| { | |
| "best_metric": 0.6129801869392395, | |
| "best_model_checkpoint": "Action_model/checkpoint-300", | |
| "epoch": 10.0, | |
| "eval_steps": 100, | |
| "global_step": 2680, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 2.570383071899414, | |
| "learning_rate": 9.96268656716418e-05, | |
| "loss": 0.1841, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 6.266295433044434, | |
| "learning_rate": 9.925373134328359e-05, | |
| "loss": 0.2301, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 8.001986503601074, | |
| "learning_rate": 9.888059701492539e-05, | |
| "loss": 0.2533, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 5.319194316864014, | |
| "learning_rate": 9.850746268656717e-05, | |
| "loss": 0.2436, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.9653372764587402, | |
| "learning_rate": 9.813432835820896e-05, | |
| "loss": 0.3712, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 7.348043441772461, | |
| "learning_rate": 9.776119402985075e-05, | |
| "loss": 0.3645, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 2.1969542503356934, | |
| "learning_rate": 9.738805970149254e-05, | |
| "loss": 0.4609, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 6.397550106048584, | |
| "learning_rate": 9.701492537313434e-05, | |
| "loss": 0.4755, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 6.923007488250732, | |
| "learning_rate": 9.664179104477612e-05, | |
| "loss": 0.3901, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 4.786198616027832, | |
| "learning_rate": 9.626865671641792e-05, | |
| "loss": 0.255, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_accuracy": 0.7926186291739895, | |
| "eval_loss": 0.7616190314292908, | |
| "eval_runtime": 8.7209, | |
| "eval_samples_per_second": 65.245, | |
| "eval_steps_per_second": 8.256, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 8.368223190307617, | |
| "learning_rate": 9.58955223880597e-05, | |
| "loss": 0.3784, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 4.078306198120117, | |
| "learning_rate": 9.552238805970149e-05, | |
| "loss": 0.4148, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 7.815361022949219, | |
| "learning_rate": 9.514925373134329e-05, | |
| "loss": 0.3621, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 11.498431205749512, | |
| "learning_rate": 9.477611940298507e-05, | |
| "loss": 0.3974, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 7.946558952331543, | |
| "learning_rate": 9.440298507462687e-05, | |
| "loss": 0.3856, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.3486919403076172, | |
| "learning_rate": 9.402985074626867e-05, | |
| "loss": 0.2435, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 4.267444133758545, | |
| "learning_rate": 9.365671641791045e-05, | |
| "loss": 0.3736, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 3.022345542907715, | |
| "learning_rate": 9.328358208955224e-05, | |
| "loss": 0.439, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 5.57196044921875, | |
| "learning_rate": 9.291044776119402e-05, | |
| "loss": 0.2996, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 2.636216640472412, | |
| "learning_rate": 9.253731343283582e-05, | |
| "loss": 0.2048, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_accuracy": 0.8084358523725835, | |
| "eval_loss": 0.724670946598053, | |
| "eval_runtime": 8.4461, | |
| "eval_samples_per_second": 67.368, | |
| "eval_steps_per_second": 8.525, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 1.615098237991333, | |
| "learning_rate": 9.216417910447762e-05, | |
| "loss": 0.3594, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 9.315821647644043, | |
| "learning_rate": 9.17910447761194e-05, | |
| "loss": 0.3046, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 3.669430732727051, | |
| "learning_rate": 9.14179104477612e-05, | |
| "loss": 0.4158, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 7.0882978439331055, | |
| "learning_rate": 9.104477611940299e-05, | |
| "loss": 0.3477, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 1.1667325496673584, | |
| "learning_rate": 9.067164179104479e-05, | |
| "loss": 0.316, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 1.482625961303711, | |
| "learning_rate": 9.029850746268657e-05, | |
| "loss": 0.3922, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 0.20793116092681885, | |
| "learning_rate": 8.992537313432836e-05, | |
| "loss": 0.3751, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 6.772298812866211, | |
| "learning_rate": 8.955223880597016e-05, | |
| "loss": 0.3269, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 5.833349227905273, | |
| "learning_rate": 8.917910447761194e-05, | |
| "loss": 0.3026, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 6.349458694458008, | |
| "learning_rate": 8.880597014925374e-05, | |
| "loss": 0.3763, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_accuracy": 0.8330404217926186, | |
| "eval_loss": 0.6129801869392395, | |
| "eval_runtime": 8.4095, | |
| "eval_samples_per_second": 67.661, | |
| "eval_steps_per_second": 8.562, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 4.767229080200195, | |
| "learning_rate": 8.843283582089554e-05, | |
| "loss": 0.3808, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "grad_norm": 12.675297737121582, | |
| "learning_rate": 8.805970149253732e-05, | |
| "loss": 0.3766, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 3.8118245601654053, | |
| "learning_rate": 8.76865671641791e-05, | |
| "loss": 0.2642, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "grad_norm": 8.736045837402344, | |
| "learning_rate": 8.731343283582089e-05, | |
| "loss": 0.3041, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "grad_norm": 6.683359146118164, | |
| "learning_rate": 8.694029850746269e-05, | |
| "loss": 0.1352, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "grad_norm": 4.780521392822266, | |
| "learning_rate": 8.656716417910447e-05, | |
| "loss": 0.4005, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "grad_norm": 9.654714584350586, | |
| "learning_rate": 8.619402985074627e-05, | |
| "loss": 0.3646, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 4.174666881561279, | |
| "learning_rate": 8.582089552238807e-05, | |
| "loss": 0.2353, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 7.596667289733887, | |
| "learning_rate": 8.548507462686568e-05, | |
| "loss": 0.3991, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "grad_norm": 5.592709064483643, | |
| "learning_rate": 8.511194029850747e-05, | |
| "loss": 0.307, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "eval_accuracy": 0.789103690685413, | |
| "eval_loss": 0.8137023448944092, | |
| "eval_runtime": 8.3292, | |
| "eval_samples_per_second": 68.314, | |
| "eval_steps_per_second": 8.644, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "grad_norm": 2.232590675354004, | |
| "learning_rate": 8.473880597014926e-05, | |
| "loss": 0.4669, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "grad_norm": 4.276609897613525, | |
| "learning_rate": 8.436567164179105e-05, | |
| "loss": 0.3831, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 7.262507915496826, | |
| "learning_rate": 8.399253731343283e-05, | |
| "loss": 0.3472, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 7.258556365966797, | |
| "learning_rate": 8.361940298507463e-05, | |
| "loss": 0.2396, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 4.945961952209473, | |
| "learning_rate": 8.324626865671642e-05, | |
| "loss": 0.2433, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 5.138702392578125, | |
| "learning_rate": 8.287313432835821e-05, | |
| "loss": 0.2947, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 1.1640909910202026, | |
| "learning_rate": 8.25e-05, | |
| "loss": 0.4791, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 4.626485824584961, | |
| "learning_rate": 8.21268656716418e-05, | |
| "loss": 0.286, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "grad_norm": 5.178492069244385, | |
| "learning_rate": 8.17537313432836e-05, | |
| "loss": 0.3202, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 7.854339122772217, | |
| "learning_rate": 8.138059701492538e-05, | |
| "loss": 0.3542, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_accuracy": 0.8014059753954306, | |
| "eval_loss": 0.6611581444740295, | |
| "eval_runtime": 8.5853, | |
| "eval_samples_per_second": 66.276, | |
| "eval_steps_per_second": 8.386, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 1.429740071296692, | |
| "learning_rate": 8.100746268656717e-05, | |
| "loss": 0.3039, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 2.9776551723480225, | |
| "learning_rate": 8.063432835820895e-05, | |
| "loss": 0.3825, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 10.557899475097656, | |
| "learning_rate": 8.026119402985075e-05, | |
| "loss": 0.5109, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 1.448002815246582, | |
| "learning_rate": 7.988805970149255e-05, | |
| "loss": 0.3421, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 4.500860691070557, | |
| "learning_rate": 7.951492537313433e-05, | |
| "loss": 0.3008, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 8.077374458312988, | |
| "learning_rate": 7.914179104477613e-05, | |
| "loss": 0.27, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 0.16809479892253876, | |
| "learning_rate": 7.876865671641792e-05, | |
| "loss": 0.2184, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 4.892763137817383, | |
| "learning_rate": 7.83955223880597e-05, | |
| "loss": 0.1479, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 8.35221004486084, | |
| "learning_rate": 7.80223880597015e-05, | |
| "loss": 0.3498, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 12.043429374694824, | |
| "learning_rate": 7.764925373134328e-05, | |
| "loss": 0.3518, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_accuracy": 0.8189806678383128, | |
| "eval_loss": 0.6964564919471741, | |
| "eval_runtime": 8.3878, | |
| "eval_samples_per_second": 67.837, | |
| "eval_steps_per_second": 8.584, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "grad_norm": 3.7737715244293213, | |
| "learning_rate": 7.727611940298508e-05, | |
| "loss": 0.3532, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "grad_norm": 4.282881736755371, | |
| "learning_rate": 7.690298507462687e-05, | |
| "loss": 0.2214, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 6.733531475067139, | |
| "learning_rate": 7.652985074626866e-05, | |
| "loss": 0.2709, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "grad_norm": 2.567267417907715, | |
| "learning_rate": 7.615671641791045e-05, | |
| "loss": 0.3725, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "grad_norm": 3.120966911315918, | |
| "learning_rate": 7.578358208955223e-05, | |
| "loss": 0.3036, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 6.505622386932373, | |
| "learning_rate": 7.541044776119403e-05, | |
| "loss": 0.2426, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 4.887637615203857, | |
| "learning_rate": 7.503731343283582e-05, | |
| "loss": 0.281, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "grad_norm": 9.790969848632812, | |
| "learning_rate": 7.466417910447762e-05, | |
| "loss": 0.4504, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "grad_norm": 4.354789733886719, | |
| "learning_rate": 7.429104477611941e-05, | |
| "loss": 0.4094, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "grad_norm": 5.015912055969238, | |
| "learning_rate": 7.39179104477612e-05, | |
| "loss": 0.3706, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "eval_accuracy": 0.804920913884007, | |
| "eval_loss": 0.7254143357276917, | |
| "eval_runtime": 8.3242, | |
| "eval_samples_per_second": 68.355, | |
| "eval_steps_per_second": 8.649, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 5.382541656494141, | |
| "learning_rate": 7.3544776119403e-05, | |
| "loss": 0.1722, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "grad_norm": 5.573971748352051, | |
| "learning_rate": 7.317164179104478e-05, | |
| "loss": 0.327, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "grad_norm": 3.5606117248535156, | |
| "learning_rate": 7.279850746268657e-05, | |
| "loss": 0.2702, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "grad_norm": 1.7398028373718262, | |
| "learning_rate": 7.242537313432837e-05, | |
| "loss": 0.238, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 2.7511751651763916, | |
| "learning_rate": 7.205223880597015e-05, | |
| "loss": 0.1848, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "grad_norm": 3.381510019302368, | |
| "learning_rate": 7.167910447761195e-05, | |
| "loss": 0.2261, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "grad_norm": 4.65634298324585, | |
| "learning_rate": 7.130597014925373e-05, | |
| "loss": 0.237, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "grad_norm": 10.35020923614502, | |
| "learning_rate": 7.093283582089553e-05, | |
| "loss": 0.3012, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 8.878485679626465, | |
| "learning_rate": 7.055970149253732e-05, | |
| "loss": 0.4094, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "grad_norm": 2.9728074073791504, | |
| "learning_rate": 7.01865671641791e-05, | |
| "loss": 0.4084, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.8101933216168717, | |
| "eval_loss": 0.6746156811714172, | |
| "eval_runtime": 8.2718, | |
| "eval_samples_per_second": 68.788, | |
| "eval_steps_per_second": 8.704, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 4.835368633270264, | |
| "learning_rate": 6.98134328358209e-05, | |
| "loss": 0.3152, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "grad_norm": 2.9197049140930176, | |
| "learning_rate": 6.944029850746268e-05, | |
| "loss": 0.3433, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 5.646128177642822, | |
| "learning_rate": 6.906716417910448e-05, | |
| "loss": 0.2604, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "grad_norm": 3.860607862472534, | |
| "learning_rate": 6.869402985074627e-05, | |
| "loss": 0.2831, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "grad_norm": 0.1358175426721573, | |
| "learning_rate": 6.832089552238807e-05, | |
| "loss": 0.242, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "grad_norm": 1.1011104583740234, | |
| "learning_rate": 6.794776119402985e-05, | |
| "loss": 0.2621, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "grad_norm": 7.837879180908203, | |
| "learning_rate": 6.757462686567164e-05, | |
| "loss": 0.249, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "grad_norm": 6.8647613525390625, | |
| "learning_rate": 6.720149253731343e-05, | |
| "loss": 0.3398, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "grad_norm": 2.8186678886413574, | |
| "learning_rate": 6.682835820895522e-05, | |
| "loss": 0.3092, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 4.623282432556152, | |
| "learning_rate": 6.645522388059702e-05, | |
| "loss": 0.2533, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "eval_accuracy": 0.8189806678383128, | |
| "eval_loss": 0.6866591572761536, | |
| "eval_runtime": 8.3143, | |
| "eval_samples_per_second": 68.436, | |
| "eval_steps_per_second": 8.66, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 4.85120964050293, | |
| "learning_rate": 6.608208955223882e-05, | |
| "loss": 0.2279, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "grad_norm": 0.7263774275779724, | |
| "learning_rate": 6.57089552238806e-05, | |
| "loss": 0.1725, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "grad_norm": 6.813180923461914, | |
| "learning_rate": 6.53358208955224e-05, | |
| "loss": 0.3304, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "grad_norm": 8.58501148223877, | |
| "learning_rate": 6.496268656716418e-05, | |
| "loss": 0.1864, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "grad_norm": 2.814436435699463, | |
| "learning_rate": 6.458955223880597e-05, | |
| "loss": 0.1496, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "grad_norm": 8.36603832244873, | |
| "learning_rate": 6.421641791044777e-05, | |
| "loss": 0.208, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "grad_norm": 3.5715956687927246, | |
| "learning_rate": 6.384328358208955e-05, | |
| "loss": 0.2429, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "grad_norm": 4.983556270599365, | |
| "learning_rate": 6.347014925373135e-05, | |
| "loss": 0.4053, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "grad_norm": 4.936723232269287, | |
| "learning_rate": 6.309701492537313e-05, | |
| "loss": 0.1545, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "grad_norm": 6.59185791015625, | |
| "learning_rate": 6.272388059701493e-05, | |
| "loss": 0.3147, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "eval_accuracy": 0.8189806678383128, | |
| "eval_loss": 0.7077136635780334, | |
| "eval_runtime": 8.3117, | |
| "eval_samples_per_second": 68.457, | |
| "eval_steps_per_second": 8.662, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "grad_norm": 9.348366737365723, | |
| "learning_rate": 6.235074626865672e-05, | |
| "loss": 0.3634, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "grad_norm": 9.918521881103516, | |
| "learning_rate": 6.19776119402985e-05, | |
| "loss": 0.3151, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 5.687044143676758, | |
| "learning_rate": 6.16044776119403e-05, | |
| "loss": 0.3088, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "grad_norm": 3.8347887992858887, | |
| "learning_rate": 6.123134328358209e-05, | |
| "loss": 0.2128, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "grad_norm": 5.380050182342529, | |
| "learning_rate": 6.0858208955223884e-05, | |
| "loss": 0.255, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "grad_norm": 8.848828315734863, | |
| "learning_rate": 6.0485074626865676e-05, | |
| "loss": 0.2794, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "grad_norm": 3.9666404724121094, | |
| "learning_rate": 6.011194029850746e-05, | |
| "loss": 0.1954, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "grad_norm": 0.3369455635547638, | |
| "learning_rate": 5.973880597014926e-05, | |
| "loss": 0.2298, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "grad_norm": 16.327823638916016, | |
| "learning_rate": 5.9365671641791044e-05, | |
| "loss": 0.2504, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "grad_norm": 7.070168495178223, | |
| "learning_rate": 5.8992537313432835e-05, | |
| "loss": 0.3182, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "eval_accuracy": 0.8189806678383128, | |
| "eval_loss": 0.6661401987075806, | |
| "eval_runtime": 8.2263, | |
| "eval_samples_per_second": 69.169, | |
| "eval_steps_per_second": 8.752, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "grad_norm": 2.853975534439087, | |
| "learning_rate": 5.8619402985074634e-05, | |
| "loss": 0.201, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "grad_norm": 0.958690881729126, | |
| "learning_rate": 5.824626865671642e-05, | |
| "loss": 0.1833, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "grad_norm": 3.4794461727142334, | |
| "learning_rate": 5.787313432835822e-05, | |
| "loss": 0.2796, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "grad_norm": 4.793296813964844, | |
| "learning_rate": 5.7499999999999995e-05, | |
| "loss": 0.2281, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "grad_norm": 6.200154781341553, | |
| "learning_rate": 5.712686567164179e-05, | |
| "loss": 0.2814, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "grad_norm": 5.616389274597168, | |
| "learning_rate": 5.675373134328359e-05, | |
| "loss": 0.1656, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "grad_norm": 9.382554054260254, | |
| "learning_rate": 5.6380597014925376e-05, | |
| "loss": 0.19, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 3.526240587234497, | |
| "learning_rate": 5.600746268656717e-05, | |
| "loss": 0.2063, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "grad_norm": 3.494896650314331, | |
| "learning_rate": 5.563432835820895e-05, | |
| "loss": 0.1681, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "grad_norm": 5.764057636260986, | |
| "learning_rate": 5.526119402985075e-05, | |
| "loss": 0.2248, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "eval_accuracy": 0.8418277680140598, | |
| "eval_loss": 0.6632041335105896, | |
| "eval_runtime": 8.1661, | |
| "eval_samples_per_second": 69.679, | |
| "eval_steps_per_second": 8.817, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "grad_norm": 4.680635452270508, | |
| "learning_rate": 5.488805970149254e-05, | |
| "loss": 0.2179, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "grad_norm": 10.24306869506836, | |
| "learning_rate": 5.451492537313433e-05, | |
| "loss": 0.2187, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "grad_norm": 3.054690361022949, | |
| "learning_rate": 5.4141791044776126e-05, | |
| "loss": 0.1729, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "grad_norm": 4.907272815704346, | |
| "learning_rate": 5.376865671641791e-05, | |
| "loss": 0.2762, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "grad_norm": 4.774748802185059, | |
| "learning_rate": 5.33955223880597e-05, | |
| "loss": 0.1965, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "grad_norm": 5.757875919342041, | |
| "learning_rate": 5.30223880597015e-05, | |
| "loss": 0.1564, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "grad_norm": 0.3608088791370392, | |
| "learning_rate": 5.2649253731343286e-05, | |
| "loss": 0.0946, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "grad_norm": 3.6289939880371094, | |
| "learning_rate": 5.227611940298508e-05, | |
| "loss": 0.3364, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "grad_norm": 5.132009029388428, | |
| "learning_rate": 5.190298507462686e-05, | |
| "loss": 0.231, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "grad_norm": 1.0347099304199219, | |
| "learning_rate": 5.152985074626866e-05, | |
| "loss": 0.1617, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "eval_accuracy": 0.8172231985940246, | |
| "eval_loss": 0.7277125716209412, | |
| "eval_runtime": 8.4693, | |
| "eval_samples_per_second": 67.184, | |
| "eval_steps_per_second": 8.501, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "grad_norm": 2.5996298789978027, | |
| "learning_rate": 5.115671641791045e-05, | |
| "loss": 0.385, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "grad_norm": 3.724181890487671, | |
| "learning_rate": 5.078358208955224e-05, | |
| "loss": 0.1786, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "grad_norm": 2.150557518005371, | |
| "learning_rate": 5.0410447761194035e-05, | |
| "loss": 0.2122, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.8813323974609375, | |
| "learning_rate": 5.003731343283582e-05, | |
| "loss": 0.2425, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "grad_norm": 0.896369457244873, | |
| "learning_rate": 4.966417910447762e-05, | |
| "loss": 0.2208, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "grad_norm": 9.002110481262207, | |
| "learning_rate": 4.92910447761194e-05, | |
| "loss": 0.1432, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "grad_norm": 9.619662284851074, | |
| "learning_rate": 4.8917910447761195e-05, | |
| "loss": 0.1347, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "grad_norm": 3.5148773193359375, | |
| "learning_rate": 4.8544776119402986e-05, | |
| "loss": 0.2837, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "grad_norm": 7.631669044494629, | |
| "learning_rate": 4.817164179104478e-05, | |
| "loss": 0.1887, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "grad_norm": 11.738872528076172, | |
| "learning_rate": 4.779850746268657e-05, | |
| "loss": 0.2578, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "eval_accuracy": 0.8189806678383128, | |
| "eval_loss": 0.7114442586898804, | |
| "eval_runtime": 8.2672, | |
| "eval_samples_per_second": 68.826, | |
| "eval_steps_per_second": 8.709, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "grad_norm": 6.67802095413208, | |
| "learning_rate": 4.742537313432836e-05, | |
| "loss": 0.2527, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "grad_norm": 4.491325378417969, | |
| "learning_rate": 4.705223880597015e-05, | |
| "loss": 0.2386, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "grad_norm": 1.1810379028320312, | |
| "learning_rate": 4.667910447761194e-05, | |
| "loss": 0.1693, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 5.37, | |
| "grad_norm": 6.075868129730225, | |
| "learning_rate": 4.6305970149253736e-05, | |
| "loss": 0.167, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "grad_norm": 2.315635919570923, | |
| "learning_rate": 4.593283582089553e-05, | |
| "loss": 0.2243, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "grad_norm": 10.839255332946777, | |
| "learning_rate": 4.555970149253732e-05, | |
| "loss": 0.2414, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "grad_norm": 4.562304496765137, | |
| "learning_rate": 4.5186567164179104e-05, | |
| "loss": 0.264, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "grad_norm": 1.8821789026260376, | |
| "learning_rate": 4.4813432835820895e-05, | |
| "loss": 0.1407, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "grad_norm": 8.406396865844727, | |
| "learning_rate": 4.4440298507462694e-05, | |
| "loss": 0.1454, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "grad_norm": 0.2816010117530823, | |
| "learning_rate": 4.406716417910448e-05, | |
| "loss": 0.1864, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "eval_accuracy": 0.8172231985940246, | |
| "eval_loss": 0.755394458770752, | |
| "eval_runtime": 8.2598, | |
| "eval_samples_per_second": 68.888, | |
| "eval_steps_per_second": 8.717, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.63, | |
| "grad_norm": 6.619854927062988, | |
| "learning_rate": 4.369402985074627e-05, | |
| "loss": 0.2806, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "grad_norm": 2.056018829345703, | |
| "learning_rate": 4.332089552238806e-05, | |
| "loss": 0.2583, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "grad_norm": 0.966521680355072, | |
| "learning_rate": 4.294776119402985e-05, | |
| "loss": 0.0997, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "grad_norm": 2.8261241912841797, | |
| "learning_rate": 4.2574626865671645e-05, | |
| "loss": 0.1604, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "grad_norm": 3.089912176132202, | |
| "learning_rate": 4.2201492537313436e-05, | |
| "loss": 0.2775, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 5.82, | |
| "grad_norm": 7.935690879821777, | |
| "learning_rate": 4.182835820895523e-05, | |
| "loss": 0.2522, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "grad_norm": 0.7999266982078552, | |
| "learning_rate": 4.145522388059702e-05, | |
| "loss": 0.0752, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 5.9, | |
| "grad_norm": 6.0712480545043945, | |
| "learning_rate": 4.1082089552238804e-05, | |
| "loss": 0.1933, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "grad_norm": 10.768308639526367, | |
| "learning_rate": 4.07089552238806e-05, | |
| "loss": 0.1664, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "grad_norm": 9.641716003417969, | |
| "learning_rate": 4.0335820895522394e-05, | |
| "loss": 0.3134, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "eval_accuracy": 0.8154657293497364, | |
| "eval_loss": 0.7593356966972351, | |
| "eval_runtime": 8.4455, | |
| "eval_samples_per_second": 67.373, | |
| "eval_steps_per_second": 8.525, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "grad_norm": 6.7538838386535645, | |
| "learning_rate": 3.996268656716418e-05, | |
| "loss": 0.1747, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "grad_norm": 6.237377166748047, | |
| "learning_rate": 3.958955223880597e-05, | |
| "loss": 0.2406, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "grad_norm": 7.950930118560791, | |
| "learning_rate": 3.921641791044776e-05, | |
| "loss": 0.1884, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "grad_norm": 4.41484260559082, | |
| "learning_rate": 3.8843283582089554e-05, | |
| "loss": 0.1445, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 6.16, | |
| "grad_norm": 6.339887619018555, | |
| "learning_rate": 3.8470149253731345e-05, | |
| "loss": 0.2906, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "grad_norm": 7.597599983215332, | |
| "learning_rate": 3.809701492537314e-05, | |
| "loss": 0.1576, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "grad_norm": 2.379629373550415, | |
| "learning_rate": 3.772388059701493e-05, | |
| "loss": 0.2016, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 6.27, | |
| "grad_norm": 2.7694478034973145, | |
| "learning_rate": 3.735074626865671e-05, | |
| "loss": 0.1188, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "grad_norm": 2.1837210655212402, | |
| "learning_rate": 3.6977611940298505e-05, | |
| "loss": 0.1908, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "grad_norm": 4.4665350914001465, | |
| "learning_rate": 3.66044776119403e-05, | |
| "loss": 0.24, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "eval_accuracy": 0.8260105448154658, | |
| "eval_loss": 0.7510848641395569, | |
| "eval_runtime": 8.2044, | |
| "eval_samples_per_second": 69.353, | |
| "eval_steps_per_second": 8.776, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "grad_norm": 4.6551995277404785, | |
| "learning_rate": 3.6231343283582095e-05, | |
| "loss": 0.1631, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 6.42, | |
| "grad_norm": 1.098407506942749, | |
| "learning_rate": 3.585820895522388e-05, | |
| "loss": 0.0912, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 6.46, | |
| "grad_norm": 0.37138649821281433, | |
| "learning_rate": 3.548507462686567e-05, | |
| "loss": 0.2621, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "grad_norm": 7.4571757316589355, | |
| "learning_rate": 3.511194029850746e-05, | |
| "loss": 0.268, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "grad_norm": 0.5180323123931885, | |
| "learning_rate": 3.4738805970149254e-05, | |
| "loss": 0.2135, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 6.57, | |
| "grad_norm": 1.0866820812225342, | |
| "learning_rate": 3.4365671641791046e-05, | |
| "loss": 0.1489, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "grad_norm": 8.90451717376709, | |
| "learning_rate": 3.399253731343284e-05, | |
| "loss": 0.288, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "grad_norm": 1.1608803272247314, | |
| "learning_rate": 3.361940298507463e-05, | |
| "loss": 0.18, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "grad_norm": 2.9207170009613037, | |
| "learning_rate": 3.3246268656716414e-05, | |
| "loss": 0.2414, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.2674783170223236, | |
| "learning_rate": 3.287313432835821e-05, | |
| "loss": 0.2359, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "eval_accuracy": 0.8137082601054482, | |
| "eval_loss": 0.7501537203788757, | |
| "eval_runtime": 8.1528, | |
| "eval_samples_per_second": 69.792, | |
| "eval_steps_per_second": 8.831, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "grad_norm": 8.241676330566406, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.1975, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "grad_norm": 2.0347325801849365, | |
| "learning_rate": 3.2126865671641796e-05, | |
| "loss": 0.218, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "grad_norm": 1.0338706970214844, | |
| "learning_rate": 3.175373134328358e-05, | |
| "loss": 0.1437, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "grad_norm": 0.34902578592300415, | |
| "learning_rate": 3.138059701492537e-05, | |
| "loss": 0.1883, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "grad_norm": 6.642534255981445, | |
| "learning_rate": 3.100746268656717e-05, | |
| "loss": 0.2513, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "grad_norm": 4.432920455932617, | |
| "learning_rate": 3.0634328358208955e-05, | |
| "loss": 0.1058, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "grad_norm": 4.381640434265137, | |
| "learning_rate": 3.0261194029850747e-05, | |
| "loss": 0.2114, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "grad_norm": 7.730411529541016, | |
| "learning_rate": 2.9888059701492538e-05, | |
| "loss": 0.2542, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "grad_norm": 7.122923851013184, | |
| "learning_rate": 2.9514925373134326e-05, | |
| "loss": 0.2594, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "grad_norm": 1.411278486251831, | |
| "learning_rate": 2.9141791044776125e-05, | |
| "loss": 0.2322, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "eval_accuracy": 0.8347978910369068, | |
| "eval_loss": 0.6952534317970276, | |
| "eval_runtime": 8.3769, | |
| "eval_samples_per_second": 67.925, | |
| "eval_steps_per_second": 8.595, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 7.13, | |
| "grad_norm": 2.219285011291504, | |
| "learning_rate": 2.8768656716417913e-05, | |
| "loss": 0.1344, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 7.16, | |
| "grad_norm": 6.302455902099609, | |
| "learning_rate": 2.8395522388059705e-05, | |
| "loss": 0.2098, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "grad_norm": 1.2837783098220825, | |
| "learning_rate": 2.8022388059701493e-05, | |
| "loss": 0.0906, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 7.24, | |
| "grad_norm": 6.604355335235596, | |
| "learning_rate": 2.7649253731343284e-05, | |
| "loss": 0.2352, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 7.28, | |
| "grad_norm": 9.916419982910156, | |
| "learning_rate": 2.727611940298508e-05, | |
| "loss": 0.1422, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "grad_norm": 2.7665014266967773, | |
| "learning_rate": 2.6902985074626868e-05, | |
| "loss": 0.1722, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 7.35, | |
| "grad_norm": 0.24231348931789398, | |
| "learning_rate": 2.652985074626866e-05, | |
| "loss": 0.2935, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 7.39, | |
| "grad_norm": 0.8025885224342346, | |
| "learning_rate": 2.6156716417910447e-05, | |
| "loss": 0.157, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "grad_norm": 1.6752264499664307, | |
| "learning_rate": 2.578358208955224e-05, | |
| "loss": 0.1256, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 7.46, | |
| "grad_norm": 2.404883861541748, | |
| "learning_rate": 2.5410447761194027e-05, | |
| "loss": 0.1514, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.46, | |
| "eval_accuracy": 0.8260105448154658, | |
| "eval_loss": 0.7120960354804993, | |
| "eval_runtime": 8.1425, | |
| "eval_samples_per_second": 69.88, | |
| "eval_steps_per_second": 8.842, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 5.409728050231934, | |
| "learning_rate": 2.5037313432835825e-05, | |
| "loss": 0.222, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "grad_norm": 3.949014663696289, | |
| "learning_rate": 2.4664179104477614e-05, | |
| "loss": 0.245, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 7.57, | |
| "grad_norm": 8.40086555480957, | |
| "learning_rate": 2.4291044776119405e-05, | |
| "loss": 0.1408, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 7.61, | |
| "grad_norm": 7.694955348968506, | |
| "learning_rate": 2.3917910447761197e-05, | |
| "loss": 0.2072, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "grad_norm": 1.9109055995941162, | |
| "learning_rate": 2.3544776119402985e-05, | |
| "loss": 0.145, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "grad_norm": 12.803776741027832, | |
| "learning_rate": 2.3171641791044777e-05, | |
| "loss": 0.1274, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "grad_norm": 3.3325235843658447, | |
| "learning_rate": 2.2798507462686568e-05, | |
| "loss": 0.1564, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 7.76, | |
| "grad_norm": 1.105327844619751, | |
| "learning_rate": 2.242537313432836e-05, | |
| "loss": 0.2008, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "grad_norm": 1.7592620849609375, | |
| "learning_rate": 2.2052238805970148e-05, | |
| "loss": 0.203, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "grad_norm": 0.13264060020446777, | |
| "learning_rate": 2.1679104477611943e-05, | |
| "loss": 0.2089, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "eval_accuracy": 0.827768014059754, | |
| "eval_loss": 0.693087637424469, | |
| "eval_runtime": 8.2375, | |
| "eval_samples_per_second": 69.074, | |
| "eval_steps_per_second": 8.741, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "grad_norm": 5.904381275177002, | |
| "learning_rate": 2.130597014925373e-05, | |
| "loss": 0.1754, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "grad_norm": 1.7469266653060913, | |
| "learning_rate": 2.0932835820895526e-05, | |
| "loss": 0.1322, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 7.95, | |
| "grad_norm": 4.313326835632324, | |
| "learning_rate": 2.0559701492537314e-05, | |
| "loss": 0.1418, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "grad_norm": 0.14211903512477875, | |
| "learning_rate": 2.0186567164179106e-05, | |
| "loss": 0.1534, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "grad_norm": 5.527184009552002, | |
| "learning_rate": 1.9813432835820897e-05, | |
| "loss": 0.2122, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "grad_norm": 0.2312430739402771, | |
| "learning_rate": 1.9440298507462686e-05, | |
| "loss": 0.1617, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "grad_norm": 0.23949085175991058, | |
| "learning_rate": 1.906716417910448e-05, | |
| "loss": 0.1286, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 8.13, | |
| "grad_norm": 0.1903185099363327, | |
| "learning_rate": 1.869402985074627e-05, | |
| "loss": 0.0846, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 8.17, | |
| "grad_norm": 0.08518023788928986, | |
| "learning_rate": 1.832089552238806e-05, | |
| "loss": 0.0801, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 8.21, | |
| "grad_norm": 4.424215793609619, | |
| "learning_rate": 1.7947761194029852e-05, | |
| "loss": 0.2245, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 8.21, | |
| "eval_accuracy": 0.8330404217926186, | |
| "eval_loss": 0.7087014317512512, | |
| "eval_runtime": 8.1117, | |
| "eval_samples_per_second": 70.145, | |
| "eval_steps_per_second": 8.876, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 8.25, | |
| "grad_norm": 7.247931480407715, | |
| "learning_rate": 1.7574626865671644e-05, | |
| "loss": 0.0722, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "grad_norm": 4.80264949798584, | |
| "learning_rate": 1.7201492537313435e-05, | |
| "loss": 0.0844, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "grad_norm": 8.001790046691895, | |
| "learning_rate": 1.6828358208955223e-05, | |
| "loss": 0.1077, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 8.36, | |
| "grad_norm": 5.419641017913818, | |
| "learning_rate": 1.6455223880597015e-05, | |
| "loss": 0.1627, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "grad_norm": 0.031686268746852875, | |
| "learning_rate": 1.6082089552238806e-05, | |
| "loss": 0.0984, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 8.43, | |
| "grad_norm": 6.095193862915039, | |
| "learning_rate": 1.5708955223880598e-05, | |
| "loss": 0.1756, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "grad_norm": 5.179446220397949, | |
| "learning_rate": 1.5335820895522386e-05, | |
| "loss": 0.1708, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 8.51, | |
| "grad_norm": 4.06497049331665, | |
| "learning_rate": 1.496268656716418e-05, | |
| "loss": 0.1493, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 8.54, | |
| "grad_norm": 1.4721342325210571, | |
| "learning_rate": 1.458955223880597e-05, | |
| "loss": 0.2587, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 8.58, | |
| "grad_norm": 4.418783664703369, | |
| "learning_rate": 1.4216417910447763e-05, | |
| "loss": 0.1328, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 8.58, | |
| "eval_accuracy": 0.8312829525483304, | |
| "eval_loss": 0.700339674949646, | |
| "eval_runtime": 8.481, | |
| "eval_samples_per_second": 67.091, | |
| "eval_steps_per_second": 8.49, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 8.62, | |
| "grad_norm": 1.5734038352966309, | |
| "learning_rate": 1.3843283582089553e-05, | |
| "loss": 0.165, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "grad_norm": 2.624784231185913, | |
| "learning_rate": 1.3470149253731342e-05, | |
| "loss": 0.0837, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 8.69, | |
| "grad_norm": 2.7039573192596436, | |
| "learning_rate": 1.3097014925373136e-05, | |
| "loss": 0.2098, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 8.73, | |
| "grad_norm": 6.542816638946533, | |
| "learning_rate": 1.2723880597014926e-05, | |
| "loss": 0.129, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "grad_norm": 2.9511120319366455, | |
| "learning_rate": 1.2350746268656717e-05, | |
| "loss": 0.1762, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 8.81, | |
| "grad_norm": 3.435502529144287, | |
| "learning_rate": 1.1977611940298509e-05, | |
| "loss": 0.1345, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 8.84, | |
| "grad_norm": 2.1689364910125732, | |
| "learning_rate": 1.1604477611940299e-05, | |
| "loss": 0.1011, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "grad_norm": 2.3366479873657227, | |
| "learning_rate": 1.123134328358209e-05, | |
| "loss": 0.1733, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 8.92, | |
| "grad_norm": 5.928171634674072, | |
| "learning_rate": 1.085820895522388e-05, | |
| "loss": 0.1089, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "grad_norm": 0.08636012673377991, | |
| "learning_rate": 1.0485074626865672e-05, | |
| "loss": 0.1304, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "eval_accuracy": 0.8224956063268892, | |
| "eval_loss": 0.7306046485900879, | |
| "eval_runtime": 8.4262, | |
| "eval_samples_per_second": 67.528, | |
| "eval_steps_per_second": 8.545, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "grad_norm": 0.14256200194358826, | |
| "learning_rate": 1.0111940298507463e-05, | |
| "loss": 0.1506, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "grad_norm": 0.4166848659515381, | |
| "learning_rate": 9.738805970149255e-06, | |
| "loss": 0.2058, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "grad_norm": 0.3997032344341278, | |
| "learning_rate": 9.365671641791045e-06, | |
| "loss": 0.0482, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 9.1, | |
| "grad_norm": 9.076058387756348, | |
| "learning_rate": 8.992537313432836e-06, | |
| "loss": 0.2201, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 9.14, | |
| "grad_norm": 4.368849277496338, | |
| "learning_rate": 8.619402985074628e-06, | |
| "loss": 0.1288, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 9.18, | |
| "grad_norm": 4.311466693878174, | |
| "learning_rate": 8.24626865671642e-06, | |
| "loss": 0.3058, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 9.22, | |
| "grad_norm": 0.2911408543586731, | |
| "learning_rate": 7.87313432835821e-06, | |
| "loss": 0.1303, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "grad_norm": 5.493233680725098, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.0915, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 9.29, | |
| "grad_norm": 0.09431172162294388, | |
| "learning_rate": 7.126865671641792e-06, | |
| "loss": 0.0954, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 9.33, | |
| "grad_norm": 1.8603869676589966, | |
| "learning_rate": 6.7537313432835825e-06, | |
| "loss": 0.1514, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 9.33, | |
| "eval_accuracy": 0.8260105448154658, | |
| "eval_loss": 0.7162156701087952, | |
| "eval_runtime": 8.3201, | |
| "eval_samples_per_second": 68.389, | |
| "eval_steps_per_second": 8.654, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 9.37, | |
| "grad_norm": 4.870584964752197, | |
| "learning_rate": 6.380597014925374e-06, | |
| "loss": 0.1354, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 9.4, | |
| "grad_norm": 2.316840410232544, | |
| "learning_rate": 6.007462686567165e-06, | |
| "loss": 0.1348, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "grad_norm": 1.9005101919174194, | |
| "learning_rate": 5.6343283582089556e-06, | |
| "loss": 0.1755, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "grad_norm": 0.1674620360136032, | |
| "learning_rate": 5.261194029850746e-06, | |
| "loss": 0.0878, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 9.51, | |
| "grad_norm": 5.729959011077881, | |
| "learning_rate": 4.888059701492537e-06, | |
| "loss": 0.1637, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 9.55, | |
| "grad_norm": 0.02724504843354225, | |
| "learning_rate": 4.514925373134329e-06, | |
| "loss": 0.1603, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 9.59, | |
| "grad_norm": 2.728663921356201, | |
| "learning_rate": 4.141791044776119e-06, | |
| "loss": 0.1152, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 9.63, | |
| "grad_norm": 8.920695304870605, | |
| "learning_rate": 3.7686567164179105e-06, | |
| "loss": 0.1964, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 9.66, | |
| "grad_norm": 2.3974239826202393, | |
| "learning_rate": 3.3955223880597013e-06, | |
| "loss": 0.0842, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 9.7, | |
| "grad_norm": 1.6431355476379395, | |
| "learning_rate": 3.022388059701493e-06, | |
| "loss": 0.2571, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 9.7, | |
| "eval_accuracy": 0.8347978910369068, | |
| "eval_loss": 0.7012546062469482, | |
| "eval_runtime": 8.3265, | |
| "eval_samples_per_second": 68.336, | |
| "eval_steps_per_second": 8.647, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "grad_norm": 0.10621854662895203, | |
| "learning_rate": 2.6492537313432836e-06, | |
| "loss": 0.2632, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "grad_norm": 4.150152206420898, | |
| "learning_rate": 2.2761194029850747e-06, | |
| "loss": 0.2804, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 9.81, | |
| "grad_norm": 4.01139497756958, | |
| "learning_rate": 1.9029850746268657e-06, | |
| "loss": 0.1696, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 9.85, | |
| "grad_norm": 4.7402262687683105, | |
| "learning_rate": 1.5298507462686568e-06, | |
| "loss": 0.1891, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "grad_norm": 4.460111141204834, | |
| "learning_rate": 1.1567164179104478e-06, | |
| "loss": 0.1178, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "grad_norm": 5.822507858276367, | |
| "learning_rate": 7.835820895522387e-07, | |
| "loss": 0.089, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 9.96, | |
| "grad_norm": 2.4408085346221924, | |
| "learning_rate": 4.1044776119402984e-07, | |
| "loss": 0.158, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 10.792135238647461, | |
| "learning_rate": 3.7313432835820895e-08, | |
| "loss": 0.2038, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 2680, | |
| "total_flos": 3.3230947683690086e+18, | |
| "train_loss": 0.23535207314277762, | |
| "train_runtime": 1371.8304, | |
| "train_samples_per_second": 31.258, | |
| "train_steps_per_second": 1.954 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2680, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "total_flos": 3.3230947683690086e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |