|
{ |
|
"best_metric": 1.0327022075653076, |
|
"best_model_checkpoint": "./ryan_model3272024/checkpoint-500", |
|
"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": 2.4977924823760986, |
|
"learning_rate": 0.00019840255591054313, |
|
"loss": 1.9988, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"grad_norm": 2.3356950283050537, |
|
"learning_rate": 0.00019680511182108628, |
|
"loss": 1.6981, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"grad_norm": 2.1436116695404053, |
|
"learning_rate": 0.0001952076677316294, |
|
"loss": 1.6585, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.13, |
|
"grad_norm": 1.9430339336395264, |
|
"learning_rate": 0.00019361022364217253, |
|
"loss": 1.6929, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"grad_norm": 1.7540597915649414, |
|
"learning_rate": 0.00019201277955271565, |
|
"loss": 1.4822, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"grad_norm": 2.9073445796966553, |
|
"learning_rate": 0.0001904153354632588, |
|
"loss": 1.5968, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"grad_norm": 1.6934313774108887, |
|
"learning_rate": 0.00018881789137380192, |
|
"loss": 1.4684, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"grad_norm": 3.451613426208496, |
|
"learning_rate": 0.00018722044728434505, |
|
"loss": 1.4425, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"grad_norm": 1.747863531112671, |
|
"learning_rate": 0.0001856230031948882, |
|
"loss": 1.3641, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"grad_norm": 2.263026714324951, |
|
"learning_rate": 0.00018402555910543132, |
|
"loss": 1.3578, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"eval_loss": 1.2057933807373047, |
|
"eval_na_accuracy": 0.944, |
|
"eval_ordinal_accuracy": 0.54, |
|
"eval_ordinal_mae": 90.9845891143661, |
|
"eval_runtime": 46.445, |
|
"eval_samples_per_second": 10.765, |
|
"eval_steps_per_second": 1.356, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"grad_norm": 2.7085518836975098, |
|
"learning_rate": 0.00018242811501597444, |
|
"loss": 1.3463, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"grad_norm": 2.2408878803253174, |
|
"learning_rate": 0.00018083067092651756, |
|
"loss": 1.1704, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"grad_norm": 2.1588001251220703, |
|
"learning_rate": 0.00017923322683706071, |
|
"loss": 1.24, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"grad_norm": 2.582127094268799, |
|
"learning_rate": 0.00017763578274760384, |
|
"loss": 1.2735, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"grad_norm": 1.9934537410736084, |
|
"learning_rate": 0.000176038338658147, |
|
"loss": 1.344, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"grad_norm": 2.1636180877685547, |
|
"learning_rate": 0.0001744408945686901, |
|
"loss": 1.1532, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"grad_norm": 3.9455337524414062, |
|
"learning_rate": 0.00017284345047923323, |
|
"loss": 1.1231, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"grad_norm": 3.46232271194458, |
|
"learning_rate": 0.00017124600638977638, |
|
"loss": 1.0945, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.61, |
|
"grad_norm": 2.913996696472168, |
|
"learning_rate": 0.00016964856230031948, |
|
"loss": 1.3168, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"grad_norm": 2.16892671585083, |
|
"learning_rate": 0.00016805111821086263, |
|
"loss": 1.089, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"eval_loss": 1.0986785888671875, |
|
"eval_na_accuracy": 0.95, |
|
"eval_ordinal_accuracy": 0.548, |
|
"eval_ordinal_mae": 120.70968111835792, |
|
"eval_runtime": 20.2841, |
|
"eval_samples_per_second": 24.65, |
|
"eval_steps_per_second": 3.106, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"grad_norm": 2.9793784618377686, |
|
"learning_rate": 0.00016645367412140575, |
|
"loss": 1.1788, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"grad_norm": 2.662822723388672, |
|
"learning_rate": 0.0001648562300319489, |
|
"loss": 1.0067, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 0.73, |
|
"grad_norm": 2.3921396732330322, |
|
"learning_rate": 0.00016325878594249202, |
|
"loss": 1.2542, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 0.77, |
|
"grad_norm": 2.670137882232666, |
|
"learning_rate": 0.00016166134185303515, |
|
"loss": 1.1284, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"grad_norm": 3.212069511413574, |
|
"learning_rate": 0.0001600638977635783, |
|
"loss": 1.3407, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.83, |
|
"grad_norm": 2.284740447998047, |
|
"learning_rate": 0.00015846645367412142, |
|
"loss": 1.1876, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"grad_norm": 2.98988938331604, |
|
"learning_rate": 0.00015686900958466454, |
|
"loss": 1.0993, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"grad_norm": 3.687721014022827, |
|
"learning_rate": 0.00015527156549520767, |
|
"loss": 1.1273, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"grad_norm": 2.0068211555480957, |
|
"learning_rate": 0.00015367412140575082, |
|
"loss": 1.0204, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"grad_norm": 2.719209909439087, |
|
"learning_rate": 0.00015207667731629394, |
|
"loss": 0.924, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"eval_loss": 1.083807110786438, |
|
"eval_na_accuracy": 0.946, |
|
"eval_ordinal_accuracy": 0.568, |
|
"eval_ordinal_mae": 76.79816162186209, |
|
"eval_runtime": 20.3196, |
|
"eval_samples_per_second": 24.607, |
|
"eval_steps_per_second": 3.1, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.99, |
|
"grad_norm": 6.174163341522217, |
|
"learning_rate": 0.00015063897763578277, |
|
"loss": 1.0079, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"grad_norm": 1.8171716928482056, |
|
"learning_rate": 0.0001490415335463259, |
|
"loss": 0.9573, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 1.05, |
|
"grad_norm": 1.7882392406463623, |
|
"learning_rate": 0.00014744408945686902, |
|
"loss": 0.8008, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 1.09, |
|
"grad_norm": 2.8711063861846924, |
|
"learning_rate": 0.00014584664536741214, |
|
"loss": 0.7867, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 1.12, |
|
"grad_norm": 3.8461153507232666, |
|
"learning_rate": 0.00014424920127795526, |
|
"loss": 0.7375, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 1.15, |
|
"grad_norm": 3.599330186843872, |
|
"learning_rate": 0.0001426517571884984, |
|
"loss": 0.8057, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 1.18, |
|
"grad_norm": 1.9864487648010254, |
|
"learning_rate": 0.00014105431309904153, |
|
"loss": 0.7876, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 1.21, |
|
"grad_norm": 3.377739191055298, |
|
"learning_rate": 0.00013945686900958468, |
|
"loss": 0.7835, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 1.25, |
|
"grad_norm": 3.5138072967529297, |
|
"learning_rate": 0.0001378594249201278, |
|
"loss": 0.8265, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 1.28, |
|
"grad_norm": 3.5262341499328613, |
|
"learning_rate": 0.00013626198083067093, |
|
"loss": 0.694, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 1.28, |
|
"eval_loss": 1.068002700805664, |
|
"eval_na_accuracy": 0.942, |
|
"eval_ordinal_accuracy": 0.556, |
|
"eval_ordinal_mae": 105.63124973691814, |
|
"eval_runtime": 19.9749, |
|
"eval_samples_per_second": 25.031, |
|
"eval_steps_per_second": 3.154, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 1.31, |
|
"grad_norm": 2.814100980758667, |
|
"learning_rate": 0.00013466453674121405, |
|
"loss": 0.7246, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 1.34, |
|
"grad_norm": 1.9095064401626587, |
|
"learning_rate": 0.00013306709265175718, |
|
"loss": 0.768, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 1.37, |
|
"grad_norm": 2.849485158920288, |
|
"learning_rate": 0.00013146964856230033, |
|
"loss": 0.6669, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 1.41, |
|
"grad_norm": 2.2935235500335693, |
|
"learning_rate": 0.00012987220447284345, |
|
"loss": 0.6708, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 1.44, |
|
"grad_norm": 2.6063146591186523, |
|
"learning_rate": 0.0001282747603833866, |
|
"loss": 0.7481, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 1.47, |
|
"grad_norm": 2.329845428466797, |
|
"learning_rate": 0.00012667731629392972, |
|
"loss": 0.7341, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 1.5, |
|
"grad_norm": 4.184568881988525, |
|
"learning_rate": 0.00012507987220447287, |
|
"loss": 0.785, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 1.53, |
|
"grad_norm": 2.142719268798828, |
|
"learning_rate": 0.00012348242811501597, |
|
"loss": 0.702, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 1.57, |
|
"grad_norm": 2.7038168907165527, |
|
"learning_rate": 0.0001218849840255591, |
|
"loss": 0.8301, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"grad_norm": 3.093721389770508, |
|
"learning_rate": 0.00012028753993610224, |
|
"loss": 0.7739, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"eval_loss": 1.0327022075653076, |
|
"eval_na_accuracy": 0.956, |
|
"eval_ordinal_accuracy": 0.568, |
|
"eval_ordinal_mae": 56.34840202213265, |
|
"eval_runtime": 20.2051, |
|
"eval_samples_per_second": 24.746, |
|
"eval_steps_per_second": 3.118, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 1.63, |
|
"grad_norm": 3.727419853210449, |
|
"learning_rate": 0.00011869009584664536, |
|
"loss": 0.7536, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 1.66, |
|
"grad_norm": 3.3148770332336426, |
|
"learning_rate": 0.00011709265175718851, |
|
"loss": 0.7675, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 1.69, |
|
"grad_norm": 3.7952725887298584, |
|
"learning_rate": 0.00011549520766773163, |
|
"loss": 0.7543, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 1.73, |
|
"grad_norm": 3.9252896308898926, |
|
"learning_rate": 0.00011389776357827477, |
|
"loss": 0.7754, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 1.76, |
|
"grad_norm": 3.045405387878418, |
|
"learning_rate": 0.0001123003194888179, |
|
"loss": 0.5713, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 1.79, |
|
"grad_norm": 1.8544007539749146, |
|
"learning_rate": 0.00011070287539936102, |
|
"loss": 0.5592, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 1.82, |
|
"grad_norm": 3.626079797744751, |
|
"learning_rate": 0.00010910543130990417, |
|
"loss": 0.6631, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 1.85, |
|
"grad_norm": 4.355627536773682, |
|
"learning_rate": 0.00010750798722044728, |
|
"loss": 0.5981, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 1.88, |
|
"grad_norm": 4.025274276733398, |
|
"learning_rate": 0.00010591054313099043, |
|
"loss": 0.7375, |
|
"step": 590 |
|
}, |
|
{ |
|
"epoch": 1.92, |
|
"grad_norm": 2.506023645401001, |
|
"learning_rate": 0.00010431309904153355, |
|
"loss": 0.5935, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 1.92, |
|
"eval_loss": 1.0479341745376587, |
|
"eval_na_accuracy": 0.932, |
|
"eval_ordinal_accuracy": 0.598, |
|
"eval_ordinal_mae": 50.45202396943141, |
|
"eval_runtime": 20.1021, |
|
"eval_samples_per_second": 24.873, |
|
"eval_steps_per_second": 3.134, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 1.95, |
|
"grad_norm": 4.494864463806152, |
|
"learning_rate": 0.00010271565495207669, |
|
"loss": 0.651, |
|
"step": 610 |
|
}, |
|
{ |
|
"epoch": 1.98, |
|
"grad_norm": 4.595503807067871, |
|
"learning_rate": 0.00010111821086261981, |
|
"loss": 0.6743, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 2.01, |
|
"grad_norm": 3.0126616954803467, |
|
"learning_rate": 9.952076677316294e-05, |
|
"loss": 0.5441, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 2.04, |
|
"grad_norm": 2.2599124908447266, |
|
"learning_rate": 9.792332268370608e-05, |
|
"loss": 0.3816, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 2.08, |
|
"grad_norm": 1.8868279457092285, |
|
"learning_rate": 9.63258785942492e-05, |
|
"loss": 0.4175, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 2.11, |
|
"grad_norm": 2.2594425678253174, |
|
"learning_rate": 9.472843450479234e-05, |
|
"loss": 0.3208, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 2.14, |
|
"grad_norm": 1.4436126947402954, |
|
"learning_rate": 9.313099041533548e-05, |
|
"loss": 0.2459, |
|
"step": 670 |
|
}, |
|
{ |
|
"epoch": 2.17, |
|
"grad_norm": 2.018686532974243, |
|
"learning_rate": 9.15335463258786e-05, |
|
"loss": 0.2556, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 2.2, |
|
"grad_norm": 3.8970091342926025, |
|
"learning_rate": 8.993610223642172e-05, |
|
"loss": 0.2912, |
|
"step": 690 |
|
}, |
|
{ |
|
"epoch": 2.24, |
|
"grad_norm": 7.18654727935791, |
|
"learning_rate": 8.833865814696486e-05, |
|
"loss": 0.3525, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 2.24, |
|
"eval_loss": 1.1914987564086914, |
|
"eval_na_accuracy": 0.94, |
|
"eval_ordinal_accuracy": 0.578, |
|
"eval_ordinal_mae": 68.50985479798727, |
|
"eval_runtime": 20.035, |
|
"eval_samples_per_second": 24.956, |
|
"eval_steps_per_second": 3.145, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 2.27, |
|
"grad_norm": 3.8287854194641113, |
|
"learning_rate": 8.6741214057508e-05, |
|
"loss": 0.3517, |
|
"step": 710 |
|
}, |
|
{ |
|
"epoch": 2.3, |
|
"grad_norm": 4.602190971374512, |
|
"learning_rate": 8.514376996805112e-05, |
|
"loss": 0.3595, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 2.33, |
|
"grad_norm": 4.8052978515625, |
|
"learning_rate": 8.354632587859425e-05, |
|
"loss": 0.3686, |
|
"step": 730 |
|
}, |
|
{ |
|
"epoch": 2.36, |
|
"grad_norm": 1.684538722038269, |
|
"learning_rate": 8.194888178913739e-05, |
|
"loss": 0.3499, |
|
"step": 740 |
|
}, |
|
{ |
|
"epoch": 2.4, |
|
"grad_norm": 2.9729549884796143, |
|
"learning_rate": 8.035143769968051e-05, |
|
"loss": 0.3116, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 2.43, |
|
"grad_norm": 4.066229343414307, |
|
"learning_rate": 7.875399361022364e-05, |
|
"loss": 0.301, |
|
"step": 760 |
|
}, |
|
{ |
|
"epoch": 2.46, |
|
"grad_norm": 4.297142505645752, |
|
"learning_rate": 7.715654952076677e-05, |
|
"loss": 0.3501, |
|
"step": 770 |
|
}, |
|
{ |
|
"epoch": 2.49, |
|
"grad_norm": 3.189189910888672, |
|
"learning_rate": 7.555910543130991e-05, |
|
"loss": 0.4095, |
|
"step": 780 |
|
}, |
|
{ |
|
"epoch": 2.52, |
|
"grad_norm": 4.354482173919678, |
|
"learning_rate": 7.396166134185304e-05, |
|
"loss": 0.2547, |
|
"step": 790 |
|
}, |
|
{ |
|
"epoch": 2.56, |
|
"grad_norm": 2.2913010120391846, |
|
"learning_rate": 7.236421725239617e-05, |
|
"loss": 0.2385, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 2.56, |
|
"eval_loss": 1.1303021907806396, |
|
"eval_na_accuracy": 0.948, |
|
"eval_ordinal_accuracy": 0.586, |
|
"eval_ordinal_mae": 43.02211729006702, |
|
"eval_runtime": 20.2963, |
|
"eval_samples_per_second": 24.635, |
|
"eval_steps_per_second": 3.104, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 2.59, |
|
"grad_norm": 4.907588481903076, |
|
"learning_rate": 7.07667731629393e-05, |
|
"loss": 0.2578, |
|
"step": 810 |
|
}, |
|
{ |
|
"epoch": 2.62, |
|
"grad_norm": 5.770906925201416, |
|
"learning_rate": 6.916932907348244e-05, |
|
"loss": 0.3875, |
|
"step": 820 |
|
}, |
|
{ |
|
"epoch": 2.65, |
|
"grad_norm": 3.4392714500427246, |
|
"learning_rate": 6.757188498402556e-05, |
|
"loss": 0.3435, |
|
"step": 830 |
|
}, |
|
{ |
|
"epoch": 2.68, |
|
"grad_norm": 1.785447359085083, |
|
"learning_rate": 6.597444089456869e-05, |
|
"loss": 0.2681, |
|
"step": 840 |
|
}, |
|
{ |
|
"epoch": 2.72, |
|
"grad_norm": 3.527557373046875, |
|
"learning_rate": 6.437699680511182e-05, |
|
"loss": 0.3525, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 2.75, |
|
"grad_norm": 0.4556908905506134, |
|
"learning_rate": 6.277955271565496e-05, |
|
"loss": 0.2462, |
|
"step": 860 |
|
}, |
|
{ |
|
"epoch": 2.78, |
|
"grad_norm": 2.7756271362304688, |
|
"learning_rate": 6.118210862619808e-05, |
|
"loss": 0.3107, |
|
"step": 870 |
|
}, |
|
{ |
|
"epoch": 2.81, |
|
"grad_norm": 4.53245735168457, |
|
"learning_rate": 5.958466453674122e-05, |
|
"loss": 0.319, |
|
"step": 880 |
|
}, |
|
{ |
|
"epoch": 2.84, |
|
"grad_norm": 1.7556302547454834, |
|
"learning_rate": 5.7987220447284354e-05, |
|
"loss": 0.2754, |
|
"step": 890 |
|
}, |
|
{ |
|
"epoch": 2.88, |
|
"grad_norm": 4.295087814331055, |
|
"learning_rate": 5.6389776357827484e-05, |
|
"loss": 0.3423, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 2.88, |
|
"eval_loss": 1.1767209768295288, |
|
"eval_na_accuracy": 0.94, |
|
"eval_ordinal_accuracy": 0.604, |
|
"eval_ordinal_mae": 72.14369884862379, |
|
"eval_runtime": 19.1966, |
|
"eval_samples_per_second": 26.046, |
|
"eval_steps_per_second": 3.282, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 2.91, |
|
"grad_norm": 3.532813549041748, |
|
"learning_rate": 5.479233226837061e-05, |
|
"loss": 0.2459, |
|
"step": 910 |
|
}, |
|
{ |
|
"epoch": 2.94, |
|
"grad_norm": 2.6231563091278076, |
|
"learning_rate": 5.3194888178913736e-05, |
|
"loss": 0.2713, |
|
"step": 920 |
|
}, |
|
{ |
|
"epoch": 2.97, |
|
"grad_norm": 1.7921823263168335, |
|
"learning_rate": 5.159744408945687e-05, |
|
"loss": 0.1915, |
|
"step": 930 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"grad_norm": 0.42470985651016235, |
|
"learning_rate": 5e-05, |
|
"loss": 0.2758, |
|
"step": 940 |
|
}, |
|
{ |
|
"epoch": 3.04, |
|
"grad_norm": 2.1311330795288086, |
|
"learning_rate": 4.840255591054313e-05, |
|
"loss": 0.1447, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 3.07, |
|
"grad_norm": 2.371814489364624, |
|
"learning_rate": 4.680511182108626e-05, |
|
"loss": 0.0985, |
|
"step": 960 |
|
}, |
|
{ |
|
"epoch": 3.1, |
|
"grad_norm": 0.3315708637237549, |
|
"learning_rate": 4.520766773162939e-05, |
|
"loss": 0.0887, |
|
"step": 970 |
|
}, |
|
{ |
|
"epoch": 3.13, |
|
"grad_norm": 5.331906795501709, |
|
"learning_rate": 4.361022364217253e-05, |
|
"loss": 0.1125, |
|
"step": 980 |
|
}, |
|
{ |
|
"epoch": 3.16, |
|
"grad_norm": 1.143639326095581, |
|
"learning_rate": 4.201277955271566e-05, |
|
"loss": 0.1029, |
|
"step": 990 |
|
}, |
|
{ |
|
"epoch": 3.19, |
|
"grad_norm": 0.28226566314697266, |
|
"learning_rate": 4.041533546325879e-05, |
|
"loss": 0.0674, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 3.19, |
|
"eval_loss": 1.2294036149978638, |
|
"eval_na_accuracy": 0.938, |
|
"eval_ordinal_accuracy": 0.606, |
|
"eval_ordinal_mae": 28.07016432112176, |
|
"eval_runtime": 20.2957, |
|
"eval_samples_per_second": 24.636, |
|
"eval_steps_per_second": 3.104, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 3.23, |
|
"grad_norm": 0.1891467422246933, |
|
"learning_rate": 3.8817891373801916e-05, |
|
"loss": 0.0911, |
|
"step": 1010 |
|
}, |
|
{ |
|
"epoch": 3.26, |
|
"grad_norm": 0.14374223351478577, |
|
"learning_rate": 3.722044728434505e-05, |
|
"loss": 0.1007, |
|
"step": 1020 |
|
}, |
|
{ |
|
"epoch": 3.29, |
|
"grad_norm": 0.22984538972377777, |
|
"learning_rate": 3.562300319488818e-05, |
|
"loss": 0.0686, |
|
"step": 1030 |
|
}, |
|
{ |
|
"epoch": 3.32, |
|
"grad_norm": 6.988112926483154, |
|
"learning_rate": 3.402555910543131e-05, |
|
"loss": 0.0882, |
|
"step": 1040 |
|
}, |
|
{ |
|
"epoch": 3.35, |
|
"grad_norm": 0.12949563562870026, |
|
"learning_rate": 3.242811501597444e-05, |
|
"loss": 0.1098, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 3.39, |
|
"grad_norm": 0.40354716777801514, |
|
"learning_rate": 3.083067092651757e-05, |
|
"loss": 0.0809, |
|
"step": 1060 |
|
}, |
|
{ |
|
"epoch": 3.42, |
|
"grad_norm": 0.6469861268997192, |
|
"learning_rate": 2.9233226837060707e-05, |
|
"loss": 0.0998, |
|
"step": 1070 |
|
}, |
|
{ |
|
"epoch": 3.45, |
|
"grad_norm": 2.1838839054107666, |
|
"learning_rate": 2.7635782747603834e-05, |
|
"loss": 0.0923, |
|
"step": 1080 |
|
}, |
|
{ |
|
"epoch": 3.48, |
|
"grad_norm": 1.0453554391860962, |
|
"learning_rate": 2.6038338658146967e-05, |
|
"loss": 0.0669, |
|
"step": 1090 |
|
}, |
|
{ |
|
"epoch": 3.51, |
|
"grad_norm": 0.6662182807922363, |
|
"learning_rate": 2.44408945686901e-05, |
|
"loss": 0.1206, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 3.51, |
|
"eval_loss": 1.2335716485977173, |
|
"eval_na_accuracy": 0.938, |
|
"eval_ordinal_accuracy": 0.616, |
|
"eval_ordinal_mae": 65.07940023336745, |
|
"eval_runtime": 20.0687, |
|
"eval_samples_per_second": 24.914, |
|
"eval_steps_per_second": 3.139, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 3.55, |
|
"grad_norm": 3.52840256690979, |
|
"learning_rate": 2.284345047923323e-05, |
|
"loss": 0.1402, |
|
"step": 1110 |
|
}, |
|
{ |
|
"epoch": 3.58, |
|
"grad_norm": 0.1548507809638977, |
|
"learning_rate": 2.124600638977636e-05, |
|
"loss": 0.1038, |
|
"step": 1120 |
|
}, |
|
{ |
|
"epoch": 3.61, |
|
"grad_norm": 3.753204584121704, |
|
"learning_rate": 1.964856230031949e-05, |
|
"loss": 0.1237, |
|
"step": 1130 |
|
}, |
|
{ |
|
"epoch": 3.64, |
|
"grad_norm": 2.5928821563720703, |
|
"learning_rate": 1.805111821086262e-05, |
|
"loss": 0.1653, |
|
"step": 1140 |
|
}, |
|
{ |
|
"epoch": 3.67, |
|
"grad_norm": 0.26777195930480957, |
|
"learning_rate": 1.645367412140575e-05, |
|
"loss": 0.0924, |
|
"step": 1150 |
|
}, |
|
{ |
|
"epoch": 3.71, |
|
"grad_norm": 0.11366520822048187, |
|
"learning_rate": 1.485623003194888e-05, |
|
"loss": 0.0376, |
|
"step": 1160 |
|
}, |
|
{ |
|
"epoch": 3.74, |
|
"grad_norm": 2.3529770374298096, |
|
"learning_rate": 1.3258785942492014e-05, |
|
"loss": 0.1034, |
|
"step": 1170 |
|
}, |
|
{ |
|
"epoch": 3.77, |
|
"grad_norm": 0.16830460727214813, |
|
"learning_rate": 1.1661341853035145e-05, |
|
"loss": 0.107, |
|
"step": 1180 |
|
}, |
|
{ |
|
"epoch": 3.8, |
|
"grad_norm": 0.9607967734336853, |
|
"learning_rate": 1.0063897763578276e-05, |
|
"loss": 0.0751, |
|
"step": 1190 |
|
}, |
|
{ |
|
"epoch": 3.83, |
|
"grad_norm": 7.122812271118164, |
|
"learning_rate": 8.466453674121406e-06, |
|
"loss": 0.1261, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 3.83, |
|
"eval_loss": 1.2907226085662842, |
|
"eval_na_accuracy": 0.938, |
|
"eval_ordinal_accuracy": 0.604, |
|
"eval_ordinal_mae": 45.83343084012577, |
|
"eval_runtime": 20.016, |
|
"eval_samples_per_second": 24.98, |
|
"eval_steps_per_second": 3.147, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 3.83, |
|
"step": 1200, |
|
"total_flos": 1.4860396665534874e+18, |
|
"train_loss": 0.6305676238735517, |
|
"train_runtime": 1636.5913, |
|
"train_samples_per_second": 12.221, |
|
"train_steps_per_second": 0.765 |
|
} |
|
], |
|
"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 |
|
} |
|
|