| { | |
| "best_metric": 0.3851391077041626, | |
| "best_model_checkpoint": "./beans_outputs/checkpoint-130", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 1300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07692307692307693, | |
| "grad_norm": 1.2493035793304443, | |
| "learning_rate": 0.004961538461538462, | |
| "loss": 0.8568, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 1.4747363328933716, | |
| "learning_rate": 0.004923076923076923, | |
| "loss": 0.6606, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.23076923076923078, | |
| "grad_norm": 0.8188544511795044, | |
| "learning_rate": 0.004884615384615385, | |
| "loss": 0.3787, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 1.718248724937439, | |
| "learning_rate": 0.004846153846153846, | |
| "loss": 0.3154, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 2.2010860443115234, | |
| "learning_rate": 0.004807692307692308, | |
| "loss": 0.4932, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "grad_norm": 2.927755832672119, | |
| "learning_rate": 0.0047692307692307695, | |
| "loss": 0.561, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5384615384615384, | |
| "grad_norm": 1.2203412055969238, | |
| "learning_rate": 0.004730769230769231, | |
| "loss": 0.3309, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 0.08795254677534103, | |
| "learning_rate": 0.004692307692307693, | |
| "loss": 0.3261, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6923076923076923, | |
| "grad_norm": 0.4609662890434265, | |
| "learning_rate": 0.004653846153846154, | |
| "loss": 0.195, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 0.9046596884727478, | |
| "learning_rate": 0.004615384615384616, | |
| "loss": 0.3352, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8461538461538461, | |
| "grad_norm": 2.5637238025665283, | |
| "learning_rate": 0.0045769230769230765, | |
| "loss": 0.5961, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 0.32293158769607544, | |
| "learning_rate": 0.004538461538461539, | |
| "loss": 0.236, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 7.484982967376709, | |
| "learning_rate": 0.0045000000000000005, | |
| "loss": 0.6323, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_loss": 0.3851391077041626, | |
| "eval_runtime": 0.8932, | |
| "eval_samples_per_second": 148.91, | |
| "eval_steps_per_second": 19.034, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0769230769230769, | |
| "grad_norm": 0.12631376087665558, | |
| "learning_rate": 0.004461538461538462, | |
| "loss": 0.747, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 1.5336790084838867, | |
| "learning_rate": 0.004423076923076923, | |
| "loss": 0.3062, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 0.7460920214653015, | |
| "learning_rate": 0.004384615384615384, | |
| "loss": 0.2088, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3076923076923077, | |
| "grad_norm": 0.6954104900360107, | |
| "learning_rate": 0.004346153846153846, | |
| "loss": 0.2942, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.3846153846153846, | |
| "grad_norm": 2.2199032306671143, | |
| "learning_rate": 0.004307692307692308, | |
| "loss": 0.315, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.4615384615384617, | |
| "grad_norm": 2.654019832611084, | |
| "learning_rate": 0.004269230769230769, | |
| "loss": 0.4203, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 1.460809350013733, | |
| "learning_rate": 0.004230769230769231, | |
| "loss": 0.4812, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6153846153846154, | |
| "grad_norm": 0.9371921420097351, | |
| "learning_rate": 0.004192307692307692, | |
| "loss": 0.3436, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.6923076923076923, | |
| "grad_norm": 0.573295533657074, | |
| "learning_rate": 0.004153846153846154, | |
| "loss": 0.7394, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.7692307692307692, | |
| "grad_norm": 3.2005550861358643, | |
| "learning_rate": 0.004115384615384615, | |
| "loss": 0.5694, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 1.0854452848434448, | |
| "learning_rate": 0.004076923076923077, | |
| "loss": 0.6043, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 1.3743146657943726, | |
| "learning_rate": 0.0040384615384615385, | |
| "loss": 0.4406, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.11435157805681229, | |
| "learning_rate": 0.004, | |
| "loss": 0.4102, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6240601503759399, | |
| "eval_loss": 1.1071540117263794, | |
| "eval_runtime": 0.9165, | |
| "eval_samples_per_second": 145.121, | |
| "eval_steps_per_second": 18.549, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.076923076923077, | |
| "grad_norm": 0.9636590480804443, | |
| "learning_rate": 0.003961538461538462, | |
| "loss": 0.4669, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "grad_norm": 1.4855515956878662, | |
| "learning_rate": 0.003923076923076923, | |
| "loss": 0.4618, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.230769230769231, | |
| "grad_norm": 1.6462602615356445, | |
| "learning_rate": 0.003884615384615385, | |
| "loss": 0.5266, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 2.7035958766937256, | |
| "learning_rate": 0.0038461538461538464, | |
| "loss": 0.6045, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.3846153846153846, | |
| "grad_norm": 1.7253210544586182, | |
| "learning_rate": 0.0038076923076923075, | |
| "loss": 0.4711, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 0.6617446541786194, | |
| "learning_rate": 0.003769230769230769, | |
| "loss": 0.5785, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.5384615384615383, | |
| "grad_norm": 0.6448426842689514, | |
| "learning_rate": 0.003730769230769231, | |
| "loss": 0.7907, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.6153846153846154, | |
| "grad_norm": 3.4137613773345947, | |
| "learning_rate": 0.0036923076923076927, | |
| "loss": 0.9229, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 5.124917507171631, | |
| "learning_rate": 0.003653846153846154, | |
| "loss": 1.0287, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": 1.2190972566604614, | |
| "learning_rate": 0.0036153846153846154, | |
| "loss": 0.7791, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.8461538461538463, | |
| "grad_norm": 0.6898893713951111, | |
| "learning_rate": 0.003576923076923077, | |
| "loss": 1.4277, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.9230769230769234, | |
| "grad_norm": 2.563831090927124, | |
| "learning_rate": 0.003538461538461539, | |
| "loss": 1.3955, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.519380569458008, | |
| "learning_rate": 0.0034999999999999996, | |
| "loss": 0.8964, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6616541353383458, | |
| "eval_loss": 0.8284351825714111, | |
| "eval_runtime": 0.9347, | |
| "eval_samples_per_second": 142.291, | |
| "eval_steps_per_second": 18.188, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 4.3924150466918945, | |
| "learning_rate": 0.0034615384615384616, | |
| "loss": 1.2627, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.1538461538461537, | |
| "grad_norm": 1.773996114730835, | |
| "learning_rate": 0.003423076923076923, | |
| "loss": 1.3465, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.230769230769231, | |
| "grad_norm": 3.1790945529937744, | |
| "learning_rate": 0.003384615384615385, | |
| "loss": 1.2822, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.3076923076923075, | |
| "grad_norm": 5.48622465133667, | |
| "learning_rate": 0.003346153846153846, | |
| "loss": 1.0807, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.3846153846153846, | |
| "grad_norm": 3.6509623527526855, | |
| "learning_rate": 0.0033076923076923075, | |
| "loss": 1.1806, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.4615384615384617, | |
| "grad_norm": 2.3183326721191406, | |
| "learning_rate": 0.0032692307692307695, | |
| "loss": 1.2542, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.5384615384615383, | |
| "grad_norm": 3.2686471939086914, | |
| "learning_rate": 0.003230769230769231, | |
| "loss": 1.2217, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.6153846153846154, | |
| "grad_norm": 0.33250316977500916, | |
| "learning_rate": 0.003192307692307692, | |
| "loss": 1.2107, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "grad_norm": 2.4644064903259277, | |
| "learning_rate": 0.0031538461538461538, | |
| "loss": 1.2287, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.769230769230769, | |
| "grad_norm": 0.9661989808082581, | |
| "learning_rate": 0.0031153846153846153, | |
| "loss": 1.1392, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 1.0765918493270874, | |
| "learning_rate": 0.0030769230769230774, | |
| "loss": 1.1576, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.9230769230769234, | |
| "grad_norm": 0.5606147646903992, | |
| "learning_rate": 0.0030384615384615385, | |
| "loss": 1.1148, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.787473678588867, | |
| "learning_rate": 0.003, | |
| "loss": 1.155, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3383458646616541, | |
| "eval_loss": 1.1070177555084229, | |
| "eval_runtime": 0.9062, | |
| "eval_samples_per_second": 146.767, | |
| "eval_steps_per_second": 18.76, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.076923076923077, | |
| "grad_norm": 1.8376567363739014, | |
| "learning_rate": 0.0029615384615384616, | |
| "loss": 1.16, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.153846153846154, | |
| "grad_norm": 0.1573391705751419, | |
| "learning_rate": 0.002923076923076923, | |
| "loss": 1.1133, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.230769230769231, | |
| "grad_norm": 1.7338229417800903, | |
| "learning_rate": 0.0028846153846153843, | |
| "loss": 1.1604, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.3076923076923075, | |
| "grad_norm": 1.532076358795166, | |
| "learning_rate": 0.002846153846153846, | |
| "loss": 1.171, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.384615384615385, | |
| "grad_norm": 1.2591376304626465, | |
| "learning_rate": 0.002807692307692308, | |
| "loss": 1.116, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.461538461538462, | |
| "grad_norm": 1.7697715759277344, | |
| "learning_rate": 0.0027692307692307695, | |
| "loss": 1.1409, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.538461538461538, | |
| "grad_norm": 1.0740739107131958, | |
| "learning_rate": 0.0027307692307692306, | |
| "loss": 1.1458, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 0.46824851632118225, | |
| "learning_rate": 0.002692307692307692, | |
| "loss": 1.1056, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.6923076923076925, | |
| "grad_norm": 1.8973933458328247, | |
| "learning_rate": 0.0026538461538461538, | |
| "loss": 1.1024, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.769230769230769, | |
| "grad_norm": 0.06765652447938919, | |
| "learning_rate": 0.0026153846153846158, | |
| "loss": 1.1514, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.846153846153846, | |
| "grad_norm": 0.8284403085708618, | |
| "learning_rate": 0.002576923076923077, | |
| "loss": 1.1799, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "grad_norm": 1.611846923828125, | |
| "learning_rate": 0.0025384615384615385, | |
| "loss": 1.169, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.048553705215454, | |
| "learning_rate": 0.0025, | |
| "loss": 1.1069, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.3308270676691729, | |
| "eval_loss": 1.1081749200820923, | |
| "eval_runtime": 0.9355, | |
| "eval_samples_per_second": 142.176, | |
| "eval_steps_per_second": 18.173, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.076923076923077, | |
| "grad_norm": 0.6719330549240112, | |
| "learning_rate": 0.0024615384615384616, | |
| "loss": 1.1054, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 5.153846153846154, | |
| "grad_norm": 0.8374503254890442, | |
| "learning_rate": 0.002423076923076923, | |
| "loss": 1.1134, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 5.230769230769231, | |
| "grad_norm": 0.5713789463043213, | |
| "learning_rate": 0.0023846153846153848, | |
| "loss": 1.1534, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 5.3076923076923075, | |
| "grad_norm": 1.0185028314590454, | |
| "learning_rate": 0.0023461538461538463, | |
| "loss": 1.1368, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 5.384615384615385, | |
| "grad_norm": 1.021331787109375, | |
| "learning_rate": 0.002307692307692308, | |
| "loss": 1.094, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.461538461538462, | |
| "grad_norm": 0.9150602221488953, | |
| "learning_rate": 0.0022692307692307695, | |
| "loss": 1.1187, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 5.538461538461538, | |
| "grad_norm": 0.741632878780365, | |
| "learning_rate": 0.002230769230769231, | |
| "loss": 1.1129, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 5.615384615384615, | |
| "grad_norm": 1.5859191417694092, | |
| "learning_rate": 0.002192307692307692, | |
| "loss": 1.135, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 5.6923076923076925, | |
| "grad_norm": 2.2667315006256104, | |
| "learning_rate": 0.002153846153846154, | |
| "loss": 1.1474, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 5.769230769230769, | |
| "grad_norm": 1.3794482946395874, | |
| "learning_rate": 0.0021153846153846153, | |
| "loss": 1.1095, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 5.846153846153846, | |
| "grad_norm": 1.2423471212387085, | |
| "learning_rate": 0.002076923076923077, | |
| "loss": 1.1057, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 5.923076923076923, | |
| "grad_norm": 4.28483247756958, | |
| "learning_rate": 0.0020384615384615385, | |
| "loss": 1.1093, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 4.998449802398682, | |
| "learning_rate": 0.002, | |
| "loss": 1.1748, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.3308270676691729, | |
| "eval_loss": 1.1041723489761353, | |
| "eval_runtime": 0.9442, | |
| "eval_samples_per_second": 140.864, | |
| "eval_steps_per_second": 18.005, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.076923076923077, | |
| "grad_norm": 2.81730318069458, | |
| "learning_rate": 0.0019615384615384616, | |
| "loss": 1.1381, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 2.5270557403564453, | |
| "learning_rate": 0.0019230769230769232, | |
| "loss": 1.1104, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.230769230769231, | |
| "grad_norm": 0.8537469506263733, | |
| "learning_rate": 0.0018846153846153845, | |
| "loss": 1.1327, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 6.3076923076923075, | |
| "grad_norm": 0.8818097710609436, | |
| "learning_rate": 0.0018461538461538463, | |
| "loss": 1.1437, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 6.384615384615385, | |
| "grad_norm": 2.496903896331787, | |
| "learning_rate": 0.0018076923076923077, | |
| "loss": 1.1251, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 6.461538461538462, | |
| "grad_norm": 3.2999167442321777, | |
| "learning_rate": 0.0017692307692307695, | |
| "loss": 1.1373, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 6.538461538461538, | |
| "grad_norm": 2.9537723064422607, | |
| "learning_rate": 0.0017307692307692308, | |
| "loss": 1.176, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 6.615384615384615, | |
| "grad_norm": 1.5148764848709106, | |
| "learning_rate": 0.0016923076923076924, | |
| "loss": 1.1146, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 6.6923076923076925, | |
| "grad_norm": 1.3614485263824463, | |
| "learning_rate": 0.0016538461538461537, | |
| "loss": 1.0895, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 6.769230769230769, | |
| "grad_norm": 0.6686407327651978, | |
| "learning_rate": 0.0016153846153846155, | |
| "loss": 1.1245, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 6.846153846153846, | |
| "grad_norm": 1.6276365518569946, | |
| "learning_rate": 0.0015769230769230769, | |
| "loss": 1.1412, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 6.923076923076923, | |
| "grad_norm": 1.4945542812347412, | |
| "learning_rate": 0.0015384615384615387, | |
| "loss": 1.1247, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.74285888671875, | |
| "learning_rate": 0.0015, | |
| "loss": 1.1514, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.3308270676691729, | |
| "eval_loss": 1.1007708311080933, | |
| "eval_runtime": 0.9264, | |
| "eval_samples_per_second": 143.562, | |
| "eval_steps_per_second": 18.35, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 7.076923076923077, | |
| "grad_norm": 2.248867988586426, | |
| "learning_rate": 0.0014615384615384616, | |
| "loss": 1.0913, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 7.153846153846154, | |
| "grad_norm": 2.5912275314331055, | |
| "learning_rate": 0.001423076923076923, | |
| "loss": 1.1193, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 7.230769230769231, | |
| "grad_norm": 3.573967695236206, | |
| "learning_rate": 0.0013846153846153847, | |
| "loss": 1.1352, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 7.3076923076923075, | |
| "grad_norm": 2.077554225921631, | |
| "learning_rate": 0.001346153846153846, | |
| "loss": 1.0947, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 7.384615384615385, | |
| "grad_norm": 1.6364086866378784, | |
| "learning_rate": 0.0013076923076923079, | |
| "loss": 1.1296, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 7.461538461538462, | |
| "grad_norm": 1.1806069612503052, | |
| "learning_rate": 0.0012692307692307692, | |
| "loss": 1.1033, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 7.538461538461538, | |
| "grad_norm": 2.4788713455200195, | |
| "learning_rate": 0.0012307692307692308, | |
| "loss": 1.1148, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 7.615384615384615, | |
| "grad_norm": 0.8245519399642944, | |
| "learning_rate": 0.0011923076923076924, | |
| "loss": 1.1121, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 1.568903923034668, | |
| "learning_rate": 0.001153846153846154, | |
| "loss": 1.1163, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.769230769230769, | |
| "grad_norm": 2.2075676918029785, | |
| "learning_rate": 0.0011153846153846155, | |
| "loss": 1.0956, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 7.846153846153846, | |
| "grad_norm": 1.0866578817367554, | |
| "learning_rate": 0.001076923076923077, | |
| "loss": 1.1065, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 7.923076923076923, | |
| "grad_norm": 2.4340384006500244, | |
| "learning_rate": 0.0010384615384615384, | |
| "loss": 1.1118, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.1427950859069824, | |
| "learning_rate": 0.001, | |
| "loss": 1.0956, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.3383458646616541, | |
| "eval_loss": 1.1004995107650757, | |
| "eval_runtime": 0.9133, | |
| "eval_samples_per_second": 145.621, | |
| "eval_steps_per_second": 18.613, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 8.076923076923077, | |
| "grad_norm": 0.689738392829895, | |
| "learning_rate": 0.0009615384615384616, | |
| "loss": 1.1119, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 8.153846153846153, | |
| "grad_norm": 0.7051054835319519, | |
| "learning_rate": 0.0009230769230769232, | |
| "loss": 1.1076, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 8.23076923076923, | |
| "grad_norm": 0.8233575820922852, | |
| "learning_rate": 0.0008846153846153847, | |
| "loss": 1.1, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 8.307692307692308, | |
| "grad_norm": 0.4401038885116577, | |
| "learning_rate": 0.0008461538461538462, | |
| "loss": 1.1231, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 8.384615384615385, | |
| "grad_norm": 2.0618515014648438, | |
| "learning_rate": 0.0008076923076923078, | |
| "loss": 1.0877, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 8.461538461538462, | |
| "grad_norm": 1.629764199256897, | |
| "learning_rate": 0.0007692307692307693, | |
| "loss": 1.1165, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 8.538461538461538, | |
| "grad_norm": 0.9111972451210022, | |
| "learning_rate": 0.0007307692307692308, | |
| "loss": 1.1033, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 8.615384615384615, | |
| "grad_norm": 2.03430438041687, | |
| "learning_rate": 0.0006923076923076924, | |
| "loss": 1.0906, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 8.692307692307692, | |
| "grad_norm": 3.2121763229370117, | |
| "learning_rate": 0.0006538461538461539, | |
| "loss": 1.126, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 8.76923076923077, | |
| "grad_norm": 0.34755992889404297, | |
| "learning_rate": 0.0006153846153846154, | |
| "loss": 1.1051, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 8.846153846153847, | |
| "grad_norm": 0.7230011224746704, | |
| "learning_rate": 0.000576923076923077, | |
| "loss": 1.0721, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 8.923076923076923, | |
| "grad_norm": 2.8672590255737305, | |
| "learning_rate": 0.0005384615384615385, | |
| "loss": 1.0711, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 2.920103073120117, | |
| "learning_rate": 0.0005, | |
| "loss": 1.1448, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.3308270676691729, | |
| "eval_loss": 1.111855149269104, | |
| "eval_runtime": 0.9175, | |
| "eval_samples_per_second": 144.96, | |
| "eval_steps_per_second": 18.529, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 9.076923076923077, | |
| "grad_norm": 0.26905354857444763, | |
| "learning_rate": 0.0004615384615384616, | |
| "loss": 1.1121, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 9.153846153846153, | |
| "grad_norm": 0.22876305878162384, | |
| "learning_rate": 0.0004230769230769231, | |
| "loss": 1.0967, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 0.24429693818092346, | |
| "learning_rate": 0.00038461538461538467, | |
| "loss": 1.1023, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 9.307692307692308, | |
| "grad_norm": 0.6317471861839294, | |
| "learning_rate": 0.0003461538461538462, | |
| "loss": 1.1062, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 9.384615384615385, | |
| "grad_norm": 0.41996484994888306, | |
| "learning_rate": 0.0003076923076923077, | |
| "loss": 1.1162, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 9.461538461538462, | |
| "grad_norm": 1.6321758031845093, | |
| "learning_rate": 0.0002692307692307693, | |
| "loss": 1.0976, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 9.538461538461538, | |
| "grad_norm": 0.966174840927124, | |
| "learning_rate": 0.0002307692307692308, | |
| "loss": 1.0979, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 9.615384615384615, | |
| "grad_norm": 0.23766568303108215, | |
| "learning_rate": 0.00019230769230769233, | |
| "loss": 1.1012, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 9.692307692307692, | |
| "grad_norm": 0.930920422077179, | |
| "learning_rate": 0.00015384615384615385, | |
| "loss": 1.0991, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 9.76923076923077, | |
| "grad_norm": 0.34004735946655273, | |
| "learning_rate": 0.0001153846153846154, | |
| "loss": 1.1009, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 9.846153846153847, | |
| "grad_norm": 0.663216233253479, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 1.0997, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 9.923076923076923, | |
| "grad_norm": 0.9082524180412292, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 1.0998, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.356846570968628, | |
| "learning_rate": 0.0, | |
| "loss": 1.0991, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.3308270676691729, | |
| "eval_loss": 1.0983037948608398, | |
| "eval_runtime": 0.9323, | |
| "eval_samples_per_second": 142.651, | |
| "eval_steps_per_second": 18.234, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 1300, | |
| "total_flos": 8.400578669044531e+17, | |
| "train_loss": 0.9618942783429072, | |
| "train_runtime": 164.295, | |
| "train_samples_per_second": 62.936, | |
| "train_steps_per_second": 7.913 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 8.400578669044531e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |