f1": 0.9229339286881953, | |
| "eval_runtime": 5.8751, | |
| "eval_samples_per_second": 56.169, | |
| "eval_steps_per_second": 7.149, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.1454545454545455, | |
| "grad_norm": 232.57371520996094, | |
| "learning_rate": 8.737864077669904e-06, | |
| "loss": 1.323, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.3878787878787877, | |
| "grad_norm": 72.19599914550781, | |
| "learning_rate": 9.708737864077671e-06, | |
| "loss": 1.005, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.6303030303030304, | |
| "grad_norm": 55.63515853881836, | |
| "learning_rate": 9.924078091106291e-06, | |
| "loss": 0.6989, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.8727272727272726, | |
| "grad_norm": 56.631591796875, | |
| "learning_rate": 9.815618221258135e-06, | |
| "loss": 0.7527, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.3025018870830536, | |
| "eval_macro_f1": 0.9009343690194753, | |
| "eval_runtime": 6.0811, | |
| "eval_samples_per_second": 54.266, | |
| "eval_steps_per_second": 6.907, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 3.096969696969697, | |
| "grad_norm": 40.3350715637207, | |
| "learning_rate": 9.70715835140998e-06, | |
| "loss": 0.6221, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.3393939393939394, | |
| "grad_norm": 95.85454559326172, | |
| "learning_rate": 9.598698481561823e-06, | |
| "loss": 0.9071, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.581818181818182, | |
| "grad_norm": 94.18701934814453, | |
| "learning_rate": 9.490238611713667e-06, | |
| "loss": 0.7042, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.824242424242424, | |
| "grad_norm": 477.8069763183594, | |
| "learning_rate": 9.38177874186551e-06, | |
| "loss": 0.635, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.22365985810756683, | |
| "eval_macro_f1": 0.9049918736939866, | |
| "eval_runtime": 5.9294, | |
| "eval_samples_per_second": 55.655, | |
| "eval_steps_per_second": 7.083, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 4.048484848484849, | |
| "grad_norm": 110.27919006347656, | |
| "learning_rate": 9.273318872017354e-06, | |
| "loss": 1.1381, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.290909090909091, | |
| "grad_norm": 58.735958099365234, | |
| "learning_rate": 9.1648590021692e-06, | |
| "loss": 0.7225, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.533333333333333, | |
| "grad_norm": 75.20926666259766, | |
| "learning_rate": 9.056399132321042e-06, | |
| "loss": 0.4634, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.775757575757575, | |
| "grad_norm": 18.876136779785156, | |
| "learning_rate": 8.947939262472886e-06, | |
| "loss": 0.6293, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 8.281109809875488, | |
| "learning_rate": 8.83947939262473e-06, | |
| "loss": 0.6632, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.2299780696630478, | |
| "eval_macro_f1": 0.9176304185040354, | |
| "eval_runtime": 6.0281, | |
| "eval_samples_per_second": 54.744, | |
| "eval_steps_per_second": 6.967, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.242424242424242, | |
| "grad_norm": 13.756321907043457, | |
| "learning_rate": 8.731019522776574e-06, | |
| "loss": 0.4708, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.484848484848484, | |
| "grad_norm": 59.22605895996094, | |
| "learning_rate": 8.622559652928418e-06, | |
| "loss": 0.7127, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 5.7272727272727275, | |
| "grad_norm": 32.43043899536133, | |
| "learning_rate": 8.514099783080262e-06, | |
| "loss": 0.5682, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 5.96969696969697, | |
| "grad_norm": 54.722599029541016, | |
| "learning_rate": 8.405639913232104e-06, | |
| "loss": 0.8667, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.2767850160598755, | |
| "eval_macro_f1": 0.9210700618192522, | |
| "eval_runtime": 6.2004, | |
| "eval_samples_per_second": 53.223, | |
| "eval_steps_per_second": 6.774, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 6.193939393939394, | |
| "grad_norm": 18.175823211669922, | |
| "learning_rate": 8.29718004338395e-06, | |
| "loss": 0.6752, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.4363636363636365, | |
| "grad_norm": 56.679988861083984, | |
| "learning_rate": 8.188720173535792e-06, | |
| "loss": 0.3727, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 6.678787878787879, | |
| "grad_norm": 57.3917236328125, | |
| "learning_rate": 8.080260303687636e-06, | |
| "loss": 1.0167, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 6.921212121212121, | |
| "grad_norm": 42.186038970947266, | |
| "learning_rate": 7.97180043383948e-06, | |
| "loss": 0.9377, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.29274508357048035, | |
| "eval_macro_f1": 0.9138863000931967, | |
| "eval_runtime": 6.1213, | |
| "eval_samples_per_second": 53.91, | |
| "eval_steps_per_second": 6.861, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 7.1454545454545455, | |
| "grad_norm": 41.31782531738281, | |
| "learning_rate": 7.863340563991324e-06, | |
| "loss": 0.3818, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.387878787878788, | |
| "grad_norm": 4.223178863525391, | |
| "learning_rate": 7.754880694143168e-06, | |
| "loss": 0.4503, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 7.63030303030303, | |
| "grad_norm": 35.64258575439453, | |
| "learning_rate": 7.646420824295012e-06, | |
| "loss": 0.6038, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 7.872727272727273, | |
| "grad_norm": 37.91206359863281, | |
| "learning_rate": 7.537960954446856e-06, | |
| "loss": 0.5407, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.20143219828605652, | |
| "eval_macro_f1": 0.9357970705676355, | |
| "eval_runtime": 5.9715, | |
| "eval_samples_per_second": 55.263, | |
| "eval_steps_per_second": 7.033, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 8.096969696969698, | |
| "grad_norm": 9.571391105651855, | |
| "learning_rate": 7.429501084598699e-06, | |
| "loss": 0.3311, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.33939393939394, | |
| "grad_norm": 30.14655876159668, | |
| "learning_rate": 7.321041214750543e-06, | |
| "loss": 0.5367, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 8.581818181818182, | |
| "grad_norm": 125.38350677490234, | |
| "learning_rate": 7.212581344902386e-06, | |
| "loss": 0.4511, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 8.824242424242424, | |
| "grad_norm": 283.20819091796875, | |
| "learning_rate": 7.104121475054231e-06, | |
| "loss": 0.5474, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.329227477312088, | |
| "eval_macro_f1": 0.8817302125547928, | |
| "eval_runtime": 5.984, | |
| "eval_samples_per_second": 55.147, | |
| "eval_steps_per_second": 7.019, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 9.048484848484849, | |
| "grad_norm": 31.927379608154297, | |
| "learning_rate": 6.995661605206075e-06, | |
| "loss": 0.3963, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.290909090909091, | |
| "grad_norm": 10.10098934173584, | |
| "learning_rate": 6.887201735357918e-06, | |
| "loss": 0.445, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 9.533333333333333, | |
| "grad_norm": 1.947770118713379, | |
| "learning_rate": 6.778741865509761e-06, | |
| "loss": 0.5947, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 9.775757575757575, | |
| "grad_norm": 54.11802673339844, | |
| "learning_rate": 6.670281995661606e-06, | |
| "loss": 0.6001, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.004518165718764067, | |
| "learning_rate": 6.56182212581345e-06, | |
| "loss": 0.412, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.3594599962234497, | |
| "eval_macro_f1": 0.907735321528425, | |
| "eval_runtime": 6.0125, | |
| "eval_samples_per_second": 54.885, | |
| "eval_steps_per_second": 6.985, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 10.242424242424242, | |
| "grad_norm": 0.22337216138839722, | |
| "learning_rate": 6.453362255965293e-06, | |
| "loss": 0.2798, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 10.484848484848484, | |
| "grad_norm": 39.639984130859375, | |
| "learning_rate": 6.344902386117138e-06, | |
| "loss": 0.4377, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 10.727272727272727, | |
| "grad_norm": 53.85198211669922, | |
| "learning_rate": 6.236442516268981e-06, | |
| "loss": 0.2063, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 10.969696969696969, | |
| "grad_norm": 69.08942413330078, | |
| "learning_rate": 6.127982646420825e-06, | |
| "loss": 0.2884, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.2930862307548523, | |
| "eval_macro_f1": 0.9380839806371721, | |
| "eval_runtime": 6.0147, | |
| "eval_samples_per_second": 54.866, | |
| "eval_steps_per_second": 6.983, | |
| "step": 462 | |
| }, | |
| { | |
| "epoch": 11.193939393939393, | |
| "grad_norm": 63.143218994140625, | |
| "learning_rate": 6.019522776572668e-06, | |
| "loss": 0.6075, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 11.436363636363636, | |
| "grad_norm": 7.950187683105469, | |
| "learning_rate": 5.911062906724513e-06, | |
| "loss": 0.2654, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 11.67878787878788, | |
| "grad_norm": 82.30758666992188, | |
| "learning_rate": 5.802603036876356e-06, | |
| "loss": 0.2474, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 11.921212121212122, | |
| "grad_norm": 7.340689182281494, | |
| "learning_rate": 5.6941431670282e-06, | |
| "loss": 0.2405, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.3316686451435089, | |
| "eval_macro_f1": 0.9209216589861751, | |
| "eval_runtime": 5.8916, | |
| "eval_samples_per_second": 56.012, | |
| "eval_steps_per_second": 7.129, | |
| "step": 504 | |
| }, | |
| { | |
| "epoch": 12.145454545454545, | |
| "grad_norm": 37.141632080078125, | |
| "learning_rate": 5.585683297180043e-06, | |
| "loss": 0.3349, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 12.387878787878789, | |
| "grad_norm": 34.024383544921875, | |
| "learning_rate": 5.477223427331888e-06, | |
| "loss": 0.1742, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 12.63030303030303, | |
| "grad_norm": 46.10781478881836, | |
| "learning_rate": 5.368763557483731e-06, | |
| "loss": 0.2115, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 12.872727272727273, | |
| "grad_norm": 126.67015838623047, | |
| "learning_rate": 5.260303687635575e-06, | |
| "loss": 0.8788, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.37741926312446594, | |
| "eval_macro_f1": 0.9058106453305834, | |
| "eval_runtime": 6.6329, | |
| "eval_samples_per_second": 49.752, | |
| "eval_steps_per_second": 6.332, | |
| "step": 546 | |
| }, | |
| { | |
| "epoch": 13.096969696969698, | |
| "grad_norm": 35.44662094116211, | |
| "learning_rate": 5.151843817787418e-06, | |
| "loss": 0.5591, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 13.33939393939394, | |
| "grad_norm": 57.34544372558594, | |
| "learning_rate": 5.043383947939263e-06, | |
| "loss": 0.213, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 13.581818181818182, | |
| "grad_norm": 15.285223960876465, | |
| "learning_rate": 4.934924078091107e-06, | |
| "loss": 0.203, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 13.824242424242424, | |
| "grad_norm": 28.99003028869629, | |
| "learning_rate": 4.82646420824295e-06, | |
| "loss": 0.4163, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.39865490794181824, | |
| "eval_macro_f1": 0.9196508840275697, | |
| "eval_runtime": 5.8701, | |
| "eval_samples_per_second": 56.217, | |
| "eval_steps_per_second": 7.155, | |
| "step": 588 | |
| }, | |
| { | |
| "epoch": 14.048484848484849, | |
| "grad_norm": 346.8255310058594, | |
| "learning_rate": 4.718004338394794e-06, | |
| "loss": 0.26, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 14.290909090909091, | |
| "grad_norm": 38.04654312133789, | |
| "learning_rate": 4.609544468546638e-06, | |
| "loss": 0.3813, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 14.533333333333333, | |
| "grad_norm": 34.71643829345703, | |
| "learning_rate": 4.501084598698482e-06, | |
| "loss": 0.0974, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 14.775757575757575, | |
| "grad_norm": 34.031890869140625, | |
| "learning_rate": 4.392624728850326e-06, | |
| "loss": 0.4881, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.0002771662548184395, | |
| "learning_rate": 4.284164859002169e-06, | |
| "loss": 0.4126, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.35451531410217285, | |
| "eval_macro_f1": 0.9235679411519468, | |
| "eval_runtime": 6.0428, | |
| "eval_samples_per_second": 54.611, | |
| "eval_steps_per_second": 6.95, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 15.242424242424242, | |
| "grad_norm": 82.48748779296875, | |
| "learning_rate": 4.175704989154013e-06, | |
| "loss": 0.4444, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 15.484848484848484, | |
| "grad_norm": 0.2618753910064697, | |
| "learning_rate": 4.067245119305857e-06, | |
| "loss": 0.2083, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 15.727272727272727, | |
| "grad_norm": 98.34405517578125, | |
| "learning_rate": 3.958785249457701e-06, | |
| "loss": 0.4785, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 15.969696969696969, | |
| "grad_norm": 0.37142229080200195, | |
| "learning_rate": 3.8503253796095445e-06, | |
| "loss": 0.1583, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.38117873668670654, | |
| "eval_macro_f1": 0.9268860086407444, | |
| "eval_runtime": 6.9311, | |
| "eval_samples_per_second": 47.612, | |
| "eval_steps_per_second": 6.06, | |
| "step": 672 | |
| }, | |
| { | |
| "epoch": 16.193939393939395, | |
| "grad_norm": 40.54330825805664, | |
| "learning_rate": 3.741865509761389e-06, | |
| "loss": 0.0774, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 16.436363636363637, | |
| "grad_norm": 0.23675695061683655, | |
| "learning_rate": 3.6334056399132324e-06, | |
| "loss": 0.1639, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 16.67878787878788, | |
| "grad_norm": 47.12529373168945, | |
| "learning_rate": 3.5249457700650764e-06, | |
| "loss": 0.306, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 16.921212121212122, | |
| "grad_norm": 0.3993530571460724, | |
| "learning_rate": 3.41648590021692e-06, | |
| "loss": 0.2376, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.4087267816066742, | |
| "eval_macro_f1": 0.9295990205081115, | |
| "eval_runtime": 6.1306, | |
| "eval_samples_per_second": 53.828, | |
| "eval_steps_per_second": 6.851, | |
| "step": 714 | |
| }, | |
| { | |
| "epoch": 17.145454545454545, | |
| "grad_norm": 0.34205177426338196, | |
| "learning_rate": 3.308026030368764e-06, | |
| "loss": 0.0332, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 17.387878787878787, | |
| "grad_norm": 0.5112647414207458, | |
| "learning_rate": 3.1995661605206075e-06, | |
| "loss": 0.1332, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 17.63030303030303, | |
| "grad_norm": 120.2950439453125, | |
| "learning_rate": 3.0911062906724515e-06, | |
| "loss": 0.2503, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 17.87272727272727, | |
| "grad_norm": 223.04759216308594, | |
| "learning_rate": 2.982646420824295e-06, | |
| "loss": 0.2703, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.43362897634506226, | |
| "eval_macro_f1": 0.9264924264924266, | |
| "eval_runtime": 5.861, | |
| "eval_samples_per_second": 56.305, | |
| "eval_steps_per_second": 7.166, | |
| "step": 756 | |
| }, | |
| { | |
| "epoch": 18.096969696969698, | |
| "grad_norm": 54.66193771362305, | |
| "learning_rate": 2.874186550976139e-06, | |
| "loss": 0.1274, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 18.33939393939394, | |
| "grad_norm": 54.846466064453125, | |
| "learning_rate": 2.765726681127983e-06, | |
| "loss": 0.2751, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 18.581818181818182, | |
| "grad_norm": 53.97863006591797, | |
| "learning_rate": 2.6572668112798266e-06, | |
| "loss": 0.359, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 18.824242424242424, | |
| "grad_norm": 81.63549041748047, | |
| "learning_rate": 2.5488069414316706e-06, | |
| "loss": 0.1819, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.3480012118816376, | |
| "eval_macro_f1": 0.9236528192931639, | |
| "eval_runtime": 7.0471, | |
| "eval_samples_per_second": 46.828, | |
| "eval_steps_per_second": 5.96, | |
| "step": 798 | |
| }, | |
| { | |
| "epoch": 19.048484848484847, | |
| "grad_norm": 41.54087448120117, | |
| "learning_rate": 2.440347071583514e-06, | |
| "loss": 0.6373, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 19.29090909090909, | |
| "grad_norm": 9.001028060913086, | |
| "learning_rate": 2.331887201735358e-06, | |
| "loss": 0.2971, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 19.533333333333335, | |
| "grad_norm": 114.6279525756836, | |
| "learning_rate": 2.2234273318872017e-06, | |
| "loss": 0.1943, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 19.775757575757577, | |
| "grad_norm": 18.022676467895508, | |
| "learning_rate": 2.1149674620390457e-06, | |
| "loss": 0.1207, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.0001882202341221273, | |
| "learning_rate": 2.0065075921908892e-06, | |
| "loss": 0.1324, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.4493299424648285, | |
| "eval_macro_f1": 0.9384902143522833, | |
| "eval_runtime": 6.0147, | |
| "eval_samples_per_second": 54.865, | |
| "eval_steps_per_second": 6.983, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 20.242424242424242, | |
| "grad_norm": 13.740226745605469, | |
| "learning_rate": 1.8980477223427332e-06, | |
| "loss": 0.294, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 20.484848484848484, | |
| "grad_norm": 6.870513916015625, | |
| "learning_rate": 1.7895878524945772e-06, | |
| "loss": 0.1323, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 20.727272727272727, | |
| "grad_norm": 2.74729585647583, | |
| "learning_rate": 1.681127982646421e-06, | |
| "loss": 0.019, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 20.96969696969697, | |
| "grad_norm": 17.18338966369629, | |
| "learning_rate": 1.572668112798265e-06, | |
| "loss": 0.1312, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 0.40448498725891113, | |
| "eval_macro_f1": 0.9384902143522833, | |
| "eval_runtime": 6.133, | |
| "eval_samples_per_second": 53.808, | |
| "eval_steps_per_second": 6.848, | |
| "step": 882 | |
| }, | |
| { | |
| "epoch": 21.193939393939395, | |
| "grad_norm": 108.9833755493164, | |
| "learning_rate": 1.4642082429501087e-06, | |
| "loss": 0.2499, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 21.436363636363637, | |
| "grad_norm": 0.291847825050354, | |
| "learning_rate": 1.3557483731019525e-06, | |
| "loss": 0.1708, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 21.67878787878788, | |
| "grad_norm": 35.08168029785156, | |
| "learning_rate": 1.2472885032537963e-06, | |
| "loss": 0.0802, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 21.921212121212122, | |
| "grad_norm": 0.05401836335659027, | |
| "learning_rate": 1.13882863340564e-06, | |
| "loss": 0.1662, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.3166828453540802, | |
| "eval_macro_f1": 0.9417692129092176, | |
| "eval_runtime": 6.0442, | |
| "eval_samples_per_second": 54.598, | |
| "eval_steps_per_second": 6.949, | |
| "step": 924 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1025, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 25, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0371596050603966e+19, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |