| { | |
| "best_metric": 0.2552177309989929, | |
| "best_model_checkpoint": "./ryan_model3272024/checkpoint-600", | |
| "epoch": 3.8338658146964857, | |
| "eval_steps": 100, | |
| "global_step": 1200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 1.4023665189743042, | |
| "learning_rate": 0.00019840255591054313, | |
| "loss": 0.5486, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 1.2863692045211792, | |
| "learning_rate": 0.00019680511182108628, | |
| "loss": 0.4543, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.8842328190803528, | |
| "learning_rate": 0.0001952076677316294, | |
| "loss": 0.4222, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.8728455901145935, | |
| "learning_rate": 0.00019361022364217253, | |
| "loss": 0.3764, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.6641435027122498, | |
| "learning_rate": 0.00019201277955271565, | |
| "loss": 0.3214, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 1.4344050884246826, | |
| "learning_rate": 0.0001904153354632588, | |
| "loss": 0.3286, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.8919397592544556, | |
| "learning_rate": 0.00018881789137380192, | |
| "loss": 0.33, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 1.7052876949310303, | |
| "learning_rate": 0.00018722044728434505, | |
| "loss": 0.3337, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.4728272259235382, | |
| "learning_rate": 0.0001856230031948882, | |
| "loss": 0.3784, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 1.1663854122161865, | |
| "learning_rate": 0.00018402555910543132, | |
| "loss": 0.3853, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_loss": 0.3272034823894501, | |
| "eval_na_accuracy": 0.924, | |
| "eval_ordinal_accuracy": 0.52, | |
| "eval_ordinal_mae": 1.210578082634343, | |
| "eval_runtime": 52.9914, | |
| "eval_samples_per_second": 9.435, | |
| "eval_steps_per_second": 1.189, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.8579528331756592, | |
| "learning_rate": 0.00018242811501597444, | |
| "loss": 0.3585, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 1.02351975440979, | |
| "learning_rate": 0.00018083067092651756, | |
| "loss": 0.3621, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 1.3286011219024658, | |
| "learning_rate": 0.00017923322683706071, | |
| "loss": 0.3714, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.6290095448493958, | |
| "learning_rate": 0.00017763578274760384, | |
| "loss": 0.3275, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 1.269338846206665, | |
| "learning_rate": 0.000176038338658147, | |
| "loss": 0.4287, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.6244733333587646, | |
| "learning_rate": 0.0001744408945686901, | |
| "loss": 0.3067, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 1.1287596225738525, | |
| "learning_rate": 0.00017284345047923323, | |
| "loss": 0.2982, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 1.436303734779358, | |
| "learning_rate": 0.00017124600638977638, | |
| "loss": 0.2946, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.8159350752830505, | |
| "learning_rate": 0.00016964856230031948, | |
| "loss": 0.3514, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.7363901138305664, | |
| "learning_rate": 0.00016805111821086263, | |
| "loss": 0.3396, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": 0.27412503957748413, | |
| "eval_na_accuracy": 0.94, | |
| "eval_ordinal_accuracy": 0.5644444444444444, | |
| "eval_ordinal_mae": 1.1640199238227473, | |
| "eval_runtime": 21.3186, | |
| "eval_samples_per_second": 23.454, | |
| "eval_steps_per_second": 2.955, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.6321592330932617, | |
| "learning_rate": 0.00016645367412140575, | |
| "loss": 0.3952, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.6153714656829834, | |
| "learning_rate": 0.0001648562300319489, | |
| "loss": 0.2947, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 1.3031296730041504, | |
| "learning_rate": 0.00016325878594249202, | |
| "loss": 0.3556, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 1.058060646057129, | |
| "learning_rate": 0.00016166134185303515, | |
| "loss": 0.3432, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.957135796546936, | |
| "learning_rate": 0.0001600638977635783, | |
| "loss": 0.3675, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 1.6347941160202026, | |
| "learning_rate": 0.00015846645367412142, | |
| "loss": 0.3008, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 1.1190528869628906, | |
| "learning_rate": 0.00015686900958466454, | |
| "loss": 0.2944, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.8016924858093262, | |
| "learning_rate": 0.00015527156549520767, | |
| "loss": 0.2361, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 1.3622130155563354, | |
| "learning_rate": 0.00015367412140575082, | |
| "loss": 0.3569, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.6603774428367615, | |
| "learning_rate": 0.00015207667731629394, | |
| "loss": 0.2075, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 0.2772314250469208, | |
| "eval_na_accuracy": 0.946, | |
| "eval_ordinal_accuracy": 0.5933333333333334, | |
| "eval_ordinal_mae": 1.194209214001894, | |
| "eval_runtime": 20.7347, | |
| "eval_samples_per_second": 24.114, | |
| "eval_steps_per_second": 3.038, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 1.3968242406845093, | |
| "learning_rate": 0.00015047923322683706, | |
| "loss": 0.2232, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 0.7815521359443665, | |
| "learning_rate": 0.0001488817891373802, | |
| "loss": 0.3132, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 1.1288195848464966, | |
| "learning_rate": 0.00014728434504792333, | |
| "loss": 0.255, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "grad_norm": 0.7704196572303772, | |
| "learning_rate": 0.00014568690095846646, | |
| "loss": 0.2415, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 1.9226877689361572, | |
| "learning_rate": 0.00014408945686900958, | |
| "loss": 0.1975, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 0.5694310069084167, | |
| "learning_rate": 0.00014249201277955273, | |
| "loss": 0.1722, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 1.719147801399231, | |
| "learning_rate": 0.00014089456869009585, | |
| "loss": 0.2175, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 0.9247463941574097, | |
| "learning_rate": 0.000139297124600639, | |
| "loss": 0.2088, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 1.0941154956817627, | |
| "learning_rate": 0.00013769968051118212, | |
| "loss": 0.2854, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 1.0274015665054321, | |
| "learning_rate": 0.00013610223642172525, | |
| "loss": 0.196, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_loss": 0.273777574300766, | |
| "eval_na_accuracy": 0.95, | |
| "eval_ordinal_accuracy": 0.6133333333333333, | |
| "eval_ordinal_mae": 1.198390154937903, | |
| "eval_runtime": 20.9145, | |
| "eval_samples_per_second": 23.907, | |
| "eval_steps_per_second": 3.012, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "grad_norm": 2.912687063217163, | |
| "learning_rate": 0.00013450479233226837, | |
| "loss": 0.2156, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "grad_norm": 0.6906268000602722, | |
| "learning_rate": 0.0001329073482428115, | |
| "loss": 0.1366, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 0.43070048093795776, | |
| "learning_rate": 0.00013130990415335464, | |
| "loss": 0.2174, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "grad_norm": 0.5173763632774353, | |
| "learning_rate": 0.00012971246006389777, | |
| "loss": 0.2016, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.04314386844635, | |
| "learning_rate": 0.00012811501597444092, | |
| "loss": 0.2233, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "grad_norm": 0.523073673248291, | |
| "learning_rate": 0.00012651757188498404, | |
| "loss": 0.2231, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 3.259795904159546, | |
| "learning_rate": 0.00012492012779552716, | |
| "loss": 0.2366, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "grad_norm": 0.6846562027931213, | |
| "learning_rate": 0.00012332268370607028, | |
| "loss": 0.2144, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "grad_norm": 1.2122007608413696, | |
| "learning_rate": 0.00012172523961661342, | |
| "loss": 0.2938, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 1.3790067434310913, | |
| "learning_rate": 0.00012012779552715656, | |
| "loss": 0.2228, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.26852139830589294, | |
| "eval_na_accuracy": 0.956, | |
| "eval_ordinal_accuracy": 0.62, | |
| "eval_ordinal_mae": 1.1989026491012837, | |
| "eval_runtime": 20.0158, | |
| "eval_samples_per_second": 24.98, | |
| "eval_steps_per_second": 3.148, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "grad_norm": 0.7108421921730042, | |
| "learning_rate": 0.00011853035143769968, | |
| "loss": 0.1916, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "grad_norm": 0.42910462617874146, | |
| "learning_rate": 0.00011693290734824283, | |
| "loss": 0.2478, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 0.9730465412139893, | |
| "learning_rate": 0.00011533546325878595, | |
| "loss": 0.189, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 0.9566612243652344, | |
| "learning_rate": 0.00011373801916932908, | |
| "loss": 0.1768, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.5167070627212524, | |
| "learning_rate": 0.00011214057507987221, | |
| "loss": 0.1385, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 0.5880122780799866, | |
| "learning_rate": 0.00011054313099041533, | |
| "loss": 0.1262, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 1.202286720275879, | |
| "learning_rate": 0.00010894568690095847, | |
| "loss": 0.1721, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 2.6997601985931396, | |
| "learning_rate": 0.0001073482428115016, | |
| "loss": 0.2128, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 1.1591830253601074, | |
| "learning_rate": 0.00010575079872204474, | |
| "loss": 0.2402, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.5840221643447876, | |
| "learning_rate": 0.00010415335463258787, | |
| "loss": 0.1816, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_loss": 0.2552177309989929, | |
| "eval_na_accuracy": 0.95, | |
| "eval_ordinal_accuracy": 0.6266666666666667, | |
| "eval_ordinal_mae": 1.158560517811113, | |
| "eval_runtime": 19.5011, | |
| "eval_samples_per_second": 25.64, | |
| "eval_steps_per_second": 3.231, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 0.7560299634933472, | |
| "learning_rate": 0.000102555910543131, | |
| "loss": 0.2021, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 1.8860361576080322, | |
| "learning_rate": 0.00010095846645367413, | |
| "loss": 0.2092, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 0.7235255837440491, | |
| "learning_rate": 9.936102236421726e-05, | |
| "loss": 0.1131, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 0.3656529486179352, | |
| "learning_rate": 9.77635782747604e-05, | |
| "loss": 0.0867, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.3450271785259247, | |
| "learning_rate": 9.616613418530351e-05, | |
| "loss": 0.0903, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 1.0603750944137573, | |
| "learning_rate": 9.456869009584664e-05, | |
| "loss": 0.1234, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "grad_norm": 0.6790297031402588, | |
| "learning_rate": 9.297124600638978e-05, | |
| "loss": 0.0936, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 0.5596363544464111, | |
| "learning_rate": 9.137380191693292e-05, | |
| "loss": 0.0651, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 0.5989049673080444, | |
| "learning_rate": 8.977635782747604e-05, | |
| "loss": 0.1218, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 0.9003208875656128, | |
| "learning_rate": 8.817891373801918e-05, | |
| "loss": 0.0682, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_loss": 0.27212005853652954, | |
| "eval_na_accuracy": 0.952, | |
| "eval_ordinal_accuracy": 0.6577777777777778, | |
| "eval_ordinal_mae": 1.1557789803379113, | |
| "eval_runtime": 19.5966, | |
| "eval_samples_per_second": 25.515, | |
| "eval_steps_per_second": 3.215, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "grad_norm": 0.6663013100624084, | |
| "learning_rate": 8.658146964856231e-05, | |
| "loss": 0.0714, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 1.0458776950836182, | |
| "learning_rate": 8.498402555910544e-05, | |
| "loss": 0.102, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "grad_norm": 0.9246501922607422, | |
| "learning_rate": 8.338658146964856e-05, | |
| "loss": 0.1623, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "grad_norm": 1.0837684869766235, | |
| "learning_rate": 8.17891373801917e-05, | |
| "loss": 0.0934, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.564241349697113, | |
| "learning_rate": 8.019169329073483e-05, | |
| "loss": 0.0853, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "grad_norm": 4.335838794708252, | |
| "learning_rate": 7.859424920127795e-05, | |
| "loss": 0.1246, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 0.957082211971283, | |
| "learning_rate": 7.699680511182109e-05, | |
| "loss": 0.1292, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "grad_norm": 0.9633702039718628, | |
| "learning_rate": 7.539936102236423e-05, | |
| "loss": 0.1916, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "grad_norm": 0.7254676222801208, | |
| "learning_rate": 7.380191693290735e-05, | |
| "loss": 0.1054, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 0.5885197520256042, | |
| "learning_rate": 7.220447284345049e-05, | |
| "loss": 0.0795, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_loss": 0.2753521502017975, | |
| "eval_na_accuracy": 0.948, | |
| "eval_ordinal_accuracy": 0.6333333333333333, | |
| "eval_ordinal_mae": 1.1599188842872779, | |
| "eval_runtime": 20.0506, | |
| "eval_samples_per_second": 24.937, | |
| "eval_steps_per_second": 3.142, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "grad_norm": 0.5671622157096863, | |
| "learning_rate": 7.060702875399361e-05, | |
| "loss": 0.0948, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "grad_norm": 0.9914100766181946, | |
| "learning_rate": 6.900958466453674e-05, | |
| "loss": 0.0715, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 0.4819205105304718, | |
| "learning_rate": 6.741214057507987e-05, | |
| "loss": 0.0839, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 0.3811684250831604, | |
| "learning_rate": 6.5814696485623e-05, | |
| "loss": 0.0825, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "grad_norm": 0.9750994443893433, | |
| "learning_rate": 6.421725239616614e-05, | |
| "loss": 0.0968, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 0.35765138268470764, | |
| "learning_rate": 6.261980830670928e-05, | |
| "loss": 0.1605, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "grad_norm": 0.3497343361377716, | |
| "learning_rate": 6.1022364217252406e-05, | |
| "loss": 0.0933, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "grad_norm": 0.4838835299015045, | |
| "learning_rate": 5.942492012779552e-05, | |
| "loss": 0.0859, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "grad_norm": 0.7002846002578735, | |
| "learning_rate": 5.782747603833866e-05, | |
| "loss": 0.1021, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 2.312203884124756, | |
| "learning_rate": 5.623003194888179e-05, | |
| "loss": 0.1367, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "eval_loss": 0.29526129364967346, | |
| "eval_na_accuracy": 0.946, | |
| "eval_ordinal_accuracy": 0.64, | |
| "eval_ordinal_mae": 1.166716830432415, | |
| "eval_runtime": 20.0091, | |
| "eval_samples_per_second": 24.989, | |
| "eval_steps_per_second": 3.149, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "grad_norm": 0.44126951694488525, | |
| "learning_rate": 5.4632587859424925e-05, | |
| "loss": 0.0854, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "grad_norm": 1.0075191259384155, | |
| "learning_rate": 5.3035143769968054e-05, | |
| "loss": 0.0823, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "grad_norm": 0.9991279244422913, | |
| "learning_rate": 5.1437699680511184e-05, | |
| "loss": 0.1156, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.8888081312179565, | |
| "learning_rate": 4.984025559105431e-05, | |
| "loss": 0.0876, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 0.3761376738548279, | |
| "learning_rate": 4.824281150159744e-05, | |
| "loss": 0.0452, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "grad_norm": 0.365622341632843, | |
| "learning_rate": 4.664536741214058e-05, | |
| "loss": 0.0428, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 0.35657036304473877, | |
| "learning_rate": 4.504792332268371e-05, | |
| "loss": 0.033, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "grad_norm": 0.5636401176452637, | |
| "learning_rate": 4.345047923322684e-05, | |
| "loss": 0.0356, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "grad_norm": 0.431383341550827, | |
| "learning_rate": 4.185303514376997e-05, | |
| "loss": 0.0463, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "grad_norm": 0.583328127861023, | |
| "learning_rate": 4.0255591054313104e-05, | |
| "loss": 0.0387, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "eval_loss": 0.2923290431499481, | |
| "eval_na_accuracy": 0.944, | |
| "eval_ordinal_accuracy": 0.6377777777777778, | |
| "eval_ordinal_mae": 1.2024743282463815, | |
| "eval_runtime": 19.3226, | |
| "eval_samples_per_second": 25.876, | |
| "eval_steps_per_second": 3.26, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "grad_norm": 2.440162420272827, | |
| "learning_rate": 3.8658146964856234e-05, | |
| "loss": 0.0607, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "grad_norm": 0.29546236991882324, | |
| "learning_rate": 3.7060702875399364e-05, | |
| "loss": 0.0515, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "grad_norm": 0.44689303636550903, | |
| "learning_rate": 3.546325878594249e-05, | |
| "loss": 0.0273, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "grad_norm": 0.3288978040218353, | |
| "learning_rate": 3.386581469648562e-05, | |
| "loss": 0.0352, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "grad_norm": 0.41706767678260803, | |
| "learning_rate": 3.226837060702875e-05, | |
| "loss": 0.0345, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "grad_norm": 0.31060507893562317, | |
| "learning_rate": 3.067092651757188e-05, | |
| "loss": 0.0294, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "grad_norm": 0.2541821599006653, | |
| "learning_rate": 2.907348242811502e-05, | |
| "loss": 0.0354, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "grad_norm": 0.574343740940094, | |
| "learning_rate": 2.747603833865815e-05, | |
| "loss": 0.0443, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "grad_norm": 0.47532182931900024, | |
| "learning_rate": 2.5878594249201278e-05, | |
| "loss": 0.0605, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "grad_norm": 0.45276594161987305, | |
| "learning_rate": 2.428115015974441e-05, | |
| "loss": 0.0293, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "eval_loss": 0.2884800434112549, | |
| "eval_na_accuracy": 0.948, | |
| "eval_ordinal_accuracy": 0.6644444444444444, | |
| "eval_ordinal_mae": 1.1666180535654227, | |
| "eval_runtime": 19.9365, | |
| "eval_samples_per_second": 25.08, | |
| "eval_steps_per_second": 3.16, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "grad_norm": 0.655549168586731, | |
| "learning_rate": 2.268370607028754e-05, | |
| "loss": 0.034, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "grad_norm": 0.16610193252563477, | |
| "learning_rate": 2.108626198083067e-05, | |
| "loss": 0.0319, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "grad_norm": 0.26889652013778687, | |
| "learning_rate": 1.9488817891373803e-05, | |
| "loss": 0.0479, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "grad_norm": 0.2418793886899948, | |
| "learning_rate": 1.7891373801916932e-05, | |
| "loss": 0.0322, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "grad_norm": 0.5379694104194641, | |
| "learning_rate": 1.6293929712460065e-05, | |
| "loss": 0.0393, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "grad_norm": 0.19815516471862793, | |
| "learning_rate": 1.4696485623003195e-05, | |
| "loss": 0.0217, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "grad_norm": 0.889312207698822, | |
| "learning_rate": 1.3099041533546328e-05, | |
| "loss": 0.0332, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "grad_norm": 0.2865816652774811, | |
| "learning_rate": 1.1501597444089457e-05, | |
| "loss": 0.0313, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "grad_norm": 0.5947129726409912, | |
| "learning_rate": 9.904153354632589e-06, | |
| "loss": 0.034, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "grad_norm": 0.44885268807411194, | |
| "learning_rate": 8.306709265175718e-06, | |
| "loss": 0.0286, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "eval_loss": 0.28681233525276184, | |
| "eval_na_accuracy": 0.95, | |
| "eval_ordinal_accuracy": 0.6711111111111111, | |
| "eval_ordinal_mae": 1.1625636271304554, | |
| "eval_runtime": 19.7259, | |
| "eval_samples_per_second": 25.347, | |
| "eval_steps_per_second": 3.194, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "step": 1200, | |
| "total_flos": 1.4860396665534874e+18, | |
| "train_loss": 0.17935538868109385, | |
| "train_runtime": 1702.5744, | |
| "train_samples_per_second": 11.747, | |
| "train_steps_per_second": 0.735 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1252, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "total_flos": 1.4860396665534874e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |