| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9843478260869567, | |
| "eval_steps": 50, | |
| "global_step": 429, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.034782608695652174, | |
| "grad_norm": 27.628698806495482, | |
| "learning_rate": 5e-07, | |
| "loss": 1.6404, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.06956521739130435, | |
| "grad_norm": 14.379307335524663, | |
| "learning_rate": 1e-06, | |
| "loss": 1.5079, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.10434782608695652, | |
| "grad_norm": 7.6171415097012805, | |
| "learning_rate": 9.99648681635985e-07, | |
| "loss": 1.2858, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1391304347826087, | |
| "grad_norm": 5.902380140367801, | |
| "learning_rate": 9.985952202423114e-07, | |
| "loss": 1.2292, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.17391304347826086, | |
| "grad_norm": 5.431235666701782, | |
| "learning_rate": 9.96841096220313e-07, | |
| "loss": 1.2035, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.20869565217391303, | |
| "grad_norm": 5.011022329519287, | |
| "learning_rate": 9.943887745939163e-07, | |
| "loss": 1.1615, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.24347826086956523, | |
| "grad_norm": 4.933604072753844, | |
| "learning_rate": 9.912417015456088e-07, | |
| "loss": 1.1627, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.2782608695652174, | |
| "grad_norm": 4.904429773196176, | |
| "learning_rate": 9.874042995736093e-07, | |
| "loss": 1.1468, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3130434782608696, | |
| "grad_norm": 4.566916992899494, | |
| "learning_rate": 9.828819612770495e-07, | |
| "loss": 1.1457, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 4.618513047328418, | |
| "learning_rate": 9.77681041777897e-07, | |
| "loss": 1.1363, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "eval_loss": 1.1214656829833984, | |
| "eval_runtime": 71.1557, | |
| "eval_samples_per_second": 57.423, | |
| "eval_steps_per_second": 0.899, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3826086956521739, | |
| "grad_norm": 4.482447125747761, | |
| "learning_rate": 9.718088497902707e-07, | |
| "loss": 1.1252, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.41739130434782606, | |
| "grad_norm": 4.746315495183657, | |
| "learning_rate": 9.652736373497e-07, | |
| "loss": 1.1144, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.45217391304347826, | |
| "grad_norm": 4.800644875448408, | |
| "learning_rate": 9.580845882167572e-07, | |
| "loss": 1.1048, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.48695652173913045, | |
| "grad_norm": 5.041563912063608, | |
| "learning_rate": 9.502518049713631e-07, | |
| "loss": 1.0967, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.5217391304347826, | |
| "grad_norm": 4.834159384378619, | |
| "learning_rate": 9.417862948158997e-07, | |
| "loss": 1.0922, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.5565217391304348, | |
| "grad_norm": 4.966276459660724, | |
| "learning_rate": 9.326999541070803e-07, | |
| "loss": 1.1065, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.591304347826087, | |
| "grad_norm": 4.651039844387456, | |
| "learning_rate": 9.23005551638316e-07, | |
| "loss": 1.08, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.6260869565217392, | |
| "grad_norm": 4.61712494812361, | |
| "learning_rate": 9.127167106960681e-07, | |
| "loss": 1.0871, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6608695652173913, | |
| "grad_norm": 4.965649778107791, | |
| "learning_rate": 9.018478899154066e-07, | |
| "loss": 1.0791, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 4.710879346221803, | |
| "learning_rate": 8.904143629616732e-07, | |
| "loss": 1.0856, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "eval_loss": 1.074812412261963, | |
| "eval_runtime": 70.7335, | |
| "eval_samples_per_second": 57.766, | |
| "eval_steps_per_second": 0.905, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.7304347826086957, | |
| "grad_norm": 4.887377540999055, | |
| "learning_rate": 8.784321970668053e-07, | |
| "loss": 1.0638, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.7652173913043478, | |
| "grad_norm": 4.804331292468611, | |
| "learning_rate": 8.659182304504808e-07, | |
| "loss": 1.0768, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 4.78084183802074, | |
| "learning_rate": 8.528900486578158e-07, | |
| "loss": 1.0838, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.8347826086956521, | |
| "grad_norm": 5.189973322685715, | |
| "learning_rate": 8.393659598468642e-07, | |
| "loss": 1.0633, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "grad_norm": 5.02605525534553, | |
| "learning_rate": 8.253649690606494e-07, | |
| "loss": 1.0748, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.9043478260869565, | |
| "grad_norm": 5.433069400847481, | |
| "learning_rate": 8.10906751519882e-07, | |
| "loss": 1.0623, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.9391304347826087, | |
| "grad_norm": 4.770051175238624, | |
| "learning_rate": 7.960116249738937e-07, | |
| "loss": 1.0396, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.9739130434782609, | |
| "grad_norm": 4.606677098055935, | |
| "learning_rate": 7.807005211486444e-07, | |
| "loss": 1.0662, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.008695652173913, | |
| "grad_norm": 5.498117473835195, | |
| "learning_rate": 7.649949563319227e-07, | |
| "loss": 1.0155, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "grad_norm": 4.9633963344101595, | |
| "learning_rate": 7.489170011370779e-07, | |
| "loss": 0.9426, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "eval_loss": 1.0552960634231567, | |
| "eval_runtime": 70.6918, | |
| "eval_samples_per_second": 57.8, | |
| "eval_steps_per_second": 0.905, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0782608695652174, | |
| "grad_norm": 5.4731038473793285, | |
| "learning_rate": 7.324892494877733e-07, | |
| "loss": 0.9222, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.1130434782608696, | |
| "grad_norm": 5.024873028080845, | |
| "learning_rate": 7.15734786867344e-07, | |
| "loss": 0.9223, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.1478260869565218, | |
| "grad_norm": 5.363614939434412, | |
| "learning_rate": 6.986771578773811e-07, | |
| "loss": 0.937, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.182608695652174, | |
| "grad_norm": 5.175874733453326, | |
| "learning_rate": 6.81340333151128e-07, | |
| "loss": 0.9173, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.2173913043478262, | |
| "grad_norm": 5.446257822096013, | |
| "learning_rate": 6.637486756681842e-07, | |
| "loss": 0.9232, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.2521739130434781, | |
| "grad_norm": 5.0358352027359174, | |
| "learning_rate": 6.459269065178591e-07, | |
| "loss": 0.9146, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.2869565217391306, | |
| "grad_norm": 5.287700886552525, | |
| "learning_rate": 6.279000701592794e-07, | |
| "loss": 0.919, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.3217391304347825, | |
| "grad_norm": 5.280625369358825, | |
| "learning_rate": 6.096934992270767e-07, | |
| "loss": 0.9201, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.3565217391304347, | |
| "grad_norm": 5.218195920942847, | |
| "learning_rate": 5.913327789321077e-07, | |
| "loss": 0.9151, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 5.2780690080724515, | |
| "learning_rate": 5.728437111072375e-07, | |
| "loss": 0.9144, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "eval_loss": 1.043231725692749, | |
| "eval_runtime": 70.6101, | |
| "eval_samples_per_second": 57.867, | |
| "eval_steps_per_second": 0.906, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.4260869565217391, | |
| "grad_norm": 5.5857239509669965, | |
| "learning_rate": 5.542522779487071e-07, | |
| "loss": 0.9173, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.4608695652173913, | |
| "grad_norm": 5.283495401802204, | |
| "learning_rate": 5.355846055040448e-07, | |
| "loss": 0.9102, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.4956521739130435, | |
| "grad_norm": 5.21771322708205, | |
| "learning_rate": 5.168669269578232e-07, | |
| "loss": 0.9199, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.5304347826086957, | |
| "grad_norm": 5.2351062433238935, | |
| "learning_rate": 4.981255457668624e-07, | |
| "loss": 0.9233, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.5652173913043477, | |
| "grad_norm": 5.815315560161152, | |
| "learning_rate": 4.793867986966802e-07, | |
| "loss": 0.9091, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 5.43656366363975, | |
| "learning_rate": 4.606770188111338e-07, | |
| "loss": 0.8952, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.634782608695652, | |
| "grad_norm": 5.330638617903944, | |
| "learning_rate": 4.420224984672653e-07, | |
| "loss": 0.9053, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.6695652173913045, | |
| "grad_norm": 5.555943570818251, | |
| "learning_rate": 4.2344945236734963e-07, | |
| "loss": 0.9109, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.7043478260869565, | |
| "grad_norm": 5.513097486509469, | |
| "learning_rate": 4.049839807200688e-07, | |
| "loss": 0.9037, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 5.47180452288034, | |
| "learning_rate": 3.866520325625825e-07, | |
| "loss": 0.9069, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "eval_loss": 1.029205083847046, | |
| "eval_runtime": 70.7173, | |
| "eval_samples_per_second": 57.779, | |
| "eval_steps_per_second": 0.905, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.7739130434782608, | |
| "grad_norm": 5.680893425776684, | |
| "learning_rate": 3.684793692950344e-07, | |
| "loss": 0.879, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.808695652173913, | |
| "grad_norm": 5.872951088572936, | |
| "learning_rate": 3.504915284787405e-07, | |
| "loss": 0.8932, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.8434782608695652, | |
| "grad_norm": 5.990419512227443, | |
| "learning_rate": 3.327137879489312e-07, | |
| "loss": 0.894, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.8782608695652174, | |
| "grad_norm": 5.373969959306303, | |
| "learning_rate": 3.1517113029248233e-07, | |
| "loss": 0.9063, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.9130434782608696, | |
| "grad_norm": 5.306825054838355, | |
| "learning_rate": 2.9788820774054697e-07, | |
| "loss": 0.8991, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.9478260869565216, | |
| "grad_norm": 5.452637165713125, | |
| "learning_rate": 2.8088930752543063e-07, | |
| "loss": 0.8986, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.982608695652174, | |
| "grad_norm": 5.460619773030644, | |
| "learning_rate": 2.641983177503876e-07, | |
| "loss": 0.8953, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 2.017391304347826, | |
| "grad_norm": 6.270694873773961, | |
| "learning_rate": 2.4783869382030424e-07, | |
| "loss": 0.8563, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.0521739130434784, | |
| "grad_norm": 5.795044001331771, | |
| "learning_rate": 2.3183342548044065e-07, | |
| "loss": 0.7987, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 6.0960236861916055, | |
| "learning_rate": 2.1620500450955221e-07, | |
| "loss": 0.8152, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "eval_loss": 1.039559245109558, | |
| "eval_runtime": 70.7734, | |
| "eval_samples_per_second": 57.734, | |
| "eval_steps_per_second": 0.904, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.121739130434783, | |
| "grad_norm": 5.769696534143466, | |
| "learning_rate": 2.0097539311278898e-07, | |
| "loss": 0.8055, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 2.1565217391304348, | |
| "grad_norm": 6.054826333866415, | |
| "learning_rate": 1.8616599305879333e-07, | |
| "loss": 0.8184, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.1913043478260867, | |
| "grad_norm": 5.8177110155805405, | |
| "learning_rate": 1.7179761560436097e-07, | |
| "loss": 0.8201, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 2.226086956521739, | |
| "grad_norm": 5.966607504919806, | |
| "learning_rate": 1.5789045224893538e-07, | |
| "loss": 0.8166, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.260869565217391, | |
| "grad_norm": 5.59992033676923, | |
| "learning_rate": 1.444640463600293e-07, | |
| "loss": 0.8265, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.2956521739130435, | |
| "grad_norm": 5.804095502408287, | |
| "learning_rate": 1.3153726570944828e-07, | |
| "loss": 0.8008, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.3304347826086955, | |
| "grad_norm": 5.39590256328954, | |
| "learning_rate": 1.1912827595891312e-07, | |
| "loss": 0.8131, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 2.365217391304348, | |
| "grad_norm": 5.653819193920734, | |
| "learning_rate": 1.0725451513233674e-07, | |
| "loss": 0.8219, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 5.716140358883574, | |
| "learning_rate": 9.593266911063253e-08, | |
| "loss": 0.8005, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 2.4347826086956523, | |
| "grad_norm": 6.028759811444573, | |
| "learning_rate": 8.517864818348803e-08, | |
| "loss": 0.8141, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.4347826086956523, | |
| "eval_loss": 1.0375770330429077, | |
| "eval_runtime": 70.8262, | |
| "eval_samples_per_second": 57.691, | |
| "eval_steps_per_second": 0.904, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.4695652173913043, | |
| "grad_norm": 5.635963856252332, | |
| "learning_rate": 7.500756469105818e-08, | |
| "loss": 0.8119, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 2.5043478260869563, | |
| "grad_norm": 6.062141717487416, | |
| "learning_rate": 6.543371178699442e-08, | |
| "loss": 0.7997, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.5391304347826087, | |
| "grad_norm": 5.747247864530177, | |
| "learning_rate": 5.647054335265489e-08, | |
| "loss": 0.8019, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 2.573913043478261, | |
| "grad_norm": 5.7882061983533415, | |
| "learning_rate": 4.813065509072278e-08, | |
| "loss": 0.7917, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.608695652173913, | |
| "grad_norm": 5.960090404893479, | |
| "learning_rate": 4.0425766824798814e-08, | |
| "loss": 0.8107, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.643478260869565, | |
| "grad_norm": 6.220656601688518, | |
| "learning_rate": 3.3366706029845096e-08, | |
| "loss": 0.8191, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.6782608695652175, | |
| "grad_norm": 5.532579400575541, | |
| "learning_rate": 2.696339261662156e-08, | |
| "loss": 0.8074, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 2.7130434782608694, | |
| "grad_norm": 5.970542774366677, | |
| "learning_rate": 2.122482499149869e-08, | |
| "loss": 0.802, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.747826086956522, | |
| "grad_norm": 5.973565867263469, | |
| "learning_rate": 1.6159067411235737e-08, | |
| "loss": 0.7993, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 5.77223311406108, | |
| "learning_rate": 1.177323865049512e-08, | |
| "loss": 0.8097, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "eval_loss": 1.0366120338439941, | |
| "eval_runtime": 70.731, | |
| "eval_samples_per_second": 57.768, | |
| "eval_steps_per_second": 0.905, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.8173913043478263, | |
| "grad_norm": 5.721453785688962, | |
| "learning_rate": 8.073501998017152e-09, | |
| "loss": 0.7976, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 2.8521739130434782, | |
| "grad_norm": 5.82427897146351, | |
| "learning_rate": 5.065056595513984e-09, | |
| "loss": 0.8042, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.8869565217391306, | |
| "grad_norm": 6.051738285563, | |
| "learning_rate": 2.752130131453756e-09, | |
| "loss": 0.797, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 2.9217391304347826, | |
| "grad_norm": 6.0320734666727045, | |
| "learning_rate": 1.137972900002171e-09, | |
| "loss": 0.8181, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.9565217391304346, | |
| "grad_norm": 5.640940082183562, | |
| "learning_rate": 2.2485323347054552e-10, | |
| "loss": 0.8073, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.9843478260869567, | |
| "step": 429, | |
| "total_flos": 2529344492666880.0, | |
| "train_loss": 0.9555719093287186, | |
| "train_runtime": 6085.5766, | |
| "train_samples_per_second": 18.127, | |
| "train_steps_per_second": 0.07 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 429, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2529344492666880.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |