| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.24509803921568626, | |
| "eval_steps": 10, | |
| "global_step": 550, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004456327985739751, | |
| "grad_norm": 0.7916810512542725, | |
| "learning_rate": 0.0002945454545454545, | |
| "loss": 0.3376, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.004456327985739751, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.6079411506652832, | |
| "eval_runtime": 537.6759, | |
| "eval_samples_per_second": 8.347, | |
| "eval_steps_per_second": 2.087, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008912655971479501, | |
| "grad_norm": 0.1121700331568718, | |
| "learning_rate": 0.00028909090909090904, | |
| "loss": 0.3858, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.008912655971479501, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.4899146854877472, | |
| "eval_runtime": 528.4079, | |
| "eval_samples_per_second": 8.493, | |
| "eval_steps_per_second": 2.123, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.013368983957219251, | |
| "grad_norm": 0.4192979633808136, | |
| "learning_rate": 0.0002836363636363636, | |
| "loss": 0.4091, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.013368983957219251, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.4288414418697357, | |
| "eval_runtime": 532.0654, | |
| "eval_samples_per_second": 8.435, | |
| "eval_steps_per_second": 2.109, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.017825311942959002, | |
| "grad_norm": 0.6693852543830872, | |
| "learning_rate": 0.00027818181818181815, | |
| "loss": 0.3274, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.017825311942959002, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.3423554301261902, | |
| "eval_runtime": 534.016, | |
| "eval_samples_per_second": 8.404, | |
| "eval_steps_per_second": 2.101, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.022281639928698752, | |
| "grad_norm": 2.928346633911133, | |
| "learning_rate": 0.0002727272727272727, | |
| "loss": 0.3434, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.022281639928698752, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.32134178280830383, | |
| "eval_runtime": 540.6384, | |
| "eval_samples_per_second": 8.301, | |
| "eval_steps_per_second": 2.075, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.026737967914438502, | |
| "grad_norm": 7.551202774047852, | |
| "learning_rate": 0.0002672727272727272, | |
| "loss": 0.3627, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.026737967914438502, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.24254043400287628, | |
| "eval_runtime": 532.8983, | |
| "eval_samples_per_second": 8.422, | |
| "eval_steps_per_second": 2.105, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.031194295900178252, | |
| "grad_norm": 5.043577671051025, | |
| "learning_rate": 0.0002618181818181818, | |
| "loss": 0.3062, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.031194295900178252, | |
| "eval_accuracy": 0.8770053386688232, | |
| "eval_loss": 0.16930516064167023, | |
| "eval_runtime": 531.9662, | |
| "eval_samples_per_second": 8.437, | |
| "eval_steps_per_second": 2.109, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.035650623885918005, | |
| "grad_norm": 0.5775203108787537, | |
| "learning_rate": 0.00025636363636363633, | |
| "loss": 0.2115, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.035650623885918005, | |
| "eval_accuracy": 0.8687611222267151, | |
| "eval_loss": 0.2173369973897934, | |
| "eval_runtime": 536.5487, | |
| "eval_samples_per_second": 8.365, | |
| "eval_steps_per_second": 2.091, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.040106951871657755, | |
| "grad_norm": 1.0835380554199219, | |
| "learning_rate": 0.00025090909090909086, | |
| "loss": 0.2757, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.040106951871657755, | |
| "eval_accuracy": 0.8426916003227234, | |
| "eval_loss": 0.2951839566230774, | |
| "eval_runtime": 531.3931, | |
| "eval_samples_per_second": 8.446, | |
| "eval_steps_per_second": 2.111, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.044563279857397504, | |
| "grad_norm": 0.8163406848907471, | |
| "learning_rate": 0.00024545454545454545, | |
| "loss": 0.2344, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.044563279857397504, | |
| "eval_accuracy": 0.8957219123840332, | |
| "eval_loss": 0.17275306582450867, | |
| "eval_runtime": 539.8237, | |
| "eval_samples_per_second": 8.314, | |
| "eval_steps_per_second": 2.078, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.049019607843137254, | |
| "grad_norm": 0.909106433391571, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 0.3855, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.049019607843137254, | |
| "eval_accuracy": 0.9086452722549438, | |
| "eval_loss": 0.22119583189487457, | |
| "eval_runtime": 532.7138, | |
| "eval_samples_per_second": 8.425, | |
| "eval_steps_per_second": 2.106, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.053475935828877004, | |
| "grad_norm": 4.141552448272705, | |
| "learning_rate": 0.00023454545454545454, | |
| "loss": 0.3104, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.053475935828877004, | |
| "eval_accuracy": 0.8426916003227234, | |
| "eval_loss": 0.2036559134721756, | |
| "eval_runtime": 539.9806, | |
| "eval_samples_per_second": 8.311, | |
| "eval_steps_per_second": 2.078, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.057932263814616754, | |
| "grad_norm": 0.04542813077569008, | |
| "learning_rate": 0.00022909090909090907, | |
| "loss": 0.235, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.057932263814616754, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.6018572449684143, | |
| "eval_runtime": 532.3019, | |
| "eval_samples_per_second": 8.431, | |
| "eval_steps_per_second": 2.108, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.062388591800356503, | |
| "grad_norm": 0.20932789146900177, | |
| "learning_rate": 0.00022363636363636363, | |
| "loss": 0.383, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.062388591800356503, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.1932043582201004, | |
| "eval_runtime": 534.2227, | |
| "eval_samples_per_second": 8.401, | |
| "eval_steps_per_second": 2.1, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.06684491978609626, | |
| "grad_norm": 2.113372325897217, | |
| "learning_rate": 0.00021818181818181816, | |
| "loss": 0.3633, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.06684491978609626, | |
| "eval_accuracy": 0.9097593426704407, | |
| "eval_loss": 0.27564167976379395, | |
| "eval_runtime": 533.5467, | |
| "eval_samples_per_second": 8.412, | |
| "eval_steps_per_second": 2.103, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.07130124777183601, | |
| "grad_norm": 2.0779898166656494, | |
| "learning_rate": 0.00021272727272727272, | |
| "loss": 0.2534, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.07130124777183601, | |
| "eval_accuracy": 0.9124331474304199, | |
| "eval_loss": 0.17877253890037537, | |
| "eval_runtime": 534.2313, | |
| "eval_samples_per_second": 8.401, | |
| "eval_steps_per_second": 2.1, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.07575757575757576, | |
| "grad_norm": 2.8469862937927246, | |
| "learning_rate": 0.00020727272727272725, | |
| "loss": 0.1618, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.07575757575757576, | |
| "eval_accuracy": 0.9001782536506653, | |
| "eval_loss": 0.21939000487327576, | |
| "eval_runtime": 539.3123, | |
| "eval_samples_per_second": 8.322, | |
| "eval_steps_per_second": 2.08, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.08021390374331551, | |
| "grad_norm": 0.3544740378856659, | |
| "learning_rate": 0.0002018181818181818, | |
| "loss": 0.3204, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.08021390374331551, | |
| "eval_accuracy": 0.7437611222267151, | |
| "eval_loss": 0.6187111735343933, | |
| "eval_runtime": 539.0642, | |
| "eval_samples_per_second": 8.326, | |
| "eval_steps_per_second": 2.081, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.08467023172905526, | |
| "grad_norm": 0.08713312447071075, | |
| "learning_rate": 0.00019636363636363634, | |
| "loss": 0.4719, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.08467023172905526, | |
| "eval_accuracy": 0.917557954788208, | |
| "eval_loss": 0.15933164954185486, | |
| "eval_runtime": 532.5075, | |
| "eval_samples_per_second": 8.428, | |
| "eval_steps_per_second": 2.107, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.08912655971479501, | |
| "grad_norm": 0.10667263716459274, | |
| "learning_rate": 0.0001909090909090909, | |
| "loss": 0.2124, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08912655971479501, | |
| "eval_accuracy": 0.874331533908844, | |
| "eval_loss": 0.2731720209121704, | |
| "eval_runtime": 532.8068, | |
| "eval_samples_per_second": 8.423, | |
| "eval_steps_per_second": 2.106, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.09358288770053476, | |
| "grad_norm": 0.09285986423492432, | |
| "learning_rate": 0.00018545454545454543, | |
| "loss": 0.1807, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.09358288770053476, | |
| "eval_accuracy": 0.9458556175231934, | |
| "eval_loss": 0.15097761154174805, | |
| "eval_runtime": 534.5302, | |
| "eval_samples_per_second": 8.396, | |
| "eval_steps_per_second": 2.099, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.09803921568627451, | |
| "grad_norm": 0.21800605952739716, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 0.241, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.09803921568627451, | |
| "eval_accuracy": 0.9117646813392639, | |
| "eval_loss": 0.20164452493190765, | |
| "eval_runtime": 534.0184, | |
| "eval_samples_per_second": 8.404, | |
| "eval_steps_per_second": 2.101, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.10249554367201426, | |
| "grad_norm": 2.175548553466797, | |
| "learning_rate": 0.00017454545454545452, | |
| "loss": 0.3668, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.10249554367201426, | |
| "eval_accuracy": 0.9079768061637878, | |
| "eval_loss": 0.22817277908325195, | |
| "eval_runtime": 533.4214, | |
| "eval_samples_per_second": 8.414, | |
| "eval_steps_per_second": 2.103, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.10695187165775401, | |
| "grad_norm": 2.417444944381714, | |
| "learning_rate": 0.00016909090909090907, | |
| "loss": 0.2424, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.10695187165775401, | |
| "eval_accuracy": 0.9364973306655884, | |
| "eval_loss": 0.1917591691017151, | |
| "eval_runtime": 539.0061, | |
| "eval_samples_per_second": 8.326, | |
| "eval_steps_per_second": 2.082, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.11140819964349376, | |
| "grad_norm": 1.2514077425003052, | |
| "learning_rate": 0.0001636363636363636, | |
| "loss": 0.1934, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.11140819964349376, | |
| "eval_accuracy": 0.934937596321106, | |
| "eval_loss": 0.14482024312019348, | |
| "eval_runtime": 535.7699, | |
| "eval_samples_per_second": 8.377, | |
| "eval_steps_per_second": 2.094, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.11586452762923351, | |
| "grad_norm": 7.08188533782959, | |
| "learning_rate": 0.00015818181818181816, | |
| "loss": 0.3554, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.11586452762923351, | |
| "eval_accuracy": 0.9601158499717712, | |
| "eval_loss": 0.09824846684932709, | |
| "eval_runtime": 532.0513, | |
| "eval_samples_per_second": 8.435, | |
| "eval_steps_per_second": 2.109, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.12032085561497326, | |
| "grad_norm": 1.4493603706359863, | |
| "learning_rate": 0.0001527272727272727, | |
| "loss": 0.1619, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.12032085561497326, | |
| "eval_accuracy": 0.9569964408874512, | |
| "eval_loss": 0.11313515901565552, | |
| "eval_runtime": 531.5383, | |
| "eval_samples_per_second": 8.443, | |
| "eval_steps_per_second": 2.111, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.12477718360071301, | |
| "grad_norm": 0.9487653374671936, | |
| "learning_rate": 0.00014727272727272725, | |
| "loss": 0.0925, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.12477718360071301, | |
| "eval_accuracy": 0.9748217463493347, | |
| "eval_loss": 0.06883005797863007, | |
| "eval_runtime": 533.9256, | |
| "eval_samples_per_second": 8.406, | |
| "eval_steps_per_second": 2.101, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.12923351158645277, | |
| "grad_norm": 0.0604180209338665, | |
| "learning_rate": 0.0001418181818181818, | |
| "loss": 0.1223, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.12923351158645277, | |
| "eval_accuracy": 0.9766042828559875, | |
| "eval_loss": 0.056954748928546906, | |
| "eval_runtime": 539.6876, | |
| "eval_samples_per_second": 8.316, | |
| "eval_steps_per_second": 2.079, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.13368983957219252, | |
| "grad_norm": 2.264599561691284, | |
| "learning_rate": 0.00013636363636363634, | |
| "loss": 0.1144, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.13368983957219252, | |
| "eval_accuracy": 0.9741532802581787, | |
| "eval_loss": 0.06963187456130981, | |
| "eval_runtime": 535.6675, | |
| "eval_samples_per_second": 8.378, | |
| "eval_steps_per_second": 2.095, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.13814616755793227, | |
| "grad_norm": 3.3151047229766846, | |
| "learning_rate": 0.0001309090909090909, | |
| "loss": 0.0702, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.13814616755793227, | |
| "eval_accuracy": 0.8979501128196716, | |
| "eval_loss": 0.3257082402706146, | |
| "eval_runtime": 539.7299, | |
| "eval_samples_per_second": 8.315, | |
| "eval_steps_per_second": 2.079, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.14260249554367202, | |
| "grad_norm": 0.20427709817886353, | |
| "learning_rate": 0.00012545454545454543, | |
| "loss": 0.1778, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.14260249554367202, | |
| "eval_accuracy": 0.9463012218475342, | |
| "eval_loss": 0.15733219683170319, | |
| "eval_runtime": 532.2728, | |
| "eval_samples_per_second": 8.432, | |
| "eval_steps_per_second": 2.108, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.14705882352941177, | |
| "grad_norm": 0.29525595903396606, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 0.1151, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.14705882352941177, | |
| "eval_accuracy": 0.9881907105445862, | |
| "eval_loss": 0.04204181954264641, | |
| "eval_runtime": 532.9583, | |
| "eval_samples_per_second": 8.421, | |
| "eval_steps_per_second": 2.105, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.15151515151515152, | |
| "grad_norm": 0.05631331354379654, | |
| "learning_rate": 0.00011454545454545453, | |
| "loss": 0.0784, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.15151515151515152, | |
| "eval_accuracy": 0.9870766401290894, | |
| "eval_loss": 0.041727881878614426, | |
| "eval_runtime": 533.1617, | |
| "eval_samples_per_second": 8.418, | |
| "eval_steps_per_second": 2.104, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.15597147950089127, | |
| "grad_norm": 1.579466700553894, | |
| "learning_rate": 0.00010909090909090908, | |
| "loss": 0.1782, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.15597147950089127, | |
| "eval_accuracy": 0.981951892375946, | |
| "eval_loss": 0.04734932258725166, | |
| "eval_runtime": 530.6175, | |
| "eval_samples_per_second": 8.458, | |
| "eval_steps_per_second": 2.115, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.16042780748663102, | |
| "grad_norm": 0.4887264370918274, | |
| "learning_rate": 0.00010363636363636362, | |
| "loss": 0.0328, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.16042780748663102, | |
| "eval_accuracy": 0.9826202988624573, | |
| "eval_loss": 0.0771423876285553, | |
| "eval_runtime": 539.3929, | |
| "eval_samples_per_second": 8.32, | |
| "eval_steps_per_second": 2.08, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.16488413547237077, | |
| "grad_norm": 0.05755012482404709, | |
| "learning_rate": 9.818181818181817e-05, | |
| "loss": 0.1689, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.16488413547237077, | |
| "eval_accuracy": 0.9770498871803284, | |
| "eval_loss": 0.049294665455818176, | |
| "eval_runtime": 534.2657, | |
| "eval_samples_per_second": 8.4, | |
| "eval_steps_per_second": 2.1, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.16934046345811052, | |
| "grad_norm": 0.2694285809993744, | |
| "learning_rate": 9.272727272727271e-05, | |
| "loss": 0.0219, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.16934046345811052, | |
| "eval_accuracy": 0.9402852058410645, | |
| "eval_loss": 0.14428526163101196, | |
| "eval_runtime": 531.3442, | |
| "eval_samples_per_second": 8.447, | |
| "eval_steps_per_second": 2.112, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.17379679144385027, | |
| "grad_norm": 0.12420541048049927, | |
| "learning_rate": 8.727272727272726e-05, | |
| "loss": 0.0951, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.17379679144385027, | |
| "eval_accuracy": 0.9977718591690063, | |
| "eval_loss": 0.012917960062623024, | |
| "eval_runtime": 535.2048, | |
| "eval_samples_per_second": 8.386, | |
| "eval_steps_per_second": 2.096, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.17825311942959002, | |
| "grad_norm": 0.25717079639434814, | |
| "learning_rate": 8.18181818181818e-05, | |
| "loss": 0.0244, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.17825311942959002, | |
| "eval_accuracy": 0.9968805909156799, | |
| "eval_loss": 0.010928132571280003, | |
| "eval_runtime": 531.9808, | |
| "eval_samples_per_second": 8.436, | |
| "eval_steps_per_second": 2.109, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.18270944741532977, | |
| "grad_norm": 8.522324562072754, | |
| "learning_rate": 7.636363636363635e-05, | |
| "loss": 0.0142, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.18270944741532977, | |
| "eval_accuracy": 0.9942067861557007, | |
| "eval_loss": 0.01725505292415619, | |
| "eval_runtime": 536.5062, | |
| "eval_samples_per_second": 8.365, | |
| "eval_steps_per_second": 2.091, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.18716577540106952, | |
| "grad_norm": 0.08865547925233841, | |
| "learning_rate": 7.09090909090909e-05, | |
| "loss": 0.0702, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.18716577540106952, | |
| "eval_accuracy": 0.9772727489471436, | |
| "eval_loss": 0.060311682522296906, | |
| "eval_runtime": 532.5572, | |
| "eval_samples_per_second": 8.427, | |
| "eval_steps_per_second": 2.107, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.19162210338680927, | |
| "grad_norm": 0.02239236794412136, | |
| "learning_rate": 6.545454545454545e-05, | |
| "loss": 0.0209, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.19162210338680927, | |
| "eval_accuracy": 0.9904189109802246, | |
| "eval_loss": 0.02353021316230297, | |
| "eval_runtime": 538.0122, | |
| "eval_samples_per_second": 8.342, | |
| "eval_steps_per_second": 2.085, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.19607843137254902, | |
| "grad_norm": 14.454584121704102, | |
| "learning_rate": 5.9999999999999995e-05, | |
| "loss": 0.0914, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.19607843137254902, | |
| "eval_accuracy": 0.9939839839935303, | |
| "eval_loss": 0.01657612808048725, | |
| "eval_runtime": 532.1826, | |
| "eval_samples_per_second": 8.433, | |
| "eval_steps_per_second": 2.108, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.20053475935828877, | |
| "grad_norm": 0.065692238509655, | |
| "learning_rate": 5.454545454545454e-05, | |
| "loss": 0.0645, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.20053475935828877, | |
| "eval_accuracy": 0.9886363744735718, | |
| "eval_loss": 0.029221026226878166, | |
| "eval_runtime": 532.6751, | |
| "eval_samples_per_second": 8.425, | |
| "eval_steps_per_second": 2.106, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.20499108734402852, | |
| "grad_norm": 0.03740346431732178, | |
| "learning_rate": 4.9090909090909084e-05, | |
| "loss": 0.0027, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.20499108734402852, | |
| "eval_accuracy": 0.9848484992980957, | |
| "eval_loss": 0.04155493900179863, | |
| "eval_runtime": 531.2072, | |
| "eval_samples_per_second": 8.449, | |
| "eval_steps_per_second": 2.112, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.20944741532976827, | |
| "grad_norm": 0.31382623314857483, | |
| "learning_rate": 4.363636363636363e-05, | |
| "loss": 0.1536, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.20944741532976827, | |
| "eval_accuracy": 0.9879679083824158, | |
| "eval_loss": 0.03183320537209511, | |
| "eval_runtime": 531.0891, | |
| "eval_samples_per_second": 8.451, | |
| "eval_steps_per_second": 2.113, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.21390374331550802, | |
| "grad_norm": 0.028712157160043716, | |
| "learning_rate": 3.8181818181818174e-05, | |
| "loss": 0.086, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.21390374331550802, | |
| "eval_accuracy": 0.987522304058075, | |
| "eval_loss": 0.032387156039476395, | |
| "eval_runtime": 538.4057, | |
| "eval_samples_per_second": 8.336, | |
| "eval_steps_per_second": 2.084, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.21836007130124777, | |
| "grad_norm": 0.016576524823904037, | |
| "learning_rate": 3.2727272727272725e-05, | |
| "loss": 0.0223, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.21836007130124777, | |
| "eval_accuracy": 0.9955436587333679, | |
| "eval_loss": 0.011954777874052525, | |
| "eval_runtime": 530.6249, | |
| "eval_samples_per_second": 8.458, | |
| "eval_steps_per_second": 2.114, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.22281639928698752, | |
| "grad_norm": 0.014118828810751438, | |
| "learning_rate": 2.727272727272727e-05, | |
| "loss": 0.0235, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.22281639928698752, | |
| "eval_accuracy": 0.9962121248245239, | |
| "eval_loss": 0.010361210443079472, | |
| "eval_runtime": 539.7181, | |
| "eval_samples_per_second": 8.315, | |
| "eval_steps_per_second": 2.079, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.22727272727272727, | |
| "grad_norm": 0.027222217991948128, | |
| "learning_rate": 2.1818181818181814e-05, | |
| "loss": 0.0025, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.22727272727272727, | |
| "eval_accuracy": 0.9966577291488647, | |
| "eval_loss": 0.00987223070114851, | |
| "eval_runtime": 530.4418, | |
| "eval_samples_per_second": 8.461, | |
| "eval_steps_per_second": 2.115, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.23172905525846701, | |
| "grad_norm": 9.495808601379395, | |
| "learning_rate": 1.6363636363636363e-05, | |
| "loss": 0.0254, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.23172905525846701, | |
| "eval_accuracy": 0.9962121248245239, | |
| "eval_loss": 0.010062512941658497, | |
| "eval_runtime": 532.4144, | |
| "eval_samples_per_second": 8.43, | |
| "eval_steps_per_second": 2.107, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.23618538324420676, | |
| "grad_norm": 0.02721397951245308, | |
| "learning_rate": 1.0909090909090907e-05, | |
| "loss": 0.0741, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.23618538324420676, | |
| "eval_accuracy": 0.9944295883178711, | |
| "eval_loss": 0.014333564788103104, | |
| "eval_runtime": 533.0522, | |
| "eval_samples_per_second": 8.419, | |
| "eval_steps_per_second": 2.105, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.24064171122994651, | |
| "grad_norm": 0.20209485292434692, | |
| "learning_rate": 5.454545454545454e-06, | |
| "loss": 0.0305, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.24064171122994651, | |
| "eval_accuracy": 0.9935383200645447, | |
| "eval_loss": 0.016204852610826492, | |
| "eval_runtime": 531.1995, | |
| "eval_samples_per_second": 8.449, | |
| "eval_steps_per_second": 2.112, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.24509803921568626, | |
| "grad_norm": 7.379052639007568, | |
| "learning_rate": 0.0, | |
| "loss": 0.007, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.24509803921568626, | |
| "eval_accuracy": 0.9935383200645447, | |
| "eval_loss": 0.016523057594895363, | |
| "eval_runtime": 537.9402, | |
| "eval_samples_per_second": 8.343, | |
| "eval_steps_per_second": 2.086, | |
| "step": 550 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 550, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.326400520422712e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |