| { | |
| "best_metric": 0.782608695652174, | |
| "best_model_checkpoint": "CTMAE-P2-V2-S3/checkpoint-2925", | |
| "epoch": 49.02, | |
| "eval_steps": 500, | |
| "global_step": 3250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.003076923076923077, | |
| "grad_norm": 2.4690194129943848, | |
| "learning_rate": 3.0769230769230774e-07, | |
| "loss": 0.6769, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.006153846153846154, | |
| "grad_norm": 2.5120630264282227, | |
| "learning_rate": 6.153846153846155e-07, | |
| "loss": 0.6635, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.009230769230769232, | |
| "grad_norm": 5.427048206329346, | |
| "learning_rate": 9.230769230769232e-07, | |
| "loss": 0.6582, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.012307692307692308, | |
| "grad_norm": 5.626338958740234, | |
| "learning_rate": 1.230769230769231e-06, | |
| "loss": 0.6186, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.015384615384615385, | |
| "grad_norm": 5.170588493347168, | |
| "learning_rate": 1.5384615384615387e-06, | |
| "loss": 0.5612, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.018461538461538463, | |
| "grad_norm": 5.933913707733154, | |
| "learning_rate": 1.8461538461538465e-06, | |
| "loss": 0.6138, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "eval_accuracy": 0.5434782608695652, | |
| "eval_loss": 0.7341543436050415, | |
| "eval_runtime": 16.5125, | |
| "eval_samples_per_second": 2.786, | |
| "eval_steps_per_second": 0.727, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.0015384615384615, | |
| "grad_norm": 5.487543106079102, | |
| "learning_rate": 2.153846153846154e-06, | |
| "loss": 0.5948, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.0046153846153847, | |
| "grad_norm": 5.932001113891602, | |
| "learning_rate": 2.461538461538462e-06, | |
| "loss": 0.6058, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0076923076923077, | |
| "grad_norm": 6.421907901763916, | |
| "learning_rate": 2.7692307692307697e-06, | |
| "loss": 0.6592, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0107692307692309, | |
| "grad_norm": 6.242369651794434, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "loss": 0.5283, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0138461538461538, | |
| "grad_norm": 6.2263007164001465, | |
| "learning_rate": 3.384615384615385e-06, | |
| "loss": 0.4516, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.0169230769230768, | |
| "grad_norm": 21.957347869873047, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 0.5932, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 6.019967079162598, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.5607, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_accuracy": 0.5434782608695652, | |
| "eval_loss": 0.7364471554756165, | |
| "eval_runtime": 15.5184, | |
| "eval_samples_per_second": 2.964, | |
| "eval_steps_per_second": 0.773, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.003076923076923, | |
| "grad_norm": 10.679378509521484, | |
| "learning_rate": 4.307692307692308e-06, | |
| "loss": 0.4795, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.006153846153846, | |
| "grad_norm": 12.046272277832031, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 0.5978, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.0092307692307694, | |
| "grad_norm": 14.933501243591309, | |
| "learning_rate": 4.923076923076924e-06, | |
| "loss": 0.6278, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.0123076923076924, | |
| "grad_norm": 5.484428405761719, | |
| "learning_rate": 5.230769230769232e-06, | |
| "loss": 0.6293, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.0153846153846153, | |
| "grad_norm": 5.703970909118652, | |
| "learning_rate": 5.538461538461539e-06, | |
| "loss": 0.5672, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.0184615384615383, | |
| "grad_norm": 7.624868392944336, | |
| "learning_rate": 5.846153846153847e-06, | |
| "loss": 0.5525, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "eval_accuracy": 0.5434782608695652, | |
| "eval_loss": 0.9435462951660156, | |
| "eval_runtime": 15.4928, | |
| "eval_samples_per_second": 2.969, | |
| "eval_steps_per_second": 0.775, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 3.0015384615384617, | |
| "grad_norm": 5.229469299316406, | |
| "learning_rate": 6.153846153846155e-06, | |
| "loss": 0.5915, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.0046153846153847, | |
| "grad_norm": 5.959716320037842, | |
| "learning_rate": 6.461538461538463e-06, | |
| "loss": 0.5174, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.0076923076923077, | |
| "grad_norm": 5.900456428527832, | |
| "learning_rate": 6.76923076923077e-06, | |
| "loss": 0.5029, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.0107692307692306, | |
| "grad_norm": 5.498166084289551, | |
| "learning_rate": 7.076923076923078e-06, | |
| "loss": 0.5725, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.013846153846154, | |
| "grad_norm": 6.51685094833374, | |
| "learning_rate": 7.384615384615386e-06, | |
| "loss": 0.5766, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.016923076923077, | |
| "grad_norm": 22.756736755371094, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.4172, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 5.767360687255859, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.6208, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "eval_accuracy": 0.5869565217391305, | |
| "eval_loss": 0.7286421656608582, | |
| "eval_runtime": 14.2421, | |
| "eval_samples_per_second": 3.23, | |
| "eval_steps_per_second": 0.843, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.003076923076923, | |
| "grad_norm": 9.422948837280273, | |
| "learning_rate": 8.307692307692309e-06, | |
| "loss": 0.4313, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.006153846153846, | |
| "grad_norm": 6.0982985496521, | |
| "learning_rate": 8.615384615384617e-06, | |
| "loss": 0.4412, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.009230769230769, | |
| "grad_norm": 37.13871765136719, | |
| "learning_rate": 8.923076923076925e-06, | |
| "loss": 0.5237, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.012307692307692, | |
| "grad_norm": 0.8876226544380188, | |
| "learning_rate": 9.230769230769232e-06, | |
| "loss": 0.417, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.015384615384615, | |
| "grad_norm": 11.075803756713867, | |
| "learning_rate": 9.53846153846154e-06, | |
| "loss": 1.0414, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.018461538461539, | |
| "grad_norm": 2.8011114597320557, | |
| "learning_rate": 9.846153846153848e-06, | |
| "loss": 0.6099, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "eval_accuracy": 0.5434782608695652, | |
| "eval_loss": 0.7308201193809509, | |
| "eval_runtime": 14.1793, | |
| "eval_samples_per_second": 3.244, | |
| "eval_steps_per_second": 0.846, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 5.001538461538462, | |
| "grad_norm": 1.8966983556747437, | |
| "learning_rate": 9.982905982905984e-06, | |
| "loss": 0.6785, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 5.004615384615384, | |
| "grad_norm": 1.9870883226394653, | |
| "learning_rate": 9.94871794871795e-06, | |
| "loss": 0.6534, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 5.007692307692308, | |
| "grad_norm": 3.71256160736084, | |
| "learning_rate": 9.914529914529915e-06, | |
| "loss": 0.5627, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 5.010769230769231, | |
| "grad_norm": 7.984810829162598, | |
| "learning_rate": 9.880341880341882e-06, | |
| "loss": 0.4855, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 5.013846153846154, | |
| "grad_norm": 7.3561177253723145, | |
| "learning_rate": 9.846153846153848e-06, | |
| "loss": 0.5532, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 5.016923076923077, | |
| "grad_norm": 9.686047554016113, | |
| "learning_rate": 9.811965811965812e-06, | |
| "loss": 0.5581, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "grad_norm": 11.757854461669922, | |
| "learning_rate": 9.777777777777779e-06, | |
| "loss": 0.4065, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "eval_accuracy": 0.5434782608695652, | |
| "eval_loss": 1.057287573814392, | |
| "eval_runtime": 14.2388, | |
| "eval_samples_per_second": 3.231, | |
| "eval_steps_per_second": 0.843, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 6.003076923076923, | |
| "grad_norm": 5.493303298950195, | |
| "learning_rate": 9.743589743589744e-06, | |
| "loss": 0.7007, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.006153846153846, | |
| "grad_norm": 11.533708572387695, | |
| "learning_rate": 9.70940170940171e-06, | |
| "loss": 0.5369, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 6.009230769230769, | |
| "grad_norm": 7.627615928649902, | |
| "learning_rate": 9.675213675213677e-06, | |
| "loss": 0.5327, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 6.012307692307692, | |
| "grad_norm": 16.237632751464844, | |
| "learning_rate": 9.641025641025642e-06, | |
| "loss": 0.4611, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 6.015384615384615, | |
| "grad_norm": 25.719934463500977, | |
| "learning_rate": 9.606837606837607e-06, | |
| "loss": 0.5421, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 6.018461538461539, | |
| "grad_norm": 14.450688362121582, | |
| "learning_rate": 9.572649572649575e-06, | |
| "loss": 0.4338, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "eval_accuracy": 0.5869565217391305, | |
| "eval_loss": 0.6411370635032654, | |
| "eval_runtime": 14.1729, | |
| "eval_samples_per_second": 3.246, | |
| "eval_steps_per_second": 0.847, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 7.001538461538462, | |
| "grad_norm": 3.6783626079559326, | |
| "learning_rate": 9.53846153846154e-06, | |
| "loss": 0.61, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 7.004615384615384, | |
| "grad_norm": 11.61288070678711, | |
| "learning_rate": 9.504273504273504e-06, | |
| "loss": 0.6009, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 7.007692307692308, | |
| "grad_norm": 9.435476303100586, | |
| "learning_rate": 9.470085470085471e-06, | |
| "loss": 0.7558, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 7.010769230769231, | |
| "grad_norm": 9.540996551513672, | |
| "learning_rate": 9.435897435897436e-06, | |
| "loss": 0.3977, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 7.013846153846154, | |
| "grad_norm": 4.846683025360107, | |
| "learning_rate": 9.401709401709402e-06, | |
| "loss": 0.6812, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.016923076923077, | |
| "grad_norm": 8.068705558776855, | |
| "learning_rate": 9.367521367521369e-06, | |
| "loss": 0.558, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "grad_norm": 0.9865846037864685, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 0.4706, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 0.5360137820243835, | |
| "eval_runtime": 14.2447, | |
| "eval_samples_per_second": 3.229, | |
| "eval_steps_per_second": 0.842, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 8.003076923076923, | |
| "grad_norm": 2.319526433944702, | |
| "learning_rate": 9.2991452991453e-06, | |
| "loss": 0.7282, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 8.006153846153847, | |
| "grad_norm": 5.820730686187744, | |
| "learning_rate": 9.264957264957267e-06, | |
| "loss": 0.5329, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 8.009230769230768, | |
| "grad_norm": 0.8161588311195374, | |
| "learning_rate": 9.230769230769232e-06, | |
| "loss": 0.4894, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 8.012307692307692, | |
| "grad_norm": 8.13700008392334, | |
| "learning_rate": 9.196581196581196e-06, | |
| "loss": 0.5198, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 8.015384615384615, | |
| "grad_norm": 4.7142815589904785, | |
| "learning_rate": 9.162393162393163e-06, | |
| "loss": 0.3848, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 8.018461538461539, | |
| "grad_norm": 0.2144533395767212, | |
| "learning_rate": 9.128205128205129e-06, | |
| "loss": 0.4147, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "eval_accuracy": 0.5869565217391305, | |
| "eval_loss": 1.0753138065338135, | |
| "eval_runtime": 14.2289, | |
| "eval_samples_per_second": 3.233, | |
| "eval_steps_per_second": 0.843, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 9.00153846153846, | |
| "grad_norm": 9.62121868133545, | |
| "learning_rate": 9.094017094017094e-06, | |
| "loss": 1.0112, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 9.004615384615384, | |
| "grad_norm": 33.65559005737305, | |
| "learning_rate": 9.059829059829061e-06, | |
| "loss": 0.4409, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 9.007692307692308, | |
| "grad_norm": 23.247821807861328, | |
| "learning_rate": 9.025641025641027e-06, | |
| "loss": 0.3565, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 9.010769230769231, | |
| "grad_norm": 68.15277099609375, | |
| "learning_rate": 8.991452991452992e-06, | |
| "loss": 0.7108, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 9.013846153846155, | |
| "grad_norm": 5.1095476150512695, | |
| "learning_rate": 8.957264957264959e-06, | |
| "loss": 0.5591, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 9.016923076923076, | |
| "grad_norm": 14.021625518798828, | |
| "learning_rate": 8.923076923076925e-06, | |
| "loss": 0.3854, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "grad_norm": 8.44845199584961, | |
| "learning_rate": 8.888888888888888e-06, | |
| "loss": 0.6016, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "eval_accuracy": 0.5652173913043478, | |
| "eval_loss": 1.1058659553527832, | |
| "eval_runtime": 15.3529, | |
| "eval_samples_per_second": 2.996, | |
| "eval_steps_per_second": 0.782, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 10.003076923076923, | |
| "grad_norm": 6.827932834625244, | |
| "learning_rate": 8.854700854700855e-06, | |
| "loss": 0.7955, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 10.006153846153847, | |
| "grad_norm": 3.688110589981079, | |
| "learning_rate": 8.820512820512821e-06, | |
| "loss": 0.3492, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 10.009230769230768, | |
| "grad_norm": 13.171318054199219, | |
| "learning_rate": 8.786324786324786e-06, | |
| "loss": 0.4298, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 10.012307692307692, | |
| "grad_norm": 12.102375984191895, | |
| "learning_rate": 8.752136752136753e-06, | |
| "loss": 0.6888, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 10.015384615384615, | |
| "grad_norm": 6.251157760620117, | |
| "learning_rate": 8.717948717948719e-06, | |
| "loss": 0.3471, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 10.018461538461539, | |
| "grad_norm": 13.216636657714844, | |
| "learning_rate": 8.683760683760684e-06, | |
| "loss": 0.8252, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.5254927277565002, | |
| "eval_runtime": 14.3522, | |
| "eval_samples_per_second": 3.205, | |
| "eval_steps_per_second": 0.836, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 11.00153846153846, | |
| "grad_norm": 4.716592788696289, | |
| "learning_rate": 8.649572649572651e-06, | |
| "loss": 0.3548, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 11.004615384615384, | |
| "grad_norm": 1.752955675125122, | |
| "learning_rate": 8.615384615384617e-06, | |
| "loss": 0.2913, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 11.007692307692308, | |
| "grad_norm": 11.685894966125488, | |
| "learning_rate": 8.58119658119658e-06, | |
| "loss": 0.6061, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 11.010769230769231, | |
| "grad_norm": 14.7794189453125, | |
| "learning_rate": 8.547008547008548e-06, | |
| "loss": 0.4836, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 11.013846153846155, | |
| "grad_norm": 20.174833297729492, | |
| "learning_rate": 8.512820512820513e-06, | |
| "loss": 0.5189, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 11.016923076923076, | |
| "grad_norm": 8.418707847595215, | |
| "learning_rate": 8.478632478632479e-06, | |
| "loss": 0.4754, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "grad_norm": 10.69355583190918, | |
| "learning_rate": 8.444444444444446e-06, | |
| "loss": 0.3944, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "eval_accuracy": 0.5869565217391305, | |
| "eval_loss": 0.8291409611701965, | |
| "eval_runtime": 14.4575, | |
| "eval_samples_per_second": 3.182, | |
| "eval_steps_per_second": 0.83, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 12.003076923076923, | |
| "grad_norm": 23.261892318725586, | |
| "learning_rate": 8.410256410256411e-06, | |
| "loss": 0.5586, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 12.006153846153847, | |
| "grad_norm": 57.039825439453125, | |
| "learning_rate": 8.376068376068377e-06, | |
| "loss": 0.8555, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 12.009230769230768, | |
| "grad_norm": 3.4197776317596436, | |
| "learning_rate": 8.341880341880344e-06, | |
| "loss": 0.4604, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 12.012307692307692, | |
| "grad_norm": 9.3665189743042, | |
| "learning_rate": 8.307692307692309e-06, | |
| "loss": 0.477, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 12.015384615384615, | |
| "grad_norm": 14.621077537536621, | |
| "learning_rate": 8.273504273504273e-06, | |
| "loss": 0.6194, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 12.018461538461539, | |
| "grad_norm": 18.631038665771484, | |
| "learning_rate": 8.23931623931624e-06, | |
| "loss": 0.8652, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "eval_accuracy": 0.5652173913043478, | |
| "eval_loss": 0.726616382598877, | |
| "eval_runtime": 15.1578, | |
| "eval_samples_per_second": 3.035, | |
| "eval_steps_per_second": 0.792, | |
| "step": 845 | |
| }, | |
| { | |
| "epoch": 13.00153846153846, | |
| "grad_norm": 3.469635009765625, | |
| "learning_rate": 8.205128205128205e-06, | |
| "loss": 0.689, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 13.004615384615384, | |
| "grad_norm": 5.77168083190918, | |
| "learning_rate": 8.17094017094017e-06, | |
| "loss": 0.3198, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 13.007692307692308, | |
| "grad_norm": 12.537999153137207, | |
| "learning_rate": 8.136752136752138e-06, | |
| "loss": 0.4305, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 13.010769230769231, | |
| "grad_norm": 0.7631092667579651, | |
| "learning_rate": 8.102564102564103e-06, | |
| "loss": 0.6715, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 13.013846153846155, | |
| "grad_norm": 13.839919090270996, | |
| "learning_rate": 8.068376068376069e-06, | |
| "loss": 0.4264, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 13.016923076923076, | |
| "grad_norm": 11.262467384338379, | |
| "learning_rate": 8.034188034188036e-06, | |
| "loss": 0.3711, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "grad_norm": 15.98154354095459, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.4453, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "eval_accuracy": 0.6304347826086957, | |
| "eval_loss": 0.8913701772689819, | |
| "eval_runtime": 14.2586, | |
| "eval_samples_per_second": 3.226, | |
| "eval_steps_per_second": 0.842, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 14.003076923076923, | |
| "grad_norm": 22.157976150512695, | |
| "learning_rate": 7.965811965811967e-06, | |
| "loss": 0.3566, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 14.006153846153847, | |
| "grad_norm": 4.313268661499023, | |
| "learning_rate": 7.931623931623932e-06, | |
| "loss": 0.5692, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 14.009230769230768, | |
| "grad_norm": 11.119792938232422, | |
| "learning_rate": 7.897435897435898e-06, | |
| "loss": 0.5033, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 14.012307692307692, | |
| "grad_norm": 14.92096996307373, | |
| "learning_rate": 7.863247863247863e-06, | |
| "loss": 0.3935, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 14.015384615384615, | |
| "grad_norm": 8.508167266845703, | |
| "learning_rate": 7.82905982905983e-06, | |
| "loss": 0.2881, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 14.018461538461539, | |
| "grad_norm": 31.820133209228516, | |
| "learning_rate": 7.794871794871796e-06, | |
| "loss": 0.3375, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "eval_accuracy": 0.5869565217391305, | |
| "eval_loss": 1.4724668264389038, | |
| "eval_runtime": 14.2065, | |
| "eval_samples_per_second": 3.238, | |
| "eval_steps_per_second": 0.845, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 15.00153846153846, | |
| "grad_norm": 1.2260445356369019, | |
| "learning_rate": 7.760683760683761e-06, | |
| "loss": 0.2684, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 15.004615384615384, | |
| "grad_norm": 24.864896774291992, | |
| "learning_rate": 7.726495726495728e-06, | |
| "loss": 0.2463, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 15.007692307692308, | |
| "grad_norm": 8.236894607543945, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.4154, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 15.010769230769231, | |
| "grad_norm": 72.18476867675781, | |
| "learning_rate": 7.658119658119659e-06, | |
| "loss": 0.7205, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 15.013846153846155, | |
| "grad_norm": 3.46252703666687, | |
| "learning_rate": 7.6239316239316244e-06, | |
| "loss": 0.7416, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 15.016923076923076, | |
| "grad_norm": 23.594797134399414, | |
| "learning_rate": 7.58974358974359e-06, | |
| "loss": 0.3084, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 15.02, | |
| "grad_norm": 23.40220069885254, | |
| "learning_rate": 7.555555555555556e-06, | |
| "loss": 0.5423, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 15.02, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.6293339729309082, | |
| "eval_runtime": 14.542, | |
| "eval_samples_per_second": 3.163, | |
| "eval_steps_per_second": 0.825, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 16.00307692307692, | |
| "grad_norm": 6.501556873321533, | |
| "learning_rate": 7.521367521367522e-06, | |
| "loss": 0.1966, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 16.006153846153847, | |
| "grad_norm": 1.277785062789917, | |
| "learning_rate": 7.487179487179488e-06, | |
| "loss": 0.3867, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 16.00923076923077, | |
| "grad_norm": 0.18731409311294556, | |
| "learning_rate": 7.452991452991454e-06, | |
| "loss": 0.3975, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 16.012307692307694, | |
| "grad_norm": 25.202768325805664, | |
| "learning_rate": 7.4188034188034196e-06, | |
| "loss": 0.6083, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 16.015384615384615, | |
| "grad_norm": 3.939324378967285, | |
| "learning_rate": 7.384615384615386e-06, | |
| "loss": 0.2747, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 16.018461538461537, | |
| "grad_norm": 3.0128068923950195, | |
| "learning_rate": 7.350427350427351e-06, | |
| "loss": 0.4718, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 16.02, | |
| "eval_accuracy": 0.5869565217391305, | |
| "eval_loss": 0.9326255917549133, | |
| "eval_runtime": 15.4385, | |
| "eval_samples_per_second": 2.98, | |
| "eval_steps_per_second": 0.777, | |
| "step": 1105 | |
| }, | |
| { | |
| "epoch": 17.001538461538463, | |
| "grad_norm": 29.254545211791992, | |
| "learning_rate": 7.316239316239317e-06, | |
| "loss": 0.4704, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 17.004615384615384, | |
| "grad_norm": 1.9285635948181152, | |
| "learning_rate": 7.282051282051282e-06, | |
| "loss": 0.356, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 17.00769230769231, | |
| "grad_norm": 35.04645919799805, | |
| "learning_rate": 7.247863247863248e-06, | |
| "loss": 0.5782, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 17.01076923076923, | |
| "grad_norm": 5.088606834411621, | |
| "learning_rate": 7.213675213675214e-06, | |
| "loss": 0.4747, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 17.013846153846153, | |
| "grad_norm": 14.807530403137207, | |
| "learning_rate": 7.17948717948718e-06, | |
| "loss": 0.3342, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 17.016923076923078, | |
| "grad_norm": 5.486337661743164, | |
| "learning_rate": 7.145299145299146e-06, | |
| "loss": 0.6439, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 17.02, | |
| "grad_norm": 16.447872161865234, | |
| "learning_rate": 7.111111111111112e-06, | |
| "loss": 0.4305, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 17.02, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 0.6471044421195984, | |
| "eval_runtime": 14.2174, | |
| "eval_samples_per_second": 3.235, | |
| "eval_steps_per_second": 0.844, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 18.00307692307692, | |
| "grad_norm": 22.18661880493164, | |
| "learning_rate": 7.076923076923078e-06, | |
| "loss": 0.387, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 18.006153846153847, | |
| "grad_norm": 3.238065004348755, | |
| "learning_rate": 7.0427350427350435e-06, | |
| "loss": 0.1337, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 18.00923076923077, | |
| "grad_norm": 9.553592681884766, | |
| "learning_rate": 7.008547008547009e-06, | |
| "loss": 0.5108, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 18.012307692307694, | |
| "grad_norm": 20.022808074951172, | |
| "learning_rate": 6.974358974358974e-06, | |
| "loss": 0.4185, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 18.015384615384615, | |
| "grad_norm": 32.78887939453125, | |
| "learning_rate": 6.940170940170941e-06, | |
| "loss": 0.5136, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 18.018461538461537, | |
| "grad_norm": 25.71335792541504, | |
| "learning_rate": 6.905982905982906e-06, | |
| "loss": 0.1434, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 18.02, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 1.1058884859085083, | |
| "eval_runtime": 14.2564, | |
| "eval_samples_per_second": 3.227, | |
| "eval_steps_per_second": 0.842, | |
| "step": 1235 | |
| }, | |
| { | |
| "epoch": 19.001538461538463, | |
| "grad_norm": 55.081085205078125, | |
| "learning_rate": 6.871794871794872e-06, | |
| "loss": 0.1972, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 19.004615384615384, | |
| "grad_norm": 30.474956512451172, | |
| "learning_rate": 6.837606837606839e-06, | |
| "loss": 0.4069, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 19.00769230769231, | |
| "grad_norm": 34.68502426147461, | |
| "learning_rate": 6.803418803418804e-06, | |
| "loss": 0.5577, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 19.01076923076923, | |
| "grad_norm": 3.7890563011169434, | |
| "learning_rate": 6.76923076923077e-06, | |
| "loss": 0.4651, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 19.013846153846153, | |
| "grad_norm": 2.22894024848938, | |
| "learning_rate": 6.735042735042736e-06, | |
| "loss": 0.4381, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 19.016923076923078, | |
| "grad_norm": 1.855634093284607, | |
| "learning_rate": 6.700854700854701e-06, | |
| "loss": 0.4127, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 19.02, | |
| "grad_norm": 21.70796012878418, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.3129, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 19.02, | |
| "eval_accuracy": 0.7391304347826086, | |
| "eval_loss": 0.7987100481987, | |
| "eval_runtime": 14.9435, | |
| "eval_samples_per_second": 3.078, | |
| "eval_steps_per_second": 0.803, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 20.00307692307692, | |
| "grad_norm": 37.435733795166016, | |
| "learning_rate": 6.632478632478633e-06, | |
| "loss": 0.4383, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 20.006153846153847, | |
| "grad_norm": 0.6071020364761353, | |
| "learning_rate": 6.598290598290598e-06, | |
| "loss": 0.2316, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 20.00923076923077, | |
| "grad_norm": 8.772833824157715, | |
| "learning_rate": 6.564102564102565e-06, | |
| "loss": 0.8647, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 20.012307692307694, | |
| "grad_norm": 1.2445635795593262, | |
| "learning_rate": 6.529914529914531e-06, | |
| "loss": 0.256, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 20.015384615384615, | |
| "grad_norm": 12.750873565673828, | |
| "learning_rate": 6.495726495726496e-06, | |
| "loss": 0.3799, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 20.018461538461537, | |
| "grad_norm": 0.15409302711486816, | |
| "learning_rate": 6.461538461538463e-06, | |
| "loss": 0.2712, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 20.02, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 0.9583348631858826, | |
| "eval_runtime": 14.2561, | |
| "eval_samples_per_second": 3.227, | |
| "eval_steps_per_second": 0.842, | |
| "step": 1365 | |
| }, | |
| { | |
| "epoch": 21.001538461538463, | |
| "grad_norm": 0.9794425368309021, | |
| "learning_rate": 6.427350427350428e-06, | |
| "loss": 0.2697, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 21.004615384615384, | |
| "grad_norm": 41.89208221435547, | |
| "learning_rate": 6.3931623931623935e-06, | |
| "loss": 0.1116, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 21.00769230769231, | |
| "grad_norm": 0.219721257686615, | |
| "learning_rate": 6.358974358974359e-06, | |
| "loss": 0.3221, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 21.01076923076923, | |
| "grad_norm": 51.84693908691406, | |
| "learning_rate": 6.324786324786325e-06, | |
| "loss": 0.5398, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 21.013846153846153, | |
| "grad_norm": 0.23835326731204987, | |
| "learning_rate": 6.290598290598291e-06, | |
| "loss": 0.3207, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 21.016923076923078, | |
| "grad_norm": 33.43702697753906, | |
| "learning_rate": 6.256410256410257e-06, | |
| "loss": 0.2401, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 21.02, | |
| "grad_norm": 39.07565689086914, | |
| "learning_rate": 6.222222222222223e-06, | |
| "loss": 0.5669, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 21.02, | |
| "eval_accuracy": 0.5652173913043478, | |
| "eval_loss": 1.9777171611785889, | |
| "eval_runtime": 14.2455, | |
| "eval_samples_per_second": 3.229, | |
| "eval_steps_per_second": 0.842, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 22.00307692307692, | |
| "grad_norm": 19.675718307495117, | |
| "learning_rate": 6.188034188034189e-06, | |
| "loss": 0.2085, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 22.006153846153847, | |
| "grad_norm": 47.32659912109375, | |
| "learning_rate": 6.153846153846155e-06, | |
| "loss": 0.2847, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 22.00923076923077, | |
| "grad_norm": 20.55663299560547, | |
| "learning_rate": 6.11965811965812e-06, | |
| "loss": 0.6073, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 22.012307692307694, | |
| "grad_norm": 16.39328956604004, | |
| "learning_rate": 6.085470085470086e-06, | |
| "loss": 0.2029, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 22.015384615384615, | |
| "grad_norm": 0.12228210270404816, | |
| "learning_rate": 6.051282051282051e-06, | |
| "loss": 0.3122, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 22.018461538461537, | |
| "grad_norm": 0.8447695970535278, | |
| "learning_rate": 6.0170940170940174e-06, | |
| "loss": 0.3252, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 22.02, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.9683187007904053, | |
| "eval_runtime": 14.2253, | |
| "eval_samples_per_second": 3.234, | |
| "eval_steps_per_second": 0.844, | |
| "step": 1495 | |
| }, | |
| { | |
| "epoch": 23.001538461538463, | |
| "grad_norm": 0.7028045058250427, | |
| "learning_rate": 5.982905982905983e-06, | |
| "loss": 0.5694, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 23.004615384615384, | |
| "grad_norm": 0.02848893031477928, | |
| "learning_rate": 5.948717948717949e-06, | |
| "loss": 0.1348, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 23.00769230769231, | |
| "grad_norm": 16.87662696838379, | |
| "learning_rate": 5.914529914529915e-06, | |
| "loss": 0.243, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 23.01076923076923, | |
| "grad_norm": 90.88231658935547, | |
| "learning_rate": 5.880341880341881e-06, | |
| "loss": 0.1983, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 23.013846153846153, | |
| "grad_norm": 6.538637638092041, | |
| "learning_rate": 5.846153846153847e-06, | |
| "loss": 0.2995, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 23.016923076923078, | |
| "grad_norm": 29.127273559570312, | |
| "learning_rate": 5.8119658119658126e-06, | |
| "loss": 0.4176, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 23.02, | |
| "grad_norm": 62.22664260864258, | |
| "learning_rate": 5.777777777777778e-06, | |
| "loss": 0.1086, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 23.02, | |
| "eval_accuracy": 0.7391304347826086, | |
| "eval_loss": 1.0588734149932861, | |
| "eval_runtime": 14.7624, | |
| "eval_samples_per_second": 3.116, | |
| "eval_steps_per_second": 0.813, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 24.00307692307692, | |
| "grad_norm": 0.369827002286911, | |
| "learning_rate": 5.743589743589743e-06, | |
| "loss": 0.1041, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 24.006153846153847, | |
| "grad_norm": 0.0885409414768219, | |
| "learning_rate": 5.70940170940171e-06, | |
| "loss": 0.1651, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 24.00923076923077, | |
| "grad_norm": 38.52853012084961, | |
| "learning_rate": 5.675213675213675e-06, | |
| "loss": 0.4727, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 24.012307692307694, | |
| "grad_norm": 62.92534255981445, | |
| "learning_rate": 5.641025641025641e-06, | |
| "loss": 0.5216, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 24.015384615384615, | |
| "grad_norm": 162.53988647460938, | |
| "learning_rate": 5.606837606837608e-06, | |
| "loss": 0.2298, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 24.018461538461537, | |
| "grad_norm": 11.33154582977295, | |
| "learning_rate": 5.572649572649573e-06, | |
| "loss": 0.2289, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 24.02, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 1.0725319385528564, | |
| "eval_runtime": 14.2373, | |
| "eval_samples_per_second": 3.231, | |
| "eval_steps_per_second": 0.843, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 25.001538461538463, | |
| "grad_norm": 0.03378036618232727, | |
| "learning_rate": 5.538461538461539e-06, | |
| "loss": 0.107, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 25.004615384615384, | |
| "grad_norm": 15.620013236999512, | |
| "learning_rate": 5.504273504273505e-06, | |
| "loss": 0.1683, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 25.00769230769231, | |
| "grad_norm": 35.80552673339844, | |
| "learning_rate": 5.470085470085471e-06, | |
| "loss": 0.5356, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 25.01076923076923, | |
| "grad_norm": 119.33697509765625, | |
| "learning_rate": 5.435897435897436e-06, | |
| "loss": 0.3483, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 25.013846153846153, | |
| "grad_norm": 0.039877910166978836, | |
| "learning_rate": 5.401709401709402e-06, | |
| "loss": 0.3975, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 25.016923076923078, | |
| "grad_norm": 122.86112213134766, | |
| "learning_rate": 5.367521367521367e-06, | |
| "loss": 0.0873, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 25.02, | |
| "grad_norm": 0.03186870366334915, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 0.2527, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 25.02, | |
| "eval_accuracy": 0.7608695652173914, | |
| "eval_loss": 1.0045281648635864, | |
| "eval_runtime": 14.274, | |
| "eval_samples_per_second": 3.223, | |
| "eval_steps_per_second": 0.841, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 26.00307692307692, | |
| "grad_norm": 0.012652536854147911, | |
| "learning_rate": 5.2991452991453e-06, | |
| "loss": 0.0624, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 26.006153846153847, | |
| "grad_norm": 8.27054500579834, | |
| "learning_rate": 5.264957264957265e-06, | |
| "loss": 0.2865, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 26.00923076923077, | |
| "grad_norm": 0.5706243515014648, | |
| "learning_rate": 5.230769230769232e-06, | |
| "loss": 0.283, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 26.012307692307694, | |
| "grad_norm": 0.05628874897956848, | |
| "learning_rate": 5.196581196581197e-06, | |
| "loss": 0.2724, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 26.015384615384615, | |
| "grad_norm": 3.2168116569519043, | |
| "learning_rate": 5.162393162393163e-06, | |
| "loss": 0.2868, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 26.018461538461537, | |
| "grad_norm": 1.1663625240325928, | |
| "learning_rate": 5.128205128205128e-06, | |
| "loss": 0.433, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 26.02, | |
| "eval_accuracy": 0.6304347826086957, | |
| "eval_loss": 1.157365322113037, | |
| "eval_runtime": 15.047, | |
| "eval_samples_per_second": 3.057, | |
| "eval_steps_per_second": 0.798, | |
| "step": 1755 | |
| }, | |
| { | |
| "epoch": 27.001538461538463, | |
| "grad_norm": 0.0997692197561264, | |
| "learning_rate": 5.094017094017094e-06, | |
| "loss": 0.4229, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 27.004615384615384, | |
| "grad_norm": 0.09240197390317917, | |
| "learning_rate": 5.05982905982906e-06, | |
| "loss": 0.3206, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 27.00769230769231, | |
| "grad_norm": 0.2623593509197235, | |
| "learning_rate": 5.025641025641026e-06, | |
| "loss": 0.1986, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 27.01076923076923, | |
| "grad_norm": 0.24803876876831055, | |
| "learning_rate": 4.991452991452992e-06, | |
| "loss": 0.4088, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 27.013846153846153, | |
| "grad_norm": 0.5555140972137451, | |
| "learning_rate": 4.957264957264958e-06, | |
| "loss": 0.155, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 27.016923076923078, | |
| "grad_norm": 0.22051659226417542, | |
| "learning_rate": 4.923076923076924e-06, | |
| "loss": 0.2226, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 27.02, | |
| "grad_norm": 142.2720489501953, | |
| "learning_rate": 4.888888888888889e-06, | |
| "loss": 0.3203, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 27.02, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 1.0994583368301392, | |
| "eval_runtime": 14.2217, | |
| "eval_samples_per_second": 3.234, | |
| "eval_steps_per_second": 0.844, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 28.00307692307692, | |
| "grad_norm": 1.9806488752365112, | |
| "learning_rate": 4.854700854700855e-06, | |
| "loss": 0.0561, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 28.006153846153847, | |
| "grad_norm": 0.4285343289375305, | |
| "learning_rate": 4.820512820512821e-06, | |
| "loss": 0.0691, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 28.00923076923077, | |
| "grad_norm": 0.16625083982944489, | |
| "learning_rate": 4.786324786324787e-06, | |
| "loss": 0.2259, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 28.012307692307694, | |
| "grad_norm": 1.8851540088653564, | |
| "learning_rate": 4.752136752136752e-06, | |
| "loss": 0.0057, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 28.015384615384615, | |
| "grad_norm": 23.864330291748047, | |
| "learning_rate": 4.717948717948718e-06, | |
| "loss": 0.0304, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 28.018461538461537, | |
| "grad_norm": 36.58037567138672, | |
| "learning_rate": 4.6837606837606844e-06, | |
| "loss": 0.65, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 28.02, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 1.4325636625289917, | |
| "eval_runtime": 14.2225, | |
| "eval_samples_per_second": 3.234, | |
| "eval_steps_per_second": 0.844, | |
| "step": 1885 | |
| }, | |
| { | |
| "epoch": 29.001538461538463, | |
| "grad_norm": 97.73675537109375, | |
| "learning_rate": 4.64957264957265e-06, | |
| "loss": 0.1698, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 29.004615384615384, | |
| "grad_norm": 72.0632095336914, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 0.5534, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 29.00769230769231, | |
| "grad_norm": 36.123287200927734, | |
| "learning_rate": 4.581196581196582e-06, | |
| "loss": 0.6991, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 29.01076923076923, | |
| "grad_norm": 0.9478359222412109, | |
| "learning_rate": 4.547008547008547e-06, | |
| "loss": 0.3699, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 29.013846153846153, | |
| "grad_norm": 13.153339385986328, | |
| "learning_rate": 4.512820512820513e-06, | |
| "loss": 0.0125, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 29.016923076923078, | |
| "grad_norm": 0.0068915472365915775, | |
| "learning_rate": 4.4786324786324796e-06, | |
| "loss": 0.3134, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 29.02, | |
| "grad_norm": 0.5487728714942932, | |
| "learning_rate": 4.444444444444444e-06, | |
| "loss": 0.1041, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 29.02, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 1.217459797859192, | |
| "eval_runtime": 14.1868, | |
| "eval_samples_per_second": 3.242, | |
| "eval_steps_per_second": 0.846, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 30.00307692307692, | |
| "grad_norm": 0.011832280084490776, | |
| "learning_rate": 4.4102564102564104e-06, | |
| "loss": 0.1152, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 30.006153846153847, | |
| "grad_norm": 0.02865392342209816, | |
| "learning_rate": 4.376068376068377e-06, | |
| "loss": 0.0966, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 30.00923076923077, | |
| "grad_norm": 8.51937484741211, | |
| "learning_rate": 4.341880341880342e-06, | |
| "loss": 0.1849, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 30.012307692307694, | |
| "grad_norm": 0.006231443490833044, | |
| "learning_rate": 4.307692307692308e-06, | |
| "loss": 0.4387, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 30.015384615384615, | |
| "grad_norm": 1.779493808746338, | |
| "learning_rate": 4.273504273504274e-06, | |
| "loss": 0.2136, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 30.018461538461537, | |
| "grad_norm": 119.08480072021484, | |
| "learning_rate": 4.239316239316239e-06, | |
| "loss": 0.0569, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 30.02, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 1.4498846530914307, | |
| "eval_runtime": 14.237, | |
| "eval_samples_per_second": 3.231, | |
| "eval_steps_per_second": 0.843, | |
| "step": 2015 | |
| }, | |
| { | |
| "epoch": 31.001538461538463, | |
| "grad_norm": 1.9587482213974, | |
| "learning_rate": 4.2051282051282055e-06, | |
| "loss": 0.0864, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 31.004615384615384, | |
| "grad_norm": 193.9834442138672, | |
| "learning_rate": 4.170940170940172e-06, | |
| "loss": 0.328, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 31.00769230769231, | |
| "grad_norm": 4.229808807373047, | |
| "learning_rate": 4.136752136752136e-06, | |
| "loss": 0.0106, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 31.01076923076923, | |
| "grad_norm": 0.17659415304660797, | |
| "learning_rate": 4.102564102564103e-06, | |
| "loss": 0.0987, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 31.013846153846153, | |
| "grad_norm": 185.70263671875, | |
| "learning_rate": 4.068376068376069e-06, | |
| "loss": 0.5216, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 31.016923076923078, | |
| "grad_norm": 151.5210418701172, | |
| "learning_rate": 4.034188034188034e-06, | |
| "loss": 0.3362, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 31.02, | |
| "grad_norm": 0.8233569860458374, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.2142, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 31.02, | |
| "eval_accuracy": 0.7608695652173914, | |
| "eval_loss": 1.3655930757522583, | |
| "eval_runtime": 14.1707, | |
| "eval_samples_per_second": 3.246, | |
| "eval_steps_per_second": 0.847, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 32.003076923076925, | |
| "grad_norm": 3.8791003227233887, | |
| "learning_rate": 3.965811965811966e-06, | |
| "loss": 0.0783, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 32.00615384615384, | |
| "grad_norm": 5.642085552215576, | |
| "learning_rate": 3.9316239316239315e-06, | |
| "loss": 0.0175, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 32.00923076923077, | |
| "grad_norm": 58.637088775634766, | |
| "learning_rate": 3.897435897435898e-06, | |
| "loss": 0.2936, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 32.012307692307694, | |
| "grad_norm": 0.4610566198825836, | |
| "learning_rate": 3.863247863247864e-06, | |
| "loss": 0.1162, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 32.01538461538462, | |
| "grad_norm": 0.08907013386487961, | |
| "learning_rate": 3.8290598290598295e-06, | |
| "loss": 0.2103, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 32.01846153846154, | |
| "grad_norm": 0.04717539623379707, | |
| "learning_rate": 3.794871794871795e-06, | |
| "loss": 0.343, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 32.02, | |
| "eval_accuracy": 0.7608695652173914, | |
| "eval_loss": 1.3127285242080688, | |
| "eval_runtime": 14.1293, | |
| "eval_samples_per_second": 3.256, | |
| "eval_steps_per_second": 0.849, | |
| "step": 2145 | |
| }, | |
| { | |
| "epoch": 33.00153846153846, | |
| "grad_norm": 0.09433780610561371, | |
| "learning_rate": 3.760683760683761e-06, | |
| "loss": 0.2631, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 33.004615384615384, | |
| "grad_norm": 0.08145506680011749, | |
| "learning_rate": 3.726495726495727e-06, | |
| "loss": 0.0053, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 33.00769230769231, | |
| "grad_norm": 31.62529945373535, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 0.3741, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 33.01076923076923, | |
| "grad_norm": 198.175048828125, | |
| "learning_rate": 3.6581196581196584e-06, | |
| "loss": 0.2364, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 33.01384615384615, | |
| "grad_norm": 0.010738876648247242, | |
| "learning_rate": 3.623931623931624e-06, | |
| "loss": 0.2583, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 33.01692307692308, | |
| "grad_norm": 0.07434792816638947, | |
| "learning_rate": 3.58974358974359e-06, | |
| "loss": 0.1219, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 33.02, | |
| "grad_norm": 0.04365354776382446, | |
| "learning_rate": 3.555555555555556e-06, | |
| "loss": 0.0331, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 33.02, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 1.5137012004852295, | |
| "eval_runtime": 14.1663, | |
| "eval_samples_per_second": 3.247, | |
| "eval_steps_per_second": 0.847, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 34.003076923076925, | |
| "grad_norm": 0.04650627076625824, | |
| "learning_rate": 3.5213675213675218e-06, | |
| "loss": 0.2369, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 34.00615384615384, | |
| "grad_norm": 0.016520565375685692, | |
| "learning_rate": 3.487179487179487e-06, | |
| "loss": 0.0013, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 34.00923076923077, | |
| "grad_norm": 0.010590762831270695, | |
| "learning_rate": 3.452991452991453e-06, | |
| "loss": 0.1856, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 34.012307692307694, | |
| "grad_norm": 0.21154338121414185, | |
| "learning_rate": 3.4188034188034193e-06, | |
| "loss": 0.3057, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 34.01538461538462, | |
| "grad_norm": 282.3055419921875, | |
| "learning_rate": 3.384615384615385e-06, | |
| "loss": 0.1637, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 34.01846153846154, | |
| "grad_norm": 263.7860412597656, | |
| "learning_rate": 3.3504273504273506e-06, | |
| "loss": 0.1634, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 34.02, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 1.4773932695388794, | |
| "eval_runtime": 14.2162, | |
| "eval_samples_per_second": 3.236, | |
| "eval_steps_per_second": 0.844, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 35.00153846153846, | |
| "grad_norm": 0.02735510654747486, | |
| "learning_rate": 3.3162393162393165e-06, | |
| "loss": 0.0023, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 35.004615384615384, | |
| "grad_norm": 0.04332771152257919, | |
| "learning_rate": 3.2820512820512823e-06, | |
| "loss": 0.2221, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 35.00769230769231, | |
| "grad_norm": 0.016110800206661224, | |
| "learning_rate": 3.247863247863248e-06, | |
| "loss": 0.2181, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 35.01076923076923, | |
| "grad_norm": 0.12393196672201157, | |
| "learning_rate": 3.213675213675214e-06, | |
| "loss": 0.4696, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 35.01384615384615, | |
| "grad_norm": 0.06598012894392014, | |
| "learning_rate": 3.1794871794871795e-06, | |
| "loss": 0.1672, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 35.01692307692308, | |
| "grad_norm": 0.004367148969322443, | |
| "learning_rate": 3.1452991452991453e-06, | |
| "loss": 0.0012, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 35.02, | |
| "grad_norm": 0.01092283334583044, | |
| "learning_rate": 3.1111111111111116e-06, | |
| "loss": 0.5358, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 35.02, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 1.5174051523208618, | |
| "eval_runtime": 14.2031, | |
| "eval_samples_per_second": 3.239, | |
| "eval_steps_per_second": 0.845, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 36.003076923076925, | |
| "grad_norm": 0.026499304920434952, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "loss": 0.1287, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 36.00615384615384, | |
| "grad_norm": 2.692544460296631, | |
| "learning_rate": 3.042735042735043e-06, | |
| "loss": 0.2211, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 36.00923076923077, | |
| "grad_norm": 0.026944167912006378, | |
| "learning_rate": 3.0085470085470087e-06, | |
| "loss": 0.01, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 36.012307692307694, | |
| "grad_norm": 0.16730260848999023, | |
| "learning_rate": 2.9743589743589746e-06, | |
| "loss": 0.3199, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 36.01538461538462, | |
| "grad_norm": 11.841240882873535, | |
| "learning_rate": 2.9401709401709404e-06, | |
| "loss": 0.0015, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 36.01846153846154, | |
| "grad_norm": 206.77235412597656, | |
| "learning_rate": 2.9059829059829063e-06, | |
| "loss": 0.0396, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 36.02, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 1.447475552558899, | |
| "eval_runtime": 14.233, | |
| "eval_samples_per_second": 3.232, | |
| "eval_steps_per_second": 0.843, | |
| "step": 2405 | |
| }, | |
| { | |
| "epoch": 37.00153846153846, | |
| "grad_norm": 0.008494521491229534, | |
| "learning_rate": 2.8717948717948717e-06, | |
| "loss": 0.0187, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 37.004615384615384, | |
| "grad_norm": 0.12653498351573944, | |
| "learning_rate": 2.8376068376068376e-06, | |
| "loss": 0.188, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 37.00769230769231, | |
| "grad_norm": 0.06878158450126648, | |
| "learning_rate": 2.803418803418804e-06, | |
| "loss": 0.1975, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 37.01076923076923, | |
| "grad_norm": 0.012811361812055111, | |
| "learning_rate": 2.7692307692307697e-06, | |
| "loss": 0.0616, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 37.01384615384615, | |
| "grad_norm": 0.012951454147696495, | |
| "learning_rate": 2.7350427350427355e-06, | |
| "loss": 0.0182, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 37.01692307692308, | |
| "grad_norm": 0.015709415078163147, | |
| "learning_rate": 2.700854700854701e-06, | |
| "loss": 0.1402, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 37.02, | |
| "grad_norm": 111.22488403320312, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 0.7272, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 37.02, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 1.6262131929397583, | |
| "eval_runtime": 14.2867, | |
| "eval_samples_per_second": 3.22, | |
| "eval_steps_per_second": 0.84, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 38.003076923076925, | |
| "grad_norm": 0.6310192346572876, | |
| "learning_rate": 2.6324786324786327e-06, | |
| "loss": 0.069, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 38.00615384615384, | |
| "grad_norm": 0.005475493147969246, | |
| "learning_rate": 2.5982905982905985e-06, | |
| "loss": 0.454, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 38.00923076923077, | |
| "grad_norm": 0.031099509447813034, | |
| "learning_rate": 2.564102564102564e-06, | |
| "loss": 0.2161, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 38.012307692307694, | |
| "grad_norm": 0.025538576766848564, | |
| "learning_rate": 2.52991452991453e-06, | |
| "loss": 0.0924, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 38.01538461538462, | |
| "grad_norm": 0.26595520973205566, | |
| "learning_rate": 2.495726495726496e-06, | |
| "loss": 0.4355, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 38.01846153846154, | |
| "grad_norm": 0.00920882262289524, | |
| "learning_rate": 2.461538461538462e-06, | |
| "loss": 0.3445, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 38.02, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 1.787401795387268, | |
| "eval_runtime": 14.2035, | |
| "eval_samples_per_second": 3.239, | |
| "eval_steps_per_second": 0.845, | |
| "step": 2535 | |
| }, | |
| { | |
| "epoch": 39.00153846153846, | |
| "grad_norm": 0.216685950756073, | |
| "learning_rate": 2.4273504273504274e-06, | |
| "loss": 0.1452, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 39.004615384615384, | |
| "grad_norm": 228.7432098388672, | |
| "learning_rate": 2.3931623931623937e-06, | |
| "loss": 0.1403, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 39.00769230769231, | |
| "grad_norm": 0.0731094554066658, | |
| "learning_rate": 2.358974358974359e-06, | |
| "loss": 0.0189, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 39.01076923076923, | |
| "grad_norm": 0.01636696793138981, | |
| "learning_rate": 2.324786324786325e-06, | |
| "loss": 0.1849, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 39.01384615384615, | |
| "grad_norm": 0.005581328645348549, | |
| "learning_rate": 2.290598290598291e-06, | |
| "loss": 0.3136, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 39.01692307692308, | |
| "grad_norm": 0.13789384067058563, | |
| "learning_rate": 2.2564102564102566e-06, | |
| "loss": 0.0827, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 39.02, | |
| "grad_norm": 0.01795981265604496, | |
| "learning_rate": 2.222222222222222e-06, | |
| "loss": 0.2743, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 39.02, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 1.5739110708236694, | |
| "eval_runtime": 14.2448, | |
| "eval_samples_per_second": 3.229, | |
| "eval_steps_per_second": 0.842, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 40.003076923076925, | |
| "grad_norm": 77.0948715209961, | |
| "learning_rate": 2.1880341880341884e-06, | |
| "loss": 0.1258, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 40.00615384615384, | |
| "grad_norm": 0.06270722299814224, | |
| "learning_rate": 2.153846153846154e-06, | |
| "loss": 0.1022, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 40.00923076923077, | |
| "grad_norm": 0.011491877026855946, | |
| "learning_rate": 2.1196581196581196e-06, | |
| "loss": 0.0004, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 40.012307692307694, | |
| "grad_norm": 204.40928649902344, | |
| "learning_rate": 2.085470085470086e-06, | |
| "loss": 0.0763, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 40.01538461538462, | |
| "grad_norm": 0.006064881104975939, | |
| "learning_rate": 2.0512820512820513e-06, | |
| "loss": 0.2972, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 40.01846153846154, | |
| "grad_norm": 0.01428136881440878, | |
| "learning_rate": 2.017094017094017e-06, | |
| "loss": 0.2592, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 40.02, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 1.6092571020126343, | |
| "eval_runtime": 14.2231, | |
| "eval_samples_per_second": 3.234, | |
| "eval_steps_per_second": 0.844, | |
| "step": 2665 | |
| }, | |
| { | |
| "epoch": 41.00153846153846, | |
| "grad_norm": 237.1355743408203, | |
| "learning_rate": 1.982905982905983e-06, | |
| "loss": 0.2178, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 41.004615384615384, | |
| "grad_norm": 0.01185480784624815, | |
| "learning_rate": 1.948717948717949e-06, | |
| "loss": 0.0087, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 41.00769230769231, | |
| "grad_norm": 0.01697443798184395, | |
| "learning_rate": 1.9145299145299148e-06, | |
| "loss": 0.4878, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 41.01076923076923, | |
| "grad_norm": 275.4456481933594, | |
| "learning_rate": 1.8803418803418804e-06, | |
| "loss": 0.0663, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 41.01384615384615, | |
| "grad_norm": 0.5372545719146729, | |
| "learning_rate": 1.8461538461538465e-06, | |
| "loss": 0.0006, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 41.01692307692308, | |
| "grad_norm": 0.0075569795444607735, | |
| "learning_rate": 1.811965811965812e-06, | |
| "loss": 0.0014, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 41.02, | |
| "grad_norm": 0.09308155626058578, | |
| "learning_rate": 1.777777777777778e-06, | |
| "loss": 0.1367, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 41.02, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 1.5614510774612427, | |
| "eval_runtime": 14.1903, | |
| "eval_samples_per_second": 3.242, | |
| "eval_steps_per_second": 0.846, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 42.003076923076925, | |
| "grad_norm": 0.07777335494756699, | |
| "learning_rate": 1.7435897435897436e-06, | |
| "loss": 0.0557, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 42.00615384615384, | |
| "grad_norm": 0.005379654001444578, | |
| "learning_rate": 1.7094017094017097e-06, | |
| "loss": 0.0125, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 42.00923076923077, | |
| "grad_norm": 278.01739501953125, | |
| "learning_rate": 1.6752136752136753e-06, | |
| "loss": 0.1841, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 42.012307692307694, | |
| "grad_norm": 0.15389478206634521, | |
| "learning_rate": 1.6410256410256412e-06, | |
| "loss": 0.0351, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 42.01538461538462, | |
| "grad_norm": 70.6353530883789, | |
| "learning_rate": 1.606837606837607e-06, | |
| "loss": 0.3138, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 42.01846153846154, | |
| "grad_norm": 0.006937114987522364, | |
| "learning_rate": 1.5726495726495727e-06, | |
| "loss": 0.0459, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 42.02, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 1.5237913131713867, | |
| "eval_runtime": 13.2102, | |
| "eval_samples_per_second": 3.482, | |
| "eval_steps_per_second": 0.908, | |
| "step": 2795 | |
| }, | |
| { | |
| "epoch": 43.00153846153846, | |
| "grad_norm": 0.016063353046774864, | |
| "learning_rate": 1.5384615384615387e-06, | |
| "loss": 0.0381, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 43.004615384615384, | |
| "grad_norm": 0.12228815257549286, | |
| "learning_rate": 1.5042735042735044e-06, | |
| "loss": 0.001, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 43.00769230769231, | |
| "grad_norm": 25.483070373535156, | |
| "learning_rate": 1.4700854700854702e-06, | |
| "loss": 0.1634, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 43.01076923076923, | |
| "grad_norm": 9.318077087402344, | |
| "learning_rate": 1.4358974358974359e-06, | |
| "loss": 0.317, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 43.01384615384615, | |
| "grad_norm": 0.2631111443042755, | |
| "learning_rate": 1.401709401709402e-06, | |
| "loss": 0.018, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 43.01692307692308, | |
| "grad_norm": 0.2109161764383316, | |
| "learning_rate": 1.3675213675213678e-06, | |
| "loss": 0.0432, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 43.02, | |
| "grad_norm": 0.0032740840688347816, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 0.0102, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 43.02, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 1.4446630477905273, | |
| "eval_runtime": 13.2357, | |
| "eval_samples_per_second": 3.475, | |
| "eval_steps_per_second": 0.907, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 44.003076923076925, | |
| "grad_norm": 0.050007205456495285, | |
| "learning_rate": 1.2991452991452993e-06, | |
| "loss": 0.0036, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 44.00615384615384, | |
| "grad_norm": 0.02491617575287819, | |
| "learning_rate": 1.264957264957265e-06, | |
| "loss": 0.2155, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 44.00923076923077, | |
| "grad_norm": 0.01646566204726696, | |
| "learning_rate": 1.230769230769231e-06, | |
| "loss": 0.0006, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 44.012307692307694, | |
| "grad_norm": 0.6604272723197937, | |
| "learning_rate": 1.1965811965811968e-06, | |
| "loss": 0.3125, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 44.01538461538462, | |
| "grad_norm": 0.008244900032877922, | |
| "learning_rate": 1.1623931623931625e-06, | |
| "loss": 0.1206, | |
| "step": 2910 | |
| }, | |
| { | |
| "epoch": 44.01846153846154, | |
| "grad_norm": 4.3739237785339355, | |
| "learning_rate": 1.1282051282051283e-06, | |
| "loss": 0.0084, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 44.02, | |
| "eval_accuracy": 0.782608695652174, | |
| "eval_loss": 1.3307474851608276, | |
| "eval_runtime": 14.3847, | |
| "eval_samples_per_second": 3.198, | |
| "eval_steps_per_second": 0.834, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 45.00153846153846, | |
| "grad_norm": 0.011739898473024368, | |
| "learning_rate": 1.0940170940170942e-06, | |
| "loss": 0.004, | |
| "step": 2930 | |
| }, | |
| { | |
| "epoch": 45.004615384615384, | |
| "grad_norm": 0.011195595376193523, | |
| "learning_rate": 1.0598290598290598e-06, | |
| "loss": 0.0017, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 45.00769230769231, | |
| "grad_norm": 0.007319163531064987, | |
| "learning_rate": 1.0256410256410257e-06, | |
| "loss": 0.1595, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 45.01076923076923, | |
| "grad_norm": 193.5129852294922, | |
| "learning_rate": 9.914529914529915e-07, | |
| "loss": 0.1625, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 45.01384615384615, | |
| "grad_norm": 0.01686985418200493, | |
| "learning_rate": 9.572649572649574e-07, | |
| "loss": 0.0014, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 45.01692307692308, | |
| "grad_norm": 0.09226860105991364, | |
| "learning_rate": 9.230769230769232e-07, | |
| "loss": 0.5208, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 45.02, | |
| "grad_norm": 0.21285171806812286, | |
| "learning_rate": 8.88888888888889e-07, | |
| "loss": 0.106, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 45.02, | |
| "eval_accuracy": 0.7608695652173914, | |
| "eval_loss": 1.3326114416122437, | |
| "eval_runtime": 14.1718, | |
| "eval_samples_per_second": 3.246, | |
| "eval_steps_per_second": 0.847, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 46.003076923076925, | |
| "grad_norm": 0.07202818244695663, | |
| "learning_rate": 8.547008547008548e-07, | |
| "loss": 0.0018, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 46.00615384615384, | |
| "grad_norm": 61.24966812133789, | |
| "learning_rate": 8.205128205128206e-07, | |
| "loss": 0.1323, | |
| "step": 3010 | |
| }, | |
| { | |
| "epoch": 46.00923076923077, | |
| "grad_norm": 0.006612909492105246, | |
| "learning_rate": 7.863247863247863e-07, | |
| "loss": 0.0181, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 46.012307692307694, | |
| "grad_norm": 0.028757771477103233, | |
| "learning_rate": 7.521367521367522e-07, | |
| "loss": 0.0946, | |
| "step": 3030 | |
| }, | |
| { | |
| "epoch": 46.01538461538462, | |
| "grad_norm": 0.12236111611127853, | |
| "learning_rate": 7.179487179487179e-07, | |
| "loss": 0.0006, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 46.01846153846154, | |
| "grad_norm": 0.01944558136165142, | |
| "learning_rate": 6.837606837606839e-07, | |
| "loss": 0.0796, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 46.02, | |
| "eval_accuracy": 0.7391304347826086, | |
| "eval_loss": 1.3509420156478882, | |
| "eval_runtime": 14.2379, | |
| "eval_samples_per_second": 3.231, | |
| "eval_steps_per_second": 0.843, | |
| "step": 3055 | |
| }, | |
| { | |
| "epoch": 47.00153846153846, | |
| "grad_norm": 109.69795989990234, | |
| "learning_rate": 6.495726495726496e-07, | |
| "loss": 0.0994, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 47.004615384615384, | |
| "grad_norm": 0.015416641719639301, | |
| "learning_rate": 6.153846153846155e-07, | |
| "loss": 0.114, | |
| "step": 3070 | |
| }, | |
| { | |
| "epoch": 47.00769230769231, | |
| "grad_norm": 0.023845411837100983, | |
| "learning_rate": 5.811965811965812e-07, | |
| "loss": 0.0067, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 47.01076923076923, | |
| "grad_norm": 0.007653037551790476, | |
| "learning_rate": 5.470085470085471e-07, | |
| "loss": 0.0013, | |
| "step": 3090 | |
| }, | |
| { | |
| "epoch": 47.01384615384615, | |
| "grad_norm": 0.6111039519309998, | |
| "learning_rate": 5.128205128205128e-07, | |
| "loss": 0.0334, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 47.01692307692308, | |
| "grad_norm": 0.005491567775607109, | |
| "learning_rate": 4.786324786324787e-07, | |
| "loss": 0.0007, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 47.02, | |
| "grad_norm": 0.012984632514417171, | |
| "learning_rate": 4.444444444444445e-07, | |
| "loss": 0.0004, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 47.02, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 1.5340851545333862, | |
| "eval_runtime": 14.5799, | |
| "eval_samples_per_second": 3.155, | |
| "eval_steps_per_second": 0.823, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 48.003076923076925, | |
| "grad_norm": 0.008926596492528915, | |
| "learning_rate": 4.102564102564103e-07, | |
| "loss": 0.0007, | |
| "step": 3130 | |
| }, | |
| { | |
| "epoch": 48.00615384615384, | |
| "grad_norm": 0.0157285425812006, | |
| "learning_rate": 3.760683760683761e-07, | |
| "loss": 0.157, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 48.00923076923077, | |
| "grad_norm": 0.0048833079636096954, | |
| "learning_rate": 3.4188034188034194e-07, | |
| "loss": 0.0005, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 48.012307692307694, | |
| "grad_norm": 0.10273938626050949, | |
| "learning_rate": 3.0769230769230774e-07, | |
| "loss": 0.137, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 48.01538461538462, | |
| "grad_norm": 0.011431020684540272, | |
| "learning_rate": 2.7350427350427354e-07, | |
| "loss": 0.1131, | |
| "step": 3170 | |
| }, | |
| { | |
| "epoch": 48.01846153846154, | |
| "grad_norm": 0.01966054178774357, | |
| "learning_rate": 2.3931623931623934e-07, | |
| "loss": 0.0004, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 48.02, | |
| "eval_accuracy": 0.7391304347826086, | |
| "eval_loss": 1.392978310585022, | |
| "eval_runtime": 14.2391, | |
| "eval_samples_per_second": 3.231, | |
| "eval_steps_per_second": 0.843, | |
| "step": 3185 | |
| }, | |
| { | |
| "epoch": 49.00153846153846, | |
| "grad_norm": 0.005121205002069473, | |
| "learning_rate": 2.0512820512820514e-07, | |
| "loss": 0.0005, | |
| "step": 3190 | |
| }, | |
| { | |
| "epoch": 49.004615384615384, | |
| "grad_norm": 0.007318329997360706, | |
| "learning_rate": 1.7094017094017097e-07, | |
| "loss": 0.0162, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 49.00769230769231, | |
| "grad_norm": 0.0550755113363266, | |
| "learning_rate": 1.3675213675213677e-07, | |
| "loss": 0.0344, | |
| "step": 3210 | |
| }, | |
| { | |
| "epoch": 49.01076923076923, | |
| "grad_norm": 0.004471645224839449, | |
| "learning_rate": 1.0256410256410257e-07, | |
| "loss": 0.3291, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 49.01384615384615, | |
| "grad_norm": 0.10918418318033218, | |
| "learning_rate": 6.837606837606839e-08, | |
| "loss": 0.0152, | |
| "step": 3230 | |
| }, | |
| { | |
| "epoch": 49.01692307692308, | |
| "grad_norm": 0.005762889981269836, | |
| "learning_rate": 3.418803418803419e-08, | |
| "loss": 0.0006, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "grad_norm": 50.610591888427734, | |
| "learning_rate": 0.0, | |
| "loss": 0.0033, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 1.4388333559036255, | |
| "eval_runtime": 16.7736, | |
| "eval_samples_per_second": 2.742, | |
| "eval_steps_per_second": 0.715, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "step": 3250, | |
| "total_flos": 5.708334209885798e+19, | |
| "train_loss": 0.32200761378957676, | |
| "train_runtime": 8569.5008, | |
| "train_samples_per_second": 1.517, | |
| "train_steps_per_second": 0.379 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "eval_accuracy": 0.782608695652174, | |
| "eval_loss": 1.330747127532959, | |
| "eval_runtime": 14.1891, | |
| "eval_samples_per_second": 3.242, | |
| "eval_steps_per_second": 0.846, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "eval_accuracy": 0.782608695652174, | |
| "eval_loss": 1.3307472467422485, | |
| "eval_runtime": 14.1739, | |
| "eval_samples_per_second": 3.245, | |
| "eval_steps_per_second": 0.847, | |
| "step": 3250 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 3250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.708334209885798e+19, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |