| { | |
| "best_metric": 0.8, | |
| "best_model_checkpoint": "CTMAE-P2-V3-3G-S2/checkpoint-1452", | |
| "epoch": 49.00492307692308, | |
| "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": 1.4794609546661377, | |
| "learning_rate": 3.0769230769230774e-07, | |
| "loss": 0.7229, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.006153846153846154, | |
| "grad_norm": 3.5014188289642334, | |
| "learning_rate": 6.153846153846155e-07, | |
| "loss": 0.7135, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.009230769230769232, | |
| "grad_norm": 2.8026018142700195, | |
| "learning_rate": 9.230769230769232e-07, | |
| "loss": 0.7085, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.012307692307692308, | |
| "grad_norm": 2.06451678276062, | |
| "learning_rate": 1.230769230769231e-06, | |
| "loss": 0.686, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.015384615384615385, | |
| "grad_norm": 2.966813802719116, | |
| "learning_rate": 1.5384615384615387e-06, | |
| "loss": 0.6659, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.018461538461538463, | |
| "grad_norm": 3.384611129760742, | |
| "learning_rate": 1.8461538461538465e-06, | |
| "loss": 0.6262, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.020307692307692308, | |
| "eval_accuracy": 0.4, | |
| "eval_loss": 0.8422228693962097, | |
| "eval_runtime": 16.0183, | |
| "eval_samples_per_second": 2.809, | |
| "eval_steps_per_second": 0.749, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.0012307692307691, | |
| "grad_norm": 2.9542248249053955, | |
| "learning_rate": 2.153846153846154e-06, | |
| "loss": 0.6232, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.0043076923076923, | |
| "grad_norm": 5.368092060089111, | |
| "learning_rate": 2.461538461538462e-06, | |
| "loss": 0.7361, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0073846153846153, | |
| "grad_norm": 6.5317206382751465, | |
| "learning_rate": 2.7692307692307697e-06, | |
| "loss": 0.6857, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0104615384615385, | |
| "grad_norm": 8.286185264587402, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "loss": 0.6384, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0135384615384615, | |
| "grad_norm": 11.786270141601562, | |
| "learning_rate": 3.384615384615385e-06, | |
| "loss": 0.5941, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.0166153846153847, | |
| "grad_norm": 4.752856254577637, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 0.5692, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0196923076923077, | |
| "grad_norm": 2.9264302253723145, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.5961, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0203076923076924, | |
| "eval_accuracy": 0.4, | |
| "eval_loss": 0.917664647102356, | |
| "eval_runtime": 14.9958, | |
| "eval_samples_per_second": 3.001, | |
| "eval_steps_per_second": 0.8, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 2.0024615384615383, | |
| "grad_norm": 5.976866722106934, | |
| "learning_rate": 4.307692307692308e-06, | |
| "loss": 0.7066, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.0055384615384617, | |
| "grad_norm": 3.7721126079559326, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 0.6393, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.0086153846153847, | |
| "grad_norm": 3.9207699298858643, | |
| "learning_rate": 4.923076923076924e-06, | |
| "loss": 0.54, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.0116923076923077, | |
| "grad_norm": 21.63241958618164, | |
| "learning_rate": 5.230769230769232e-06, | |
| "loss": 0.6078, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.0147692307692306, | |
| "grad_norm": 7.049718379974365, | |
| "learning_rate": 5.538461538461539e-06, | |
| "loss": 0.6306, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.017846153846154, | |
| "grad_norm": 4.662971019744873, | |
| "learning_rate": 5.846153846153847e-06, | |
| "loss": 0.6691, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.0203076923076924, | |
| "eval_accuracy": 0.4, | |
| "eval_loss": 0.86167311668396, | |
| "eval_runtime": 15.0869, | |
| "eval_samples_per_second": 2.983, | |
| "eval_steps_per_second": 0.795, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 3.0006153846153847, | |
| "grad_norm": 6.2993035316467285, | |
| "learning_rate": 6.153846153846155e-06, | |
| "loss": 0.6775, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.0036923076923077, | |
| "grad_norm": 5.143687725067139, | |
| "learning_rate": 6.461538461538463e-06, | |
| "loss": 0.542, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.0067692307692306, | |
| "grad_norm": 5.491978645324707, | |
| "learning_rate": 6.76923076923077e-06, | |
| "loss": 0.5436, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.009846153846154, | |
| "grad_norm": 5.525275707244873, | |
| "learning_rate": 7.076923076923078e-06, | |
| "loss": 0.5095, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.012923076923077, | |
| "grad_norm": 5.9389777183532715, | |
| "learning_rate": 7.384615384615386e-06, | |
| "loss": 0.685, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.016, | |
| "grad_norm": 8.1178617477417, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.5544, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.019076923076923, | |
| "grad_norm": 7.101919174194336, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.6242, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.0203076923076924, | |
| "eval_accuracy": 0.4, | |
| "eval_loss": 0.9505811333656311, | |
| "eval_runtime": 14.1232, | |
| "eval_samples_per_second": 3.186, | |
| "eval_steps_per_second": 0.85, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 4.001846153846154, | |
| "grad_norm": 2.8857269287109375, | |
| "learning_rate": 8.307692307692309e-06, | |
| "loss": 0.3628, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.004923076923077, | |
| "grad_norm": 6.900577068328857, | |
| "learning_rate": 8.615384615384617e-06, | |
| "loss": 0.9504, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.008, | |
| "grad_norm": 4.167863368988037, | |
| "learning_rate": 8.923076923076925e-06, | |
| "loss": 0.7059, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.011076923076923, | |
| "grad_norm": 3.5212466716766357, | |
| "learning_rate": 9.230769230769232e-06, | |
| "loss": 0.6493, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.014153846153846, | |
| "grad_norm": 12.61021614074707, | |
| "learning_rate": 9.53846153846154e-06, | |
| "loss": 0.5847, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.017230769230769, | |
| "grad_norm": 4.4552764892578125, | |
| "learning_rate": 9.846153846153848e-06, | |
| "loss": 0.5272, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 4.020307692307692, | |
| "grad_norm": 2.0534627437591553, | |
| "learning_rate": 9.982905982905984e-06, | |
| "loss": 0.631, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.020307692307692, | |
| "eval_accuracy": 0.4, | |
| "eval_loss": 0.7678373456001282, | |
| "eval_runtime": 14.2654, | |
| "eval_samples_per_second": 3.154, | |
| "eval_steps_per_second": 0.841, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 5.003076923076923, | |
| "grad_norm": 9.449999809265137, | |
| "learning_rate": 9.94871794871795e-06, | |
| "loss": 0.4265, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 5.006153846153846, | |
| "grad_norm": 14.458154678344727, | |
| "learning_rate": 9.914529914529915e-06, | |
| "loss": 0.6222, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 5.009230769230769, | |
| "grad_norm": 10.737525939941406, | |
| "learning_rate": 9.880341880341882e-06, | |
| "loss": 0.4503, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 5.012307692307692, | |
| "grad_norm": 17.312559127807617, | |
| "learning_rate": 9.846153846153848e-06, | |
| "loss": 0.4326, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 5.015384615384615, | |
| "grad_norm": 52.598114013671875, | |
| "learning_rate": 9.811965811965812e-06, | |
| "loss": 0.5235, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 5.018461538461539, | |
| "grad_norm": 9.938750267028809, | |
| "learning_rate": 9.777777777777779e-06, | |
| "loss": 0.6747, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 5.020307692307692, | |
| "eval_accuracy": 0.6, | |
| "eval_loss": 0.663204550743103, | |
| "eval_runtime": 14.4114, | |
| "eval_samples_per_second": 3.123, | |
| "eval_steps_per_second": 0.833, | |
| "step": 396 | |
| }, | |
| { | |
| "epoch": 6.001230769230769, | |
| "grad_norm": 39.090858459472656, | |
| "learning_rate": 9.743589743589744e-06, | |
| "loss": 0.4702, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.004307692307692, | |
| "grad_norm": 54.20093536376953, | |
| "learning_rate": 9.70940170940171e-06, | |
| "loss": 0.5863, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 6.007384615384615, | |
| "grad_norm": 19.63410186767578, | |
| "learning_rate": 9.675213675213677e-06, | |
| "loss": 0.383, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 6.010461538461539, | |
| "grad_norm": 29.095314025878906, | |
| "learning_rate": 9.641025641025642e-06, | |
| "loss": 0.7057, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 6.013538461538461, | |
| "grad_norm": 18.241823196411133, | |
| "learning_rate": 9.606837606837607e-06, | |
| "loss": 0.4765, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 6.016615384615385, | |
| "grad_norm": 35.08736038208008, | |
| "learning_rate": 9.572649572649575e-06, | |
| "loss": 0.487, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 6.019692307692308, | |
| "grad_norm": 6.303005218505859, | |
| "learning_rate": 9.53846153846154e-06, | |
| "loss": 0.6345, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 6.020307692307692, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.7090754508972168, | |
| "eval_runtime": 15.0395, | |
| "eval_samples_per_second": 2.992, | |
| "eval_steps_per_second": 0.798, | |
| "step": 462 | |
| }, | |
| { | |
| "epoch": 7.002461538461539, | |
| "grad_norm": 13.644148826599121, | |
| "learning_rate": 9.504273504273504e-06, | |
| "loss": 0.4836, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 7.005538461538461, | |
| "grad_norm": 12.159551620483398, | |
| "learning_rate": 9.470085470085471e-06, | |
| "loss": 0.5005, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 7.008615384615385, | |
| "grad_norm": 2.893991470336914, | |
| "learning_rate": 9.435897435897436e-06, | |
| "loss": 0.2963, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 7.011692307692308, | |
| "grad_norm": 13.471207618713379, | |
| "learning_rate": 9.401709401709402e-06, | |
| "loss": 0.5479, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.014769230769231, | |
| "grad_norm": 14.097221374511719, | |
| "learning_rate": 9.367521367521369e-06, | |
| "loss": 0.5449, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 7.017846153846154, | |
| "grad_norm": 62.047916412353516, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 0.6357, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 7.020307692307692, | |
| "eval_accuracy": 0.5111111111111111, | |
| "eval_loss": 1.096928358078003, | |
| "eval_runtime": 13.707, | |
| "eval_samples_per_second": 3.283, | |
| "eval_steps_per_second": 0.875, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 8.000615384615385, | |
| "grad_norm": 4.9958038330078125, | |
| "learning_rate": 9.2991452991453e-06, | |
| "loss": 0.509, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 8.003692307692308, | |
| "grad_norm": 6.135051727294922, | |
| "learning_rate": 9.264957264957267e-06, | |
| "loss": 0.3708, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 8.006769230769232, | |
| "grad_norm": 28.816497802734375, | |
| "learning_rate": 9.230769230769232e-06, | |
| "loss": 1.3963, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 8.009846153846153, | |
| "grad_norm": 16.59303092956543, | |
| "learning_rate": 9.196581196581196e-06, | |
| "loss": 0.6271, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 8.012923076923077, | |
| "grad_norm": 11.57801628112793, | |
| "learning_rate": 9.162393162393163e-06, | |
| "loss": 0.4178, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 8.016, | |
| "grad_norm": 14.430103302001953, | |
| "learning_rate": 9.128205128205129e-06, | |
| "loss": 0.5318, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 8.019076923076923, | |
| "grad_norm": 12.58498477935791, | |
| "learning_rate": 9.094017094017094e-06, | |
| "loss": 0.4717, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 8.020307692307693, | |
| "eval_accuracy": 0.6444444444444445, | |
| "eval_loss": 0.6718806624412537, | |
| "eval_runtime": 13.729, | |
| "eval_samples_per_second": 3.278, | |
| "eval_steps_per_second": 0.874, | |
| "step": 594 | |
| }, | |
| { | |
| "epoch": 9.001846153846154, | |
| "grad_norm": 9.91074275970459, | |
| "learning_rate": 9.059829059829061e-06, | |
| "loss": 0.3508, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 9.004923076923077, | |
| "grad_norm": 7.515194892883301, | |
| "learning_rate": 9.025641025641027e-06, | |
| "loss": 0.6135, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 9.008, | |
| "grad_norm": 17.925878524780273, | |
| "learning_rate": 8.991452991452992e-06, | |
| "loss": 0.3836, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 9.011076923076923, | |
| "grad_norm": 7.74843692779541, | |
| "learning_rate": 8.957264957264959e-06, | |
| "loss": 0.4139, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 9.014153846153846, | |
| "grad_norm": 4.378147125244141, | |
| "learning_rate": 8.923076923076925e-06, | |
| "loss": 0.2696, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 9.01723076923077, | |
| "grad_norm": 7.040340900421143, | |
| "learning_rate": 8.888888888888888e-06, | |
| "loss": 0.4402, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 9.020307692307693, | |
| "grad_norm": 18.362342834472656, | |
| "learning_rate": 8.854700854700855e-06, | |
| "loss": 0.827, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 9.020307692307693, | |
| "eval_accuracy": 0.7777777777777778, | |
| "eval_loss": 0.5489006638526917, | |
| "eval_runtime": 13.758, | |
| "eval_samples_per_second": 3.271, | |
| "eval_steps_per_second": 0.872, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 10.003076923076923, | |
| "grad_norm": 6.362973690032959, | |
| "learning_rate": 8.820512820512821e-06, | |
| "loss": 0.5009, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 10.006153846153847, | |
| "grad_norm": 7.7518510818481445, | |
| "learning_rate": 8.786324786324786e-06, | |
| "loss": 0.2564, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 10.009230769230768, | |
| "grad_norm": 9.082799911499023, | |
| "learning_rate": 8.752136752136753e-06, | |
| "loss": 0.3174, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 10.012307692307692, | |
| "grad_norm": 15.28465461730957, | |
| "learning_rate": 8.717948717948719e-06, | |
| "loss": 0.5038, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 10.015384615384615, | |
| "grad_norm": 39.25768280029297, | |
| "learning_rate": 8.683760683760684e-06, | |
| "loss": 0.5722, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 10.018461538461539, | |
| "grad_norm": 15.29712963104248, | |
| "learning_rate": 8.649572649572651e-06, | |
| "loss": 0.3107, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 10.020307692307693, | |
| "eval_accuracy": 0.6888888888888889, | |
| "eval_loss": 0.7149955630302429, | |
| "eval_runtime": 11.7609, | |
| "eval_samples_per_second": 3.826, | |
| "eval_steps_per_second": 1.02, | |
| "step": 726 | |
| }, | |
| { | |
| "epoch": 11.00123076923077, | |
| "grad_norm": 4.504512786865234, | |
| "learning_rate": 8.615384615384617e-06, | |
| "loss": 0.4243, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 11.004307692307693, | |
| "grad_norm": 6.083981037139893, | |
| "learning_rate": 8.58119658119658e-06, | |
| "loss": 0.2924, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 11.007384615384616, | |
| "grad_norm": 1.0353838205337524, | |
| "learning_rate": 8.547008547008548e-06, | |
| "loss": 0.5983, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 11.010461538461538, | |
| "grad_norm": 26.63617706298828, | |
| "learning_rate": 8.512820512820513e-06, | |
| "loss": 0.4385, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 11.013538461538461, | |
| "grad_norm": 73.77381134033203, | |
| "learning_rate": 8.478632478632479e-06, | |
| "loss": 0.5767, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 11.016615384615385, | |
| "grad_norm": 5.014594078063965, | |
| "learning_rate": 8.444444444444446e-06, | |
| "loss": 0.3618, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 11.019692307692308, | |
| "grad_norm": 10.796072006225586, | |
| "learning_rate": 8.410256410256411e-06, | |
| "loss": 0.3366, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 11.020307692307693, | |
| "eval_accuracy": 0.7111111111111111, | |
| "eval_loss": 0.7248274087905884, | |
| "eval_runtime": 13.7343, | |
| "eval_samples_per_second": 3.276, | |
| "eval_steps_per_second": 0.874, | |
| "step": 792 | |
| }, | |
| { | |
| "epoch": 12.002461538461539, | |
| "grad_norm": 74.85016632080078, | |
| "learning_rate": 8.376068376068377e-06, | |
| "loss": 0.1946, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 12.005538461538462, | |
| "grad_norm": 26.65357780456543, | |
| "learning_rate": 8.341880341880344e-06, | |
| "loss": 0.2482, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 12.008615384615384, | |
| "grad_norm": 30.542070388793945, | |
| "learning_rate": 8.307692307692309e-06, | |
| "loss": 0.4699, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 12.011692307692307, | |
| "grad_norm": 1.3302332162857056, | |
| "learning_rate": 8.273504273504273e-06, | |
| "loss": 0.1524, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 12.01476923076923, | |
| "grad_norm": 26.43408966064453, | |
| "learning_rate": 8.23931623931624e-06, | |
| "loss": 0.4394, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 12.017846153846154, | |
| "grad_norm": 42.91722869873047, | |
| "learning_rate": 8.205128205128205e-06, | |
| "loss": 0.8919, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 12.020307692307693, | |
| "eval_accuracy": 0.7777777777777778, | |
| "eval_loss": 0.6666911840438843, | |
| "eval_runtime": 13.732, | |
| "eval_samples_per_second": 3.277, | |
| "eval_steps_per_second": 0.874, | |
| "step": 858 | |
| }, | |
| { | |
| "epoch": 13.000615384615385, | |
| "grad_norm": 31.700401306152344, | |
| "learning_rate": 8.17094017094017e-06, | |
| "loss": 0.6633, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 13.003692307692308, | |
| "grad_norm": 0.5030388236045837, | |
| "learning_rate": 8.136752136752138e-06, | |
| "loss": 0.2475, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 13.006769230769232, | |
| "grad_norm": 42.06875991821289, | |
| "learning_rate": 8.102564102564103e-06, | |
| "loss": 0.8717, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 13.009846153846153, | |
| "grad_norm": 33.55354690551758, | |
| "learning_rate": 8.068376068376069e-06, | |
| "loss": 0.5353, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 13.012923076923077, | |
| "grad_norm": 36.84174728393555, | |
| "learning_rate": 8.034188034188036e-06, | |
| "loss": 0.3346, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 13.016, | |
| "grad_norm": 149.687255859375, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.7291, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 13.019076923076923, | |
| "grad_norm": 0.7351716160774231, | |
| "learning_rate": 7.965811965811967e-06, | |
| "loss": 0.4823, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 13.020307692307693, | |
| "eval_accuracy": 0.4222222222222222, | |
| "eval_loss": 2.105036497116089, | |
| "eval_runtime": 15.1948, | |
| "eval_samples_per_second": 2.962, | |
| "eval_steps_per_second": 0.79, | |
| "step": 924 | |
| }, | |
| { | |
| "epoch": 14.001846153846154, | |
| "grad_norm": 35.5843391418457, | |
| "learning_rate": 7.931623931623932e-06, | |
| "loss": 0.5195, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 14.004923076923077, | |
| "grad_norm": 2.578026056289673, | |
| "learning_rate": 7.897435897435898e-06, | |
| "loss": 0.1636, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 14.008, | |
| "grad_norm": 21.901315689086914, | |
| "learning_rate": 7.863247863247863e-06, | |
| "loss": 0.384, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 14.011076923076923, | |
| "grad_norm": 51.293582916259766, | |
| "learning_rate": 7.82905982905983e-06, | |
| "loss": 0.5037, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 14.014153846153846, | |
| "grad_norm": 30.774192810058594, | |
| "learning_rate": 7.794871794871796e-06, | |
| "loss": 0.8682, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 14.01723076923077, | |
| "grad_norm": 31.879384994506836, | |
| "learning_rate": 7.760683760683761e-06, | |
| "loss": 0.3589, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 14.020307692307693, | |
| "grad_norm": 22.549419403076172, | |
| "learning_rate": 7.726495726495728e-06, | |
| "loss": 0.3742, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 14.020307692307693, | |
| "eval_accuracy": 0.5777777777777777, | |
| "eval_loss": 1.0016950368881226, | |
| "eval_runtime": 13.7811, | |
| "eval_samples_per_second": 3.265, | |
| "eval_steps_per_second": 0.871, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 15.003076923076923, | |
| "grad_norm": 0.20596639811992645, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.0428, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 15.006153846153847, | |
| "grad_norm": 12.73334789276123, | |
| "learning_rate": 7.658119658119659e-06, | |
| "loss": 0.4527, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 15.009230769230768, | |
| "grad_norm": 0.39726388454437256, | |
| "learning_rate": 7.6239316239316244e-06, | |
| "loss": 0.6545, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 15.012307692307692, | |
| "grad_norm": 47.79050064086914, | |
| "learning_rate": 7.58974358974359e-06, | |
| "loss": 0.2677, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 15.015384615384615, | |
| "grad_norm": 49.48329544067383, | |
| "learning_rate": 7.555555555555556e-06, | |
| "loss": 0.4924, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 15.018461538461539, | |
| "grad_norm": 23.630064010620117, | |
| "learning_rate": 7.521367521367522e-06, | |
| "loss": 0.3399, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 15.020307692307693, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.5679446458816528, | |
| "eval_runtime": 13.7728, | |
| "eval_samples_per_second": 3.267, | |
| "eval_steps_per_second": 0.871, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 16.00123076923077, | |
| "grad_norm": 1.3588582277297974, | |
| "learning_rate": 7.487179487179488e-06, | |
| "loss": 0.1339, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 16.00430769230769, | |
| "grad_norm": 10.130599021911621, | |
| "learning_rate": 7.452991452991454e-06, | |
| "loss": 0.5679, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 16.007384615384616, | |
| "grad_norm": 1.314917802810669, | |
| "learning_rate": 7.4188034188034196e-06, | |
| "loss": 0.3619, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 16.010461538461538, | |
| "grad_norm": 24.52656364440918, | |
| "learning_rate": 7.384615384615386e-06, | |
| "loss": 0.1272, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 16.013538461538463, | |
| "grad_norm": 33.17477798461914, | |
| "learning_rate": 7.350427350427351e-06, | |
| "loss": 0.26, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 16.016615384615385, | |
| "grad_norm": 30.977619171142578, | |
| "learning_rate": 7.316239316239317e-06, | |
| "loss": 0.7437, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 16.019692307692306, | |
| "grad_norm": 1.8153678178787231, | |
| "learning_rate": 7.282051282051282e-06, | |
| "loss": 0.6571, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 16.020307692307693, | |
| "eval_accuracy": 0.6, | |
| "eval_loss": 1.352131962776184, | |
| "eval_runtime": 13.7467, | |
| "eval_samples_per_second": 3.274, | |
| "eval_steps_per_second": 0.873, | |
| "step": 1122 | |
| }, | |
| { | |
| "epoch": 17.00246153846154, | |
| "grad_norm": 47.651878356933594, | |
| "learning_rate": 7.247863247863248e-06, | |
| "loss": 0.1583, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 17.00553846153846, | |
| "grad_norm": 7.4866766929626465, | |
| "learning_rate": 7.213675213675214e-06, | |
| "loss": 0.3367, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 17.008615384615386, | |
| "grad_norm": 61.95333480834961, | |
| "learning_rate": 7.17948717948718e-06, | |
| "loss": 0.5305, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 17.011692307692307, | |
| "grad_norm": 0.09725925326347351, | |
| "learning_rate": 7.145299145299146e-06, | |
| "loss": 0.2062, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 17.014769230769232, | |
| "grad_norm": 27.84596824645996, | |
| "learning_rate": 7.111111111111112e-06, | |
| "loss": 0.5889, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 17.017846153846154, | |
| "grad_norm": 0.5110335350036621, | |
| "learning_rate": 7.076923076923078e-06, | |
| "loss": 0.2434, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 17.020307692307693, | |
| "eval_accuracy": 0.7777777777777778, | |
| "eval_loss": 0.7811753749847412, | |
| "eval_runtime": 13.7851, | |
| "eval_samples_per_second": 3.264, | |
| "eval_steps_per_second": 0.871, | |
| "step": 1188 | |
| }, | |
| { | |
| "epoch": 18.000615384615383, | |
| "grad_norm": 4.819025993347168, | |
| "learning_rate": 7.0427350427350435e-06, | |
| "loss": 0.5103, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 18.003692307692308, | |
| "grad_norm": 0.03198942542076111, | |
| "learning_rate": 7.008547008547009e-06, | |
| "loss": 0.2102, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 18.00676923076923, | |
| "grad_norm": 53.40505599975586, | |
| "learning_rate": 6.974358974358974e-06, | |
| "loss": 0.7028, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 18.009846153846155, | |
| "grad_norm": 0.3131262958049774, | |
| "learning_rate": 6.940170940170941e-06, | |
| "loss": 0.103, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 18.012923076923077, | |
| "grad_norm": 0.027441751211881638, | |
| "learning_rate": 6.905982905982906e-06, | |
| "loss": 0.3159, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 18.016, | |
| "grad_norm": 0.5056867599487305, | |
| "learning_rate": 6.871794871794872e-06, | |
| "loss": 0.1821, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 18.019076923076923, | |
| "grad_norm": 0.06143081188201904, | |
| "learning_rate": 6.837606837606839e-06, | |
| "loss": 0.5967, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 18.020307692307693, | |
| "eval_accuracy": 0.6888888888888889, | |
| "eval_loss": 0.9575199484825134, | |
| "eval_runtime": 13.783, | |
| "eval_samples_per_second": 3.265, | |
| "eval_steps_per_second": 0.871, | |
| "step": 1254 | |
| }, | |
| { | |
| "epoch": 19.001846153846152, | |
| "grad_norm": 68.97899627685547, | |
| "learning_rate": 6.803418803418804e-06, | |
| "loss": 0.5041, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 19.004923076923077, | |
| "grad_norm": 0.17927193641662598, | |
| "learning_rate": 6.76923076923077e-06, | |
| "loss": 0.152, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 19.008, | |
| "grad_norm": 63.69683074951172, | |
| "learning_rate": 6.735042735042736e-06, | |
| "loss": 0.2777, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 19.011076923076924, | |
| "grad_norm": 0.1938479095697403, | |
| "learning_rate": 6.700854700854701e-06, | |
| "loss": 0.5759, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 19.014153846153846, | |
| "grad_norm": 4.396932125091553, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.0564, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 19.017230769230768, | |
| "grad_norm": 0.019336098805069923, | |
| "learning_rate": 6.632478632478633e-06, | |
| "loss": 0.4901, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 19.020307692307693, | |
| "grad_norm": 0.24376414716243744, | |
| "learning_rate": 6.598290598290598e-06, | |
| "loss": 0.1982, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 19.020307692307693, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 1.1720720529556274, | |
| "eval_runtime": 13.8989, | |
| "eval_samples_per_second": 3.238, | |
| "eval_steps_per_second": 0.863, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 20.00307692307692, | |
| "grad_norm": 6.922430992126465, | |
| "learning_rate": 6.564102564102565e-06, | |
| "loss": 0.144, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 20.006153846153847, | |
| "grad_norm": 14.44478702545166, | |
| "learning_rate": 6.529914529914531e-06, | |
| "loss": 0.1007, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 20.00923076923077, | |
| "grad_norm": 95.26171112060547, | |
| "learning_rate": 6.495726495726496e-06, | |
| "loss": 0.4298, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 20.012307692307694, | |
| "grad_norm": 110.86443328857422, | |
| "learning_rate": 6.461538461538463e-06, | |
| "loss": 0.1362, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 20.015384615384615, | |
| "grad_norm": 0.13159264624118805, | |
| "learning_rate": 6.427350427350428e-06, | |
| "loss": 0.086, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 20.018461538461537, | |
| "grad_norm": 116.35437774658203, | |
| "learning_rate": 6.3931623931623935e-06, | |
| "loss": 0.2631, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 20.020307692307693, | |
| "eval_accuracy": 0.4666666666666667, | |
| "eval_loss": 2.5732531547546387, | |
| "eval_runtime": 15.2291, | |
| "eval_samples_per_second": 2.955, | |
| "eval_steps_per_second": 0.788, | |
| "step": 1386 | |
| }, | |
| { | |
| "epoch": 21.00123076923077, | |
| "grad_norm": 0.08772929012775421, | |
| "learning_rate": 6.358974358974359e-06, | |
| "loss": 0.3654, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 21.00430769230769, | |
| "grad_norm": 0.5012686848640442, | |
| "learning_rate": 6.324786324786325e-06, | |
| "loss": 0.3005, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 21.007384615384616, | |
| "grad_norm": 0.3065190017223358, | |
| "learning_rate": 6.290598290598291e-06, | |
| "loss": 0.2305, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 21.010461538461538, | |
| "grad_norm": 0.326849102973938, | |
| "learning_rate": 6.256410256410257e-06, | |
| "loss": 0.2465, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 21.013538461538463, | |
| "grad_norm": 58.84406661987305, | |
| "learning_rate": 6.222222222222223e-06, | |
| "loss": 0.1797, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 21.016615384615385, | |
| "grad_norm": 0.0164014995098114, | |
| "learning_rate": 6.188034188034189e-06, | |
| "loss": 0.1588, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 21.019692307692306, | |
| "grad_norm": 34.90568923950195, | |
| "learning_rate": 6.153846153846155e-06, | |
| "loss": 0.3235, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 21.020307692307693, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.9770981669425964, | |
| "eval_runtime": 13.7436, | |
| "eval_samples_per_second": 3.274, | |
| "eval_steps_per_second": 0.873, | |
| "step": 1452 | |
| }, | |
| { | |
| "epoch": 22.00246153846154, | |
| "grad_norm": 0.055868543684482574, | |
| "learning_rate": 6.11965811965812e-06, | |
| "loss": 0.0287, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 22.00553846153846, | |
| "grad_norm": 205.71531677246094, | |
| "learning_rate": 6.085470085470086e-06, | |
| "loss": 0.3236, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 22.008615384615386, | |
| "grad_norm": 0.009756018407642841, | |
| "learning_rate": 6.051282051282051e-06, | |
| "loss": 0.311, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 22.011692307692307, | |
| "grad_norm": 125.73477172851562, | |
| "learning_rate": 6.0170940170940174e-06, | |
| "loss": 0.6165, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 22.014769230769232, | |
| "grad_norm": 0.32701388001441956, | |
| "learning_rate": 5.982905982905983e-06, | |
| "loss": 0.0298, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 22.017846153846154, | |
| "grad_norm": 87.97734069824219, | |
| "learning_rate": 5.948717948717949e-06, | |
| "loss": 0.1786, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 22.020307692307693, | |
| "eval_accuracy": 0.7111111111111111, | |
| "eval_loss": 1.1978042125701904, | |
| "eval_runtime": 13.8304, | |
| "eval_samples_per_second": 3.254, | |
| "eval_steps_per_second": 0.868, | |
| "step": 1518 | |
| }, | |
| { | |
| "epoch": 23.000615384615383, | |
| "grad_norm": 104.00849151611328, | |
| "learning_rate": 5.914529914529915e-06, | |
| "loss": 0.1849, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 23.003692307692308, | |
| "grad_norm": 5.975285053253174, | |
| "learning_rate": 5.880341880341881e-06, | |
| "loss": 0.0709, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 23.00676923076923, | |
| "grad_norm": 0.09302225708961487, | |
| "learning_rate": 5.846153846153847e-06, | |
| "loss": 0.2343, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 23.009846153846155, | |
| "grad_norm": 113.04926300048828, | |
| "learning_rate": 5.8119658119658126e-06, | |
| "loss": 0.506, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 23.012923076923077, | |
| "grad_norm": 9.404973983764648, | |
| "learning_rate": 5.777777777777778e-06, | |
| "loss": 0.3141, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 23.016, | |
| "grad_norm": 0.007651908788830042, | |
| "learning_rate": 5.743589743589743e-06, | |
| "loss": 0.7863, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 23.019076923076923, | |
| "grad_norm": 3.120847225189209, | |
| "learning_rate": 5.70940170940171e-06, | |
| "loss": 0.1352, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 23.020307692307693, | |
| "eval_accuracy": 0.7777777777777778, | |
| "eval_loss": 0.86919766664505, | |
| "eval_runtime": 15.1646, | |
| "eval_samples_per_second": 2.967, | |
| "eval_steps_per_second": 0.791, | |
| "step": 1584 | |
| }, | |
| { | |
| "epoch": 24.001846153846152, | |
| "grad_norm": 0.05145508795976639, | |
| "learning_rate": 5.675213675213675e-06, | |
| "loss": 0.3145, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 24.004923076923077, | |
| "grad_norm": 44.031471252441406, | |
| "learning_rate": 5.641025641025641e-06, | |
| "loss": 0.4755, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 24.008, | |
| "grad_norm": 1.2100696563720703, | |
| "learning_rate": 5.606837606837608e-06, | |
| "loss": 0.2257, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 24.011076923076924, | |
| "grad_norm": 38.097190856933594, | |
| "learning_rate": 5.572649572649573e-06, | |
| "loss": 0.1551, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 24.014153846153846, | |
| "grad_norm": 4.676908016204834, | |
| "learning_rate": 5.538461538461539e-06, | |
| "loss": 0.1355, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 24.017230769230768, | |
| "grad_norm": 45.96378707885742, | |
| "learning_rate": 5.504273504273505e-06, | |
| "loss": 0.1754, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 24.020307692307693, | |
| "grad_norm": 0.0224474985152483, | |
| "learning_rate": 5.470085470085471e-06, | |
| "loss": 0.1709, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 24.020307692307693, | |
| "eval_accuracy": 0.7555555555555555, | |
| "eval_loss": 1.1424057483673096, | |
| "eval_runtime": 13.7879, | |
| "eval_samples_per_second": 3.264, | |
| "eval_steps_per_second": 0.87, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 25.00307692307692, | |
| "grad_norm": 0.019953293725848198, | |
| "learning_rate": 5.435897435897436e-06, | |
| "loss": 0.0806, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 25.006153846153847, | |
| "grad_norm": 0.046231500804424286, | |
| "learning_rate": 5.401709401709402e-06, | |
| "loss": 0.0549, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 25.00923076923077, | |
| "grad_norm": 0.0693158209323883, | |
| "learning_rate": 5.367521367521367e-06, | |
| "loss": 0.3095, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 25.012307692307694, | |
| "grad_norm": 20.43744468688965, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 0.1346, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 25.015384615384615, | |
| "grad_norm": 0.02532690390944481, | |
| "learning_rate": 5.2991452991453e-06, | |
| "loss": 0.3013, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 25.018461538461537, | |
| "grad_norm": 0.1781366467475891, | |
| "learning_rate": 5.264957264957265e-06, | |
| "loss": 0.0873, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 25.020307692307693, | |
| "eval_accuracy": 0.6222222222222222, | |
| "eval_loss": 1.8760292530059814, | |
| "eval_runtime": 13.807, | |
| "eval_samples_per_second": 3.259, | |
| "eval_steps_per_second": 0.869, | |
| "step": 1716 | |
| }, | |
| { | |
| "epoch": 26.00123076923077, | |
| "grad_norm": 2.7792463302612305, | |
| "learning_rate": 5.230769230769232e-06, | |
| "loss": 0.1224, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 26.00430769230769, | |
| "grad_norm": 87.59558868408203, | |
| "learning_rate": 5.196581196581197e-06, | |
| "loss": 0.2843, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 26.007384615384616, | |
| "grad_norm": 8.613319396972656, | |
| "learning_rate": 5.162393162393163e-06, | |
| "loss": 0.0209, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 26.010461538461538, | |
| "grad_norm": 0.06284106522798538, | |
| "learning_rate": 5.128205128205128e-06, | |
| "loss": 0.0257, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 26.013538461538463, | |
| "grad_norm": 9.368754386901855, | |
| "learning_rate": 5.094017094017094e-06, | |
| "loss": 0.1877, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 26.016615384615385, | |
| "grad_norm": 0.07090014964342117, | |
| "learning_rate": 5.05982905982906e-06, | |
| "loss": 0.1936, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 26.019692307692306, | |
| "grad_norm": 22.935270309448242, | |
| "learning_rate": 5.025641025641026e-06, | |
| "loss": 0.1418, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 26.020307692307693, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 1.0963565111160278, | |
| "eval_runtime": 14.0498, | |
| "eval_samples_per_second": 3.203, | |
| "eval_steps_per_second": 0.854, | |
| "step": 1782 | |
| }, | |
| { | |
| "epoch": 27.00246153846154, | |
| "grad_norm": 0.096647247672081, | |
| "learning_rate": 4.991452991452992e-06, | |
| "loss": 0.0147, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 27.00553846153846, | |
| "grad_norm": 0.10912120342254639, | |
| "learning_rate": 4.957264957264958e-06, | |
| "loss": 0.1364, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 27.008615384615386, | |
| "grad_norm": 97.78022766113281, | |
| "learning_rate": 4.923076923076924e-06, | |
| "loss": 0.2783, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 27.011692307692307, | |
| "grad_norm": 168.70310974121094, | |
| "learning_rate": 4.888888888888889e-06, | |
| "loss": 0.2077, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 27.014769230769232, | |
| "grad_norm": 0.3277463912963867, | |
| "learning_rate": 4.854700854700855e-06, | |
| "loss": 0.003, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 27.017846153846154, | |
| "grad_norm": 0.03846672177314758, | |
| "learning_rate": 4.820512820512821e-06, | |
| "loss": 0.0075, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 27.020307692307693, | |
| "eval_accuracy": 0.6222222222222222, | |
| "eval_loss": 1.9129718542099, | |
| "eval_runtime": 15.1161, | |
| "eval_samples_per_second": 2.977, | |
| "eval_steps_per_second": 0.794, | |
| "step": 1848 | |
| }, | |
| { | |
| "epoch": 28.000615384615383, | |
| "grad_norm": 9.841035842895508, | |
| "learning_rate": 4.786324786324787e-06, | |
| "loss": 0.0969, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 28.003692307692308, | |
| "grad_norm": 37.528804779052734, | |
| "learning_rate": 4.752136752136752e-06, | |
| "loss": 0.2131, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 28.00676923076923, | |
| "grad_norm": 0.033336490392684937, | |
| "learning_rate": 4.717948717948718e-06, | |
| "loss": 0.1671, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 28.009846153846155, | |
| "grad_norm": 17.54317855834961, | |
| "learning_rate": 4.6837606837606844e-06, | |
| "loss": 0.2178, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 28.012923076923077, | |
| "grad_norm": 0.035124845802783966, | |
| "learning_rate": 4.64957264957265e-06, | |
| "loss": 0.0193, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 28.016, | |
| "grad_norm": 0.01775071956217289, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 0.1719, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 28.019076923076923, | |
| "grad_norm": 18.559314727783203, | |
| "learning_rate": 4.581196581196582e-06, | |
| "loss": 0.4534, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 28.020307692307693, | |
| "eval_accuracy": 0.7777777777777778, | |
| "eval_loss": 1.1176104545593262, | |
| "eval_runtime": 13.7703, | |
| "eval_samples_per_second": 3.268, | |
| "eval_steps_per_second": 0.871, | |
| "step": 1914 | |
| }, | |
| { | |
| "epoch": 29.001846153846152, | |
| "grad_norm": 0.03852153941988945, | |
| "learning_rate": 4.547008547008547e-06, | |
| "loss": 0.0281, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 29.004923076923077, | |
| "grad_norm": 0.040109530091285706, | |
| "learning_rate": 4.512820512820513e-06, | |
| "loss": 0.0877, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 29.008, | |
| "grad_norm": 0.011689484119415283, | |
| "learning_rate": 4.4786324786324796e-06, | |
| "loss": 0.0611, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 29.011076923076924, | |
| "grad_norm": 0.053006622940301895, | |
| "learning_rate": 4.444444444444444e-06, | |
| "loss": 0.1979, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 29.014153846153846, | |
| "grad_norm": 0.04539922997355461, | |
| "learning_rate": 4.4102564102564104e-06, | |
| "loss": 0.218, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 29.017230769230768, | |
| "grad_norm": 0.6297591328620911, | |
| "learning_rate": 4.376068376068377e-06, | |
| "loss": 0.3363, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 29.020307692307693, | |
| "grad_norm": 0.01002031471580267, | |
| "learning_rate": 4.341880341880342e-06, | |
| "loss": 0.0019, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 29.020307692307693, | |
| "eval_accuracy": 0.6222222222222222, | |
| "eval_loss": 2.0683937072753906, | |
| "eval_runtime": 13.7766, | |
| "eval_samples_per_second": 3.266, | |
| "eval_steps_per_second": 0.871, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 30.00307692307692, | |
| "grad_norm": 8.234089851379395, | |
| "learning_rate": 4.307692307692308e-06, | |
| "loss": 0.1411, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 30.006153846153847, | |
| "grad_norm": 0.18931716680526733, | |
| "learning_rate": 4.273504273504274e-06, | |
| "loss": 0.1407, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 30.00923076923077, | |
| "grad_norm": 0.020825600251555443, | |
| "learning_rate": 4.239316239316239e-06, | |
| "loss": 0.0226, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 30.012307692307694, | |
| "grad_norm": 0.09039674699306488, | |
| "learning_rate": 4.2051282051282055e-06, | |
| "loss": 0.0008, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 30.015384615384615, | |
| "grad_norm": 0.036115147173404694, | |
| "learning_rate": 4.170940170940172e-06, | |
| "loss": 0.0254, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 30.018461538461537, | |
| "grad_norm": 18.284534454345703, | |
| "learning_rate": 4.136752136752136e-06, | |
| "loss": 0.4098, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 30.020307692307693, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 1.9198042154312134, | |
| "eval_runtime": 13.8755, | |
| "eval_samples_per_second": 3.243, | |
| "eval_steps_per_second": 0.865, | |
| "step": 2046 | |
| }, | |
| { | |
| "epoch": 31.00123076923077, | |
| "grad_norm": 0.01804875209927559, | |
| "learning_rate": 4.102564102564103e-06, | |
| "loss": 0.1426, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 31.00430769230769, | |
| "grad_norm": 0.08219186961650848, | |
| "learning_rate": 4.068376068376069e-06, | |
| "loss": 0.0015, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 31.007384615384616, | |
| "grad_norm": 0.007714936509728432, | |
| "learning_rate": 4.034188034188034e-06, | |
| "loss": 0.0082, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 31.010461538461538, | |
| "grad_norm": 32.443336486816406, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.0092, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 31.013538461538463, | |
| "grad_norm": 183.87896728515625, | |
| "learning_rate": 3.965811965811966e-06, | |
| "loss": 0.2114, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 31.016615384615385, | |
| "grad_norm": 0.09029895812273026, | |
| "learning_rate": 3.9316239316239315e-06, | |
| "loss": 0.0216, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 31.019692307692306, | |
| "grad_norm": 0.007797298952937126, | |
| "learning_rate": 3.897435897435898e-06, | |
| "loss": 0.0006, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 31.020307692307693, | |
| "eval_accuracy": 0.7333333333333333, | |
| "eval_loss": 1.2723807096481323, | |
| "eval_runtime": 14.8246, | |
| "eval_samples_per_second": 3.035, | |
| "eval_steps_per_second": 0.809, | |
| "step": 2112 | |
| }, | |
| { | |
| "epoch": 32.00246153846154, | |
| "grad_norm": 1.2352715730667114, | |
| "learning_rate": 3.863247863247864e-06, | |
| "loss": 0.2567, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 32.005538461538464, | |
| "grad_norm": 1.0585654973983765, | |
| "learning_rate": 3.8290598290598295e-06, | |
| "loss": 0.06, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 32.00861538461538, | |
| "grad_norm": 0.025132305920124054, | |
| "learning_rate": 3.794871794871795e-06, | |
| "loss": 0.1982, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 32.01169230769231, | |
| "grad_norm": 15.088484764099121, | |
| "learning_rate": 3.760683760683761e-06, | |
| "loss": 0.3051, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 32.01476923076923, | |
| "grad_norm": 0.003967353142797947, | |
| "learning_rate": 3.726495726495727e-06, | |
| "loss": 0.0105, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 32.01784615384615, | |
| "grad_norm": 0.07517848908901215, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 0.1891, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 32.02030769230769, | |
| "eval_accuracy": 0.6444444444444445, | |
| "eval_loss": 1.821338176727295, | |
| "eval_runtime": 14.7875, | |
| "eval_samples_per_second": 3.043, | |
| "eval_steps_per_second": 0.811, | |
| "step": 2178 | |
| }, | |
| { | |
| "epoch": 33.00061538461539, | |
| "grad_norm": 0.453707754611969, | |
| "learning_rate": 3.6581196581196584e-06, | |
| "loss": 0.251, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 33.003692307692305, | |
| "grad_norm": 0.01656286045908928, | |
| "learning_rate": 3.623931623931624e-06, | |
| "loss": 0.1222, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 33.00676923076923, | |
| "grad_norm": 0.21087659895420074, | |
| "learning_rate": 3.58974358974359e-06, | |
| "loss": 0.0307, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 33.009846153846155, | |
| "grad_norm": 0.017987869679927826, | |
| "learning_rate": 3.555555555555556e-06, | |
| "loss": 0.1697, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 33.01292307692308, | |
| "grad_norm": 0.5001537203788757, | |
| "learning_rate": 3.5213675213675218e-06, | |
| "loss": 0.0074, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 33.016, | |
| "grad_norm": 0.023721082136034966, | |
| "learning_rate": 3.487179487179487e-06, | |
| "loss": 0.0803, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 33.01907692307692, | |
| "grad_norm": 0.03775130957365036, | |
| "learning_rate": 3.452991452991453e-06, | |
| "loss": 0.0604, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 33.02030769230769, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 2.6845359802246094, | |
| "eval_runtime": 14.7376, | |
| "eval_samples_per_second": 3.053, | |
| "eval_steps_per_second": 0.814, | |
| "step": 2244 | |
| }, | |
| { | |
| "epoch": 34.00184615384615, | |
| "grad_norm": 0.006655918899923563, | |
| "learning_rate": 3.4188034188034193e-06, | |
| "loss": 0.2833, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 34.00492307692308, | |
| "grad_norm": 0.013796042650938034, | |
| "learning_rate": 3.384615384615385e-06, | |
| "loss": 0.0196, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 34.008, | |
| "grad_norm": 0.05266215279698372, | |
| "learning_rate": 3.3504273504273506e-06, | |
| "loss": 0.0876, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 34.01107692307692, | |
| "grad_norm": 0.45490553975105286, | |
| "learning_rate": 3.3162393162393165e-06, | |
| "loss": 0.0007, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 34.014153846153846, | |
| "grad_norm": 0.12290433049201965, | |
| "learning_rate": 3.2820512820512823e-06, | |
| "loss": 0.1642, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 34.01723076923077, | |
| "grad_norm": 1.6173644065856934, | |
| "learning_rate": 3.247863247863248e-06, | |
| "loss": 0.1504, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 34.02030769230769, | |
| "grad_norm": 0.007919330149888992, | |
| "learning_rate": 3.213675213675214e-06, | |
| "loss": 0.104, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 34.02030769230769, | |
| "eval_accuracy": 0.6222222222222222, | |
| "eval_loss": 2.7467894554138184, | |
| "eval_runtime": 14.7559, | |
| "eval_samples_per_second": 3.05, | |
| "eval_steps_per_second": 0.813, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 35.003076923076925, | |
| "grad_norm": 0.0022740329150110483, | |
| "learning_rate": 3.1794871794871795e-06, | |
| "loss": 0.0003, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 35.00615384615384, | |
| "grad_norm": 0.16805915534496307, | |
| "learning_rate": 3.1452991452991453e-06, | |
| "loss": 0.001, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 35.00923076923077, | |
| "grad_norm": 0.01777742989361286, | |
| "learning_rate": 3.1111111111111116e-06, | |
| "loss": 0.039, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 35.012307692307694, | |
| "grad_norm": 6.031890869140625, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "loss": 0.0026, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 35.01538461538462, | |
| "grad_norm": 17.243139266967773, | |
| "learning_rate": 3.042735042735043e-06, | |
| "loss": 0.0728, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 35.01846153846154, | |
| "grad_norm": 0.01590224727988243, | |
| "learning_rate": 3.0085470085470087e-06, | |
| "loss": 0.028, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 35.02030769230769, | |
| "eval_accuracy": 0.7333333333333333, | |
| "eval_loss": 1.4457882642745972, | |
| "eval_runtime": 13.767, | |
| "eval_samples_per_second": 3.269, | |
| "eval_steps_per_second": 0.872, | |
| "step": 2376 | |
| }, | |
| { | |
| "epoch": 36.001230769230766, | |
| "grad_norm": 256.8249816894531, | |
| "learning_rate": 2.9743589743589746e-06, | |
| "loss": 0.1384, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 36.00430769230769, | |
| "grad_norm": 0.012335916049778461, | |
| "learning_rate": 2.9401709401709404e-06, | |
| "loss": 0.3509, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 36.007384615384616, | |
| "grad_norm": 0.01209163572639227, | |
| "learning_rate": 2.9059829059829063e-06, | |
| "loss": 0.0437, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 36.01046153846154, | |
| "grad_norm": 0.15365378558635712, | |
| "learning_rate": 2.8717948717948717e-06, | |
| "loss": 0.1917, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 36.01353846153846, | |
| "grad_norm": 0.006477895192801952, | |
| "learning_rate": 2.8376068376068376e-06, | |
| "loss": 0.1432, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 36.016615384615385, | |
| "grad_norm": 0.006202262826263905, | |
| "learning_rate": 2.803418803418804e-06, | |
| "loss": 0.2241, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 36.01969230769231, | |
| "grad_norm": 0.0072620464488863945, | |
| "learning_rate": 2.7692307692307697e-06, | |
| "loss": 0.2479, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 36.02030769230769, | |
| "eval_accuracy": 0.6222222222222222, | |
| "eval_loss": 2.161320447921753, | |
| "eval_runtime": 13.7692, | |
| "eval_samples_per_second": 3.268, | |
| "eval_steps_per_second": 0.872, | |
| "step": 2442 | |
| }, | |
| { | |
| "epoch": 37.00246153846154, | |
| "grad_norm": 0.007183571346104145, | |
| "learning_rate": 2.7350427350427355e-06, | |
| "loss": 0.0003, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 37.005538461538464, | |
| "grad_norm": 0.7165050506591797, | |
| "learning_rate": 2.700854700854701e-06, | |
| "loss": 0.0026, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 37.00861538461538, | |
| "grad_norm": 0.08769255876541138, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 0.0422, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 37.01169230769231, | |
| "grad_norm": 0.17536067962646484, | |
| "learning_rate": 2.6324786324786327e-06, | |
| "loss": 0.0004, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 37.01476923076923, | |
| "grad_norm": 0.6829420924186707, | |
| "learning_rate": 2.5982905982905985e-06, | |
| "loss": 0.0898, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 37.01784615384615, | |
| "grad_norm": 0.5209939479827881, | |
| "learning_rate": 2.564102564102564e-06, | |
| "loss": 0.0967, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 37.02030769230769, | |
| "eval_accuracy": 0.7777777777777778, | |
| "eval_loss": 1.3894845247268677, | |
| "eval_runtime": 13.7479, | |
| "eval_samples_per_second": 3.273, | |
| "eval_steps_per_second": 0.873, | |
| "step": 2508 | |
| }, | |
| { | |
| "epoch": 38.00061538461539, | |
| "grad_norm": 75.50067901611328, | |
| "learning_rate": 2.52991452991453e-06, | |
| "loss": 0.0198, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 38.003692307692305, | |
| "grad_norm": 0.04112862050533295, | |
| "learning_rate": 2.495726495726496e-06, | |
| "loss": 0.2984, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 38.00676923076923, | |
| "grad_norm": 0.005224079359322786, | |
| "learning_rate": 2.461538461538462e-06, | |
| "loss": 0.0146, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 38.009846153846155, | |
| "grad_norm": 0.00474022189155221, | |
| "learning_rate": 2.4273504273504274e-06, | |
| "loss": 0.0487, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 38.01292307692308, | |
| "grad_norm": 1.4753004312515259, | |
| "learning_rate": 2.3931623931623937e-06, | |
| "loss": 0.0005, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 38.016, | |
| "grad_norm": 0.0049695721827447414, | |
| "learning_rate": 2.358974358974359e-06, | |
| "loss": 0.0004, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 38.01907692307692, | |
| "grad_norm": 117.37168884277344, | |
| "learning_rate": 2.324786324786325e-06, | |
| "loss": 0.0668, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 38.02030769230769, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 2.014683246612549, | |
| "eval_runtime": 13.7937, | |
| "eval_samples_per_second": 3.262, | |
| "eval_steps_per_second": 0.87, | |
| "step": 2574 | |
| }, | |
| { | |
| "epoch": 39.00184615384615, | |
| "grad_norm": 0.007441969588398933, | |
| "learning_rate": 2.290598290598291e-06, | |
| "loss": 0.062, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 39.00492307692308, | |
| "grad_norm": 16.319353103637695, | |
| "learning_rate": 2.2564102564102566e-06, | |
| "loss": 0.0194, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 39.008, | |
| "grad_norm": 180.9895477294922, | |
| "learning_rate": 2.222222222222222e-06, | |
| "loss": 0.1649, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 39.01107692307692, | |
| "grad_norm": 0.0043008350767195225, | |
| "learning_rate": 2.1880341880341884e-06, | |
| "loss": 0.0003, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 39.014153846153846, | |
| "grad_norm": 0.0688568577170372, | |
| "learning_rate": 2.153846153846154e-06, | |
| "loss": 0.0141, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 39.01723076923077, | |
| "grad_norm": 1.9923820495605469, | |
| "learning_rate": 2.1196581196581196e-06, | |
| "loss": 0.0107, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 39.02030769230769, | |
| "grad_norm": 0.002165107289329171, | |
| "learning_rate": 2.085470085470086e-06, | |
| "loss": 0.0004, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 39.02030769230769, | |
| "eval_accuracy": 0.6888888888888889, | |
| "eval_loss": 1.5766425132751465, | |
| "eval_runtime": 13.7894, | |
| "eval_samples_per_second": 3.263, | |
| "eval_steps_per_second": 0.87, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 40.003076923076925, | |
| "grad_norm": 0.022886795923113823, | |
| "learning_rate": 2.0512820512820513e-06, | |
| "loss": 0.0006, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 40.00615384615384, | |
| "grad_norm": 0.007079885341227055, | |
| "learning_rate": 2.017094017094017e-06, | |
| "loss": 0.0003, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 40.00923076923077, | |
| "grad_norm": 0.005172636825591326, | |
| "learning_rate": 1.982905982905983e-06, | |
| "loss": 0.1478, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 40.012307692307694, | |
| "grad_norm": 0.003111331257969141, | |
| "learning_rate": 1.948717948717949e-06, | |
| "loss": 0.154, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 40.01538461538462, | |
| "grad_norm": 0.0919618308544159, | |
| "learning_rate": 1.9145299145299148e-06, | |
| "loss": 0.0004, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 40.01846153846154, | |
| "grad_norm": 0.004653541836887598, | |
| "learning_rate": 1.8803418803418804e-06, | |
| "loss": 0.0027, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 40.02030769230769, | |
| "eval_accuracy": 0.6444444444444445, | |
| "eval_loss": 2.35329532623291, | |
| "eval_runtime": 13.9233, | |
| "eval_samples_per_second": 3.232, | |
| "eval_steps_per_second": 0.862, | |
| "step": 2706 | |
| }, | |
| { | |
| "epoch": 41.001230769230766, | |
| "grad_norm": 0.007990103214979172, | |
| "learning_rate": 1.8461538461538465e-06, | |
| "loss": 0.166, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 41.00430769230769, | |
| "grad_norm": 0.31218963861465454, | |
| "learning_rate": 1.811965811965812e-06, | |
| "loss": 0.0003, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 41.007384615384616, | |
| "grad_norm": 0.020261919125914574, | |
| "learning_rate": 1.777777777777778e-06, | |
| "loss": 0.005, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 41.01046153846154, | |
| "grad_norm": 0.08753982931375504, | |
| "learning_rate": 1.7435897435897436e-06, | |
| "loss": 0.0001, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 41.01353846153846, | |
| "grad_norm": 0.0024725000839680433, | |
| "learning_rate": 1.7094017094017097e-06, | |
| "loss": 0.0003, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 41.016615384615385, | |
| "grad_norm": 0.44845858216285706, | |
| "learning_rate": 1.6752136752136753e-06, | |
| "loss": 0.0003, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 41.01969230769231, | |
| "grad_norm": 0.10153026133775711, | |
| "learning_rate": 1.6410256410256412e-06, | |
| "loss": 0.1436, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 41.02030769230769, | |
| "eval_accuracy": 0.6444444444444445, | |
| "eval_loss": 2.149592161178589, | |
| "eval_runtime": 14.0747, | |
| "eval_samples_per_second": 3.197, | |
| "eval_steps_per_second": 0.853, | |
| "step": 2772 | |
| }, | |
| { | |
| "epoch": 42.00246153846154, | |
| "grad_norm": 0.0065581537783145905, | |
| "learning_rate": 1.606837606837607e-06, | |
| "loss": 0.1022, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 42.005538461538464, | |
| "grad_norm": 0.010586395859718323, | |
| "learning_rate": 1.5726495726495727e-06, | |
| "loss": 0.0034, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 42.00861538461538, | |
| "grad_norm": 0.012179357931017876, | |
| "learning_rate": 1.5384615384615387e-06, | |
| "loss": 0.0002, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 42.01169230769231, | |
| "grad_norm": 0.0030303443782031536, | |
| "learning_rate": 1.5042735042735044e-06, | |
| "loss": 0.164, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 42.01476923076923, | |
| "grad_norm": 0.09926377236843109, | |
| "learning_rate": 1.4700854700854702e-06, | |
| "loss": 0.0204, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 42.01784615384615, | |
| "grad_norm": 0.00442868797108531, | |
| "learning_rate": 1.4358974358974359e-06, | |
| "loss": 0.0327, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 42.02030769230769, | |
| "eval_accuracy": 0.6444444444444445, | |
| "eval_loss": 2.2866263389587402, | |
| "eval_runtime": 14.4549, | |
| "eval_samples_per_second": 3.113, | |
| "eval_steps_per_second": 0.83, | |
| "step": 2838 | |
| }, | |
| { | |
| "epoch": 43.00061538461539, | |
| "grad_norm": 0.010174613445997238, | |
| "learning_rate": 1.401709401709402e-06, | |
| "loss": 0.0003, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 43.003692307692305, | |
| "grad_norm": 0.0034826870542019606, | |
| "learning_rate": 1.3675213675213678e-06, | |
| "loss": 0.0011, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 43.00676923076923, | |
| "grad_norm": 0.006616006139665842, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 0.1652, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 43.009846153846155, | |
| "grad_norm": 0.006677310448139906, | |
| "learning_rate": 1.2991452991452993e-06, | |
| "loss": 0.0002, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 43.01292307692308, | |
| "grad_norm": 0.024410521611571312, | |
| "learning_rate": 1.264957264957265e-06, | |
| "loss": 0.0002, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 43.016, | |
| "grad_norm": 0.016464335843920708, | |
| "learning_rate": 1.230769230769231e-06, | |
| "loss": 0.0058, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 43.01907692307692, | |
| "grad_norm": 0.005353263579308987, | |
| "learning_rate": 1.1965811965811968e-06, | |
| "loss": 0.0349, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 43.02030769230769, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 2.2496182918548584, | |
| "eval_runtime": 13.6552, | |
| "eval_samples_per_second": 3.295, | |
| "eval_steps_per_second": 0.879, | |
| "step": 2904 | |
| }, | |
| { | |
| "epoch": 44.00184615384615, | |
| "grad_norm": 0.002462292555719614, | |
| "learning_rate": 1.1623931623931625e-06, | |
| "loss": 0.0001, | |
| "step": 2910 | |
| }, | |
| { | |
| "epoch": 44.00492307692308, | |
| "grad_norm": 0.04071797430515289, | |
| "learning_rate": 1.1282051282051283e-06, | |
| "loss": 0.1643, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 44.008, | |
| "grad_norm": 0.12705393135547638, | |
| "learning_rate": 1.0940170940170942e-06, | |
| "loss": 0.0002, | |
| "step": 2930 | |
| }, | |
| { | |
| "epoch": 44.01107692307692, | |
| "grad_norm": 0.027374783530831337, | |
| "learning_rate": 1.0598290598290598e-06, | |
| "loss": 0.0002, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 44.014153846153846, | |
| "grad_norm": 0.0212483499199152, | |
| "learning_rate": 1.0256410256410257e-06, | |
| "loss": 0.1404, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 44.01723076923077, | |
| "grad_norm": 0.01933380588889122, | |
| "learning_rate": 9.914529914529915e-07, | |
| "loss": 0.0002, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 44.02030769230769, | |
| "grad_norm": 0.005917856004089117, | |
| "learning_rate": 9.572649572649574e-07, | |
| "loss": 0.1178, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 44.02030769230769, | |
| "eval_accuracy": 0.6888888888888889, | |
| "eval_loss": 1.8929468393325806, | |
| "eval_runtime": 14.1233, | |
| "eval_samples_per_second": 3.186, | |
| "eval_steps_per_second": 0.85, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 45.003076923076925, | |
| "grad_norm": 0.0025133194867521524, | |
| "learning_rate": 9.230769230769232e-07, | |
| "loss": 0.0023, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 45.00615384615384, | |
| "grad_norm": 0.3587754964828491, | |
| "learning_rate": 8.88888888888889e-07, | |
| "loss": 0.0003, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 45.00923076923077, | |
| "grad_norm": 0.00992894358932972, | |
| "learning_rate": 8.547008547008548e-07, | |
| "loss": 0.0003, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 45.012307692307694, | |
| "grad_norm": 0.007983594201505184, | |
| "learning_rate": 8.205128205128206e-07, | |
| "loss": 0.0006, | |
| "step": 3010 | |
| }, | |
| { | |
| "epoch": 45.01538461538462, | |
| "grad_norm": 0.07788557559251785, | |
| "learning_rate": 7.863247863247863e-07, | |
| "loss": 0.0003, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 45.01846153846154, | |
| "grad_norm": 0.006400303915143013, | |
| "learning_rate": 7.521367521367522e-07, | |
| "loss": 0.0001, | |
| "step": 3030 | |
| }, | |
| { | |
| "epoch": 45.02030769230769, | |
| "eval_accuracy": 0.7111111111111111, | |
| "eval_loss": 1.903012990951538, | |
| "eval_runtime": 13.7223, | |
| "eval_samples_per_second": 3.279, | |
| "eval_steps_per_second": 0.874, | |
| "step": 3036 | |
| }, | |
| { | |
| "epoch": 46.001230769230766, | |
| "grad_norm": 0.01581370085477829, | |
| "learning_rate": 7.179487179487179e-07, | |
| "loss": 0.0004, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 46.00430769230769, | |
| "grad_norm": 0.006594088859856129, | |
| "learning_rate": 6.837606837606839e-07, | |
| "loss": 0.0007, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 46.007384615384616, | |
| "grad_norm": 0.00645817257463932, | |
| "learning_rate": 6.495726495726496e-07, | |
| "loss": 0.1504, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 46.01046153846154, | |
| "grad_norm": 0.002733448054641485, | |
| "learning_rate": 6.153846153846155e-07, | |
| "loss": 0.0001, | |
| "step": 3070 | |
| }, | |
| { | |
| "epoch": 46.01353846153846, | |
| "grad_norm": 0.0036937277764081955, | |
| "learning_rate": 5.811965811965812e-07, | |
| "loss": 0.0001, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 46.016615384615385, | |
| "grad_norm": 0.14585720002651215, | |
| "learning_rate": 5.470085470085471e-07, | |
| "loss": 0.1164, | |
| "step": 3090 | |
| }, | |
| { | |
| "epoch": 46.01969230769231, | |
| "grad_norm": 0.0056580170057713985, | |
| "learning_rate": 5.128205128205128e-07, | |
| "loss": 0.1289, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 46.02030769230769, | |
| "eval_accuracy": 0.6888888888888889, | |
| "eval_loss": 1.9212490320205688, | |
| "eval_runtime": 13.6271, | |
| "eval_samples_per_second": 3.302, | |
| "eval_steps_per_second": 0.881, | |
| "step": 3102 | |
| }, | |
| { | |
| "epoch": 47.00246153846154, | |
| "grad_norm": 0.0057002464309334755, | |
| "learning_rate": 4.786324786324787e-07, | |
| "loss": 0.0002, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 47.005538461538464, | |
| "grad_norm": 37.331138610839844, | |
| "learning_rate": 4.444444444444445e-07, | |
| "loss": 0.0038, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 47.00861538461538, | |
| "grad_norm": 0.003734712256118655, | |
| "learning_rate": 4.102564102564103e-07, | |
| "loss": 0.0119, | |
| "step": 3130 | |
| }, | |
| { | |
| "epoch": 47.01169230769231, | |
| "grad_norm": 0.005541326943784952, | |
| "learning_rate": 3.760683760683761e-07, | |
| "loss": 0.0003, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 47.01476923076923, | |
| "grad_norm": 0.012346943840384483, | |
| "learning_rate": 3.4188034188034194e-07, | |
| "loss": 0.0001, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 47.01784615384615, | |
| "grad_norm": 0.05111232027411461, | |
| "learning_rate": 3.0769230769230774e-07, | |
| "loss": 0.137, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 47.02030769230769, | |
| "eval_accuracy": 0.7555555555555555, | |
| "eval_loss": 1.6076961755752563, | |
| "eval_runtime": 13.6997, | |
| "eval_samples_per_second": 3.285, | |
| "eval_steps_per_second": 0.876, | |
| "step": 3168 | |
| }, | |
| { | |
| "epoch": 48.00061538461539, | |
| "grad_norm": 0.005067374091595411, | |
| "learning_rate": 2.7350427350427354e-07, | |
| "loss": 0.0002, | |
| "step": 3170 | |
| }, | |
| { | |
| "epoch": 48.003692307692305, | |
| "grad_norm": 0.5271658301353455, | |
| "learning_rate": 2.3931623931623934e-07, | |
| "loss": 0.0002, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 48.00676923076923, | |
| "grad_norm": 0.01560221053659916, | |
| "learning_rate": 2.0512820512820514e-07, | |
| "loss": 0.0459, | |
| "step": 3190 | |
| }, | |
| { | |
| "epoch": 48.009846153846155, | |
| "grad_norm": 0.007082493510097265, | |
| "learning_rate": 1.7094017094017097e-07, | |
| "loss": 0.0005, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 48.01292307692308, | |
| "grad_norm": 0.0056843808852136135, | |
| "learning_rate": 1.3675213675213677e-07, | |
| "loss": 0.0009, | |
| "step": 3210 | |
| }, | |
| { | |
| "epoch": 48.016, | |
| "grad_norm": 0.020851995795965195, | |
| "learning_rate": 1.0256410256410257e-07, | |
| "loss": 0.0197, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 48.01907692307692, | |
| "grad_norm": 0.0018140689935535192, | |
| "learning_rate": 6.837606837606839e-08, | |
| "loss": 0.0004, | |
| "step": 3230 | |
| }, | |
| { | |
| "epoch": 48.02030769230769, | |
| "eval_accuracy": 0.7333333333333333, | |
| "eval_loss": 1.7398090362548828, | |
| "eval_runtime": 13.7259, | |
| "eval_samples_per_second": 3.278, | |
| "eval_steps_per_second": 0.874, | |
| "step": 3234 | |
| }, | |
| { | |
| "epoch": 49.00184615384615, | |
| "grad_norm": 0.007604257669299841, | |
| "learning_rate": 3.418803418803419e-08, | |
| "loss": 0.0001, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 49.00492307692308, | |
| "grad_norm": 0.003877155715599656, | |
| "learning_rate": 0.0, | |
| "loss": 0.0001, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 49.00492307692308, | |
| "eval_accuracy": 0.7333333333333333, | |
| "eval_loss": 1.7422600984573364, | |
| "eval_runtime": 16.804, | |
| "eval_samples_per_second": 2.678, | |
| "eval_steps_per_second": 0.714, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 49.00492307692308, | |
| "step": 3250, | |
| "total_flos": 5.643786123050936e+19, | |
| "train_loss": 0.2673717604703628, | |
| "train_runtime": 7905.9537, | |
| "train_samples_per_second": 1.644, | |
| "train_steps_per_second": 0.411 | |
| }, | |
| { | |
| "epoch": 49.00492307692308, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.9770982265472412, | |
| "eval_runtime": 13.7887, | |
| "eval_samples_per_second": 3.264, | |
| "eval_steps_per_second": 0.87, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 49.00492307692308, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.9770981073379517, | |
| "eval_runtime": 13.7323, | |
| "eval_samples_per_second": 3.277, | |
| "eval_steps_per_second": 0.874, | |
| "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": 5.643786123050936e+19, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |