| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 50.0, | |
| "eval_steps": 50, | |
| "global_step": 5300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 8.679245283018868e-07, | |
| "loss": 130.658, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_runtime": 2.1932, | |
| "eval_samples_per_second": 327.381, | |
| "eval_steps_per_second": 10.487, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 2.716981132075472e-06, | |
| "loss": 119.614, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_runtime": 1.9516, | |
| "eval_samples_per_second": 367.905, | |
| "eval_steps_per_second": 11.785, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 4.603773584905661e-06, | |
| "loss": 84.4188, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_runtime": 1.9342, | |
| "eval_samples_per_second": 371.215, | |
| "eval_steps_per_second": 11.891, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 6.490566037735849e-06, | |
| "loss": 53.1678, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_runtime": 1.9027, | |
| "eval_samples_per_second": 377.349, | |
| "eval_steps_per_second": 12.088, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 8.377358490566039e-06, | |
| "loss": 33.2104, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "eval_runtime": 1.9458, | |
| "eval_samples_per_second": 368.991, | |
| "eval_steps_per_second": 11.82, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 1.0264150943396227e-05, | |
| "loss": 28.033, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "eval_runtime": 1.9188, | |
| "eval_samples_per_second": 374.183, | |
| "eval_steps_per_second": 11.986, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 1.2150943396226416e-05, | |
| "loss": 22.0028, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "eval_runtime": 2.939, | |
| "eval_samples_per_second": 244.303, | |
| "eval_steps_per_second": 7.826, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 1.4037735849056604e-05, | |
| "loss": 17.8244, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "eval_runtime": 2.0678, | |
| "eval_samples_per_second": 347.226, | |
| "eval_steps_per_second": 11.123, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 1.5924528301886794e-05, | |
| "loss": 14.5603, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "eval_runtime": 2.0344, | |
| "eval_samples_per_second": 352.926, | |
| "eval_steps_per_second": 11.305, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 1.7811320754716983e-05, | |
| "loss": 11.8745, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "eval_runtime": 2.0323, | |
| "eval_samples_per_second": 353.287, | |
| "eval_steps_per_second": 11.317, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "learning_rate": 1.969811320754717e-05, | |
| "loss": 9.6856, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "eval_runtime": 2.0201, | |
| "eval_samples_per_second": 355.425, | |
| "eval_steps_per_second": 11.385, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 1.999617436193858e-05, | |
| "loss": 7.8251, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "eval_runtime": 2.0209, | |
| "eval_samples_per_second": 355.286, | |
| "eval_steps_per_second": 11.381, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "learning_rate": 1.9981648322537017e-05, | |
| "loss": 6.5604, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "eval_runtime": 1.9676, | |
| "eval_samples_per_second": 364.903, | |
| "eval_steps_per_second": 11.689, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "learning_rate": 1.995629882570864e-05, | |
| "loss": 5.9968, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "eval_runtime": 1.933, | |
| "eval_samples_per_second": 371.443, | |
| "eval_steps_per_second": 11.899, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 1.992015335881735e-05, | |
| "loss": 5.3714, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "eval_runtime": 1.928, | |
| "eval_samples_per_second": 372.4, | |
| "eval_steps_per_second": 11.929, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 7.55, | |
| "learning_rate": 1.9873251115682577e-05, | |
| "loss": 5.0835, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.55, | |
| "eval_runtime": 1.9194, | |
| "eval_samples_per_second": 374.085, | |
| "eval_steps_per_second": 11.983, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 1.9815642954080055e-05, | |
| "loss": 4.8417, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "eval_runtime": 1.9205, | |
| "eval_samples_per_second": 373.866, | |
| "eval_steps_per_second": 11.976, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 1.97473913405949e-05, | |
| "loss": 4.651, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "eval_runtime": 1.9167, | |
| "eval_samples_per_second": 374.597, | |
| "eval_steps_per_second": 12.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "learning_rate": 1.966857028288687e-05, | |
| "loss": 4.5806, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "eval_runtime": 1.9211, | |
| "eval_samples_per_second": 373.737, | |
| "eval_steps_per_second": 11.972, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "learning_rate": 1.9579265249441216e-05, | |
| "loss": 4.5062, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "eval_runtime": 1.9281, | |
| "eval_samples_per_second": 372.389, | |
| "eval_steps_per_second": 11.929, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.91, | |
| "learning_rate": 1.9479573076892152e-05, | |
| "loss": 4.4466, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 9.91, | |
| "eval_runtime": 1.9203, | |
| "eval_samples_per_second": 373.895, | |
| "eval_steps_per_second": 11.977, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "learning_rate": 1.9369601865019452e-05, | |
| "loss": 4.4225, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "eval_runtime": 1.9182, | |
| "eval_samples_per_second": 374.311, | |
| "eval_steps_per_second": 11.99, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 10.85, | |
| "learning_rate": 1.9249470859531976e-05, | |
| "loss": 4.4022, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 10.85, | |
| "eval_runtime": 1.9262, | |
| "eval_samples_per_second": 372.754, | |
| "eval_steps_per_second": 11.941, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 11.32, | |
| "learning_rate": 1.9119310322765315e-05, | |
| "loss": 4.3985, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 11.32, | |
| "eval_runtime": 1.94, | |
| "eval_samples_per_second": 370.112, | |
| "eval_steps_per_second": 11.856, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 11.79, | |
| "learning_rate": 1.8979261392433685e-05, | |
| "loss": 4.3822, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 11.79, | |
| "eval_runtime": 1.9393, | |
| "eval_samples_per_second": 370.229, | |
| "eval_steps_per_second": 11.86, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 12.26, | |
| "learning_rate": 1.8829475928589272e-05, | |
| "loss": 4.3754, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 12.26, | |
| "eval_runtime": 1.9864, | |
| "eval_samples_per_second": 361.455, | |
| "eval_steps_per_second": 11.579, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 12.74, | |
| "learning_rate": 1.8670116348954945e-05, | |
| "loss": 4.3684, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 12.74, | |
| "eval_runtime": 1.9783, | |
| "eval_samples_per_second": 362.937, | |
| "eval_steps_per_second": 11.626, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 13.21, | |
| "learning_rate": 1.850135545280894e-05, | |
| "loss": 4.3725, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 13.21, | |
| "eval_runtime": 1.9796, | |
| "eval_samples_per_second": 362.707, | |
| "eval_steps_per_second": 11.619, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 13.68, | |
| "learning_rate": 1.832337623361242e-05, | |
| "loss": 4.3738, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 13.68, | |
| "eval_runtime": 1.9672, | |
| "eval_samples_per_second": 364.987, | |
| "eval_steps_per_second": 11.692, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "learning_rate": 1.8136371680583176e-05, | |
| "loss": 4.3524, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "eval_runtime": 1.917, | |
| "eval_samples_per_second": 374.54, | |
| "eval_steps_per_second": 11.998, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 14.62, | |
| "learning_rate": 1.7940544569430468e-05, | |
| "loss": 4.3487, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 14.62, | |
| "eval_runtime": 1.9212, | |
| "eval_samples_per_second": 373.718, | |
| "eval_steps_per_second": 11.971, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 15.09, | |
| "learning_rate": 1.7736107242478143e-05, | |
| "loss": 4.347, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 15.09, | |
| "eval_runtime": 1.9221, | |
| "eval_samples_per_second": 373.543, | |
| "eval_steps_per_second": 11.966, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 15.57, | |
| "learning_rate": 1.7523281378414246e-05, | |
| "loss": 4.3417, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 15.57, | |
| "eval_runtime": 2.1716, | |
| "eval_samples_per_second": 330.631, | |
| "eval_steps_per_second": 10.591, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 16.04, | |
| "learning_rate": 1.730229775191693e-05, | |
| "loss": 4.3224, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 16.04, | |
| "eval_runtime": 1.9519, | |
| "eval_samples_per_second": 367.842, | |
| "eval_steps_per_second": 11.783, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 16.51, | |
| "learning_rate": 1.7073395983417227e-05, | |
| "loss": 4.3169, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 16.51, | |
| "eval_runtime": 1.9446, | |
| "eval_samples_per_second": 369.221, | |
| "eval_steps_per_second": 11.827, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 16.98, | |
| "learning_rate": 1.6836824279270053e-05, | |
| "loss": 4.3111, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 16.98, | |
| "eval_runtime": 1.9452, | |
| "eval_samples_per_second": 369.117, | |
| "eval_steps_per_second": 11.824, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 17.45, | |
| "learning_rate": 1.6592839162615223e-05, | |
| "loss": 4.3045, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 17.45, | |
| "eval_runtime": 1.9364, | |
| "eval_samples_per_second": 370.798, | |
| "eval_steps_per_second": 11.878, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 17.92, | |
| "learning_rate": 1.6341705195220257e-05, | |
| "loss": 4.2946, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 17.92, | |
| "eval_runtime": 1.9407, | |
| "eval_samples_per_second": 369.973, | |
| "eval_steps_per_second": 11.852, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 18.4, | |
| "learning_rate": 1.6083694690606592e-05, | |
| "loss": 4.2868, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 18.4, | |
| "eval_runtime": 1.9401, | |
| "eval_samples_per_second": 370.079, | |
| "eval_steps_per_second": 11.855, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 18.87, | |
| "learning_rate": 1.581908741877034e-05, | |
| "loss": 4.2791, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 18.87, | |
| "eval_runtime": 1.9388, | |
| "eval_samples_per_second": 370.327, | |
| "eval_steps_per_second": 11.863, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 19.34, | |
| "learning_rate": 1.5548170302817683e-05, | |
| "loss": 4.2596, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 19.34, | |
| "eval_runtime": 1.9504, | |
| "eval_samples_per_second": 368.122, | |
| "eval_steps_per_second": 11.792, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 19.81, | |
| "learning_rate": 1.5271237107843925e-05, | |
| "loss": 4.253, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 19.81, | |
| "eval_runtime": 1.9966, | |
| "eval_samples_per_second": 359.617, | |
| "eval_steps_per_second": 11.52, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 20.28, | |
| "learning_rate": 1.4988588122393497e-05, | |
| "loss": 4.2556, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 20.28, | |
| "eval_runtime": 2.0261, | |
| "eval_samples_per_second": 354.367, | |
| "eval_steps_per_second": 11.352, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 20.75, | |
| "learning_rate": 1.47005298328464e-05, | |
| "loss": 4.2289, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 20.75, | |
| "eval_runtime": 2.0081, | |
| "eval_samples_per_second": 357.56, | |
| "eval_steps_per_second": 11.454, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 21.23, | |
| "learning_rate": 1.4407374591084064e-05, | |
| "loss": 4.211, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 21.23, | |
| "eval_runtime": 1.925, | |
| "eval_samples_per_second": 372.984, | |
| "eval_steps_per_second": 11.948, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 21.7, | |
| "learning_rate": 1.4109440275795071e-05, | |
| "loss": 4.2125, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 21.7, | |
| "eval_runtime": 1.9426, | |
| "eval_samples_per_second": 369.601, | |
| "eval_steps_per_second": 11.84, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 22.17, | |
| "learning_rate": 1.3807049947787954e-05, | |
| "loss": 4.1935, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 22.17, | |
| "eval_runtime": 1.944, | |
| "eval_samples_per_second": 369.342, | |
| "eval_steps_per_second": 11.831, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 22.64, | |
| "learning_rate": 1.3500531499684819e-05, | |
| "loss": 4.1803, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 22.64, | |
| "eval_runtime": 1.9484, | |
| "eval_samples_per_second": 368.517, | |
| "eval_steps_per_second": 11.805, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 23.11, | |
| "learning_rate": 1.3190217300375694e-05, | |
| "loss": 4.173, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 23.11, | |
| "eval_runtime": 1.9375, | |
| "eval_samples_per_second": 370.574, | |
| "eval_steps_per_second": 11.871, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 23.58, | |
| "learning_rate": 1.2876443834619066e-05, | |
| "loss": 4.1564, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 23.58, | |
| "eval_runtime": 1.9431, | |
| "eval_samples_per_second": 369.514, | |
| "eval_steps_per_second": 11.837, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 24.06, | |
| "learning_rate": 1.2559551338179468e-05, | |
| "loss": 4.1486, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 24.06, | |
| "eval_runtime": 1.9508, | |
| "eval_samples_per_second": 368.046, | |
| "eval_steps_per_second": 11.79, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 24.53, | |
| "learning_rate": 1.2239883428897687e-05, | |
| "loss": 4.1223, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 24.53, | |
| "eval_runtime": 1.9403, | |
| "eval_samples_per_second": 370.043, | |
| "eval_steps_per_second": 11.854, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 1.1917786734093682e-05, | |
| "loss": 4.1107, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_runtime": 1.9435, | |
| "eval_samples_per_second": 369.437, | |
| "eval_steps_per_second": 11.834, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 25.47, | |
| "learning_rate": 1.1593610514706217e-05, | |
| "loss": 4.0779, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 25.47, | |
| "eval_runtime": 1.9405, | |
| "eval_samples_per_second": 370.004, | |
| "eval_steps_per_second": 11.853, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 25.94, | |
| "learning_rate": 1.1267706286576759e-05, | |
| "loss": 4.0284, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 25.94, | |
| "eval_runtime": 1.944, | |
| "eval_samples_per_second": 369.342, | |
| "eval_steps_per_second": 11.831, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 26.42, | |
| "learning_rate": 1.094042743928831e-05, | |
| "loss": 3.9635, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 26.42, | |
| "eval_runtime": 3.7643, | |
| "eval_samples_per_second": 190.741, | |
| "eval_steps_per_second": 6.11, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 26.89, | |
| "learning_rate": 1.0612128852972474e-05, | |
| "loss": 3.9045, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 26.89, | |
| "eval_runtime": 1.9311, | |
| "eval_samples_per_second": 371.804, | |
| "eval_steps_per_second": 11.91, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 27.36, | |
| "learning_rate": 1.0283166513500267e-05, | |
| "loss": 3.8146, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 27.36, | |
| "eval_runtime": 1.9224, | |
| "eval_samples_per_second": 373.485, | |
| "eval_steps_per_second": 11.964, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 27.83, | |
| "learning_rate": 9.953897126473933e-06, | |
| "loss": 3.7245, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 27.83, | |
| "eval_runtime": 1.9864, | |
| "eval_samples_per_second": 361.449, | |
| "eval_steps_per_second": 11.578, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 28.3, | |
| "learning_rate": 9.624677730438344e-06, | |
| "loss": 3.6334, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 28.3, | |
| "eval_runtime": 1.9535, | |
| "eval_samples_per_second": 367.553, | |
| "eval_steps_per_second": 11.774, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 28.77, | |
| "learning_rate": 9.295865309731342e-06, | |
| "loss": 3.5344, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 28.77, | |
| "eval_runtime": 1.9514, | |
| "eval_samples_per_second": 367.939, | |
| "eval_steps_per_second": 11.786, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 29.25, | |
| "learning_rate": 8.96781640739291e-06, | |
| "loss": 3.4395, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 29.25, | |
| "eval_runtime": 1.9254, | |
| "eval_samples_per_second": 372.918, | |
| "eval_steps_per_second": 11.946, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 29.72, | |
| "learning_rate": 8.64088673855282e-06, | |
| "loss": 3.3242, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 29.72, | |
| "eval_runtime": 1.9225, | |
| "eval_samples_per_second": 373.464, | |
| "eval_steps_per_second": 11.963, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 30.19, | |
| "learning_rate": 8.315430804716022e-06, | |
| "loss": 3.205, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 30.19, | |
| "eval_runtime": 1.9123, | |
| "eval_samples_per_second": 375.463, | |
| "eval_steps_per_second": 12.027, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 30.66, | |
| "learning_rate": 7.991801509364023e-06, | |
| "loss": 3.1021, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 30.66, | |
| "eval_runtime": 1.9257, | |
| "eval_samples_per_second": 372.855, | |
| "eval_steps_per_second": 11.944, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 31.13, | |
| "learning_rate": 7.670349775289047e-06, | |
| "loss": 2.9841, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 31.13, | |
| "eval_runtime": 1.9263, | |
| "eval_samples_per_second": 372.732, | |
| "eval_steps_per_second": 11.94, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 31.6, | |
| "learning_rate": 7.3514241640759175e-06, | |
| "loss": 2.8855, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 31.6, | |
| "eval_runtime": 1.9276, | |
| "eval_samples_per_second": 372.486, | |
| "eval_steps_per_second": 11.932, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 32.08, | |
| "learning_rate": 7.035370498144325e-06, | |
| "loss": 2.8144, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 32.08, | |
| "eval_runtime": 1.9176, | |
| "eval_samples_per_second": 374.418, | |
| "eval_steps_per_second": 11.994, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 32.55, | |
| "learning_rate": 6.722531485761199e-06, | |
| "loss": 2.701, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 32.55, | |
| "eval_runtime": 1.9236, | |
| "eval_samples_per_second": 373.25, | |
| "eval_steps_per_second": 11.956, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 33.02, | |
| "learning_rate": 6.413246349429934e-06, | |
| "loss": 2.615, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 33.02, | |
| "eval_runtime": 1.9223, | |
| "eval_samples_per_second": 373.516, | |
| "eval_steps_per_second": 11.965, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 33.49, | |
| "learning_rate": 6.107850458059322e-06, | |
| "loss": 2.5553, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 33.49, | |
| "eval_runtime": 1.9299, | |
| "eval_samples_per_second": 372.031, | |
| "eval_steps_per_second": 11.917, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 33.96, | |
| "learning_rate": 5.8066749633110675e-06, | |
| "loss": 2.4601, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 33.96, | |
| "eval_runtime": 1.9178, | |
| "eval_samples_per_second": 374.392, | |
| "eval_steps_per_second": 11.993, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 34.43, | |
| "learning_rate": 5.510046440520228e-06, | |
| "loss": 2.4033, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 34.43, | |
| "eval_runtime": 1.9204, | |
| "eval_samples_per_second": 373.879, | |
| "eval_steps_per_second": 11.977, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 34.91, | |
| "learning_rate": 5.218286534577938e-06, | |
| "loss": 2.3601, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 34.91, | |
| "eval_runtime": 1.913, | |
| "eval_samples_per_second": 375.322, | |
| "eval_steps_per_second": 12.023, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 35.38, | |
| "learning_rate": 4.93171161116037e-06, | |
| "loss": 2.3119, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 35.38, | |
| "eval_runtime": 1.9156, | |
| "eval_samples_per_second": 374.813, | |
| "eval_steps_per_second": 12.007, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 35.85, | |
| "learning_rate": 4.656198158017416e-06, | |
| "loss": 2.2631, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 35.85, | |
| "eval_runtime": 1.9287, | |
| "eval_samples_per_second": 372.271, | |
| "eval_steps_per_second": 11.925, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 36.32, | |
| "learning_rate": 4.380800513736635e-06, | |
| "loss": 2.2201, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 36.32, | |
| "eval_runtime": 1.9261, | |
| "eval_samples_per_second": 372.768, | |
| "eval_steps_per_second": 11.941, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 36.79, | |
| "learning_rate": 4.111495967954926e-06, | |
| "loss": 2.1818, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 36.79, | |
| "eval_runtime": 1.9179, | |
| "eval_samples_per_second": 374.378, | |
| "eval_steps_per_second": 11.993, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 37.26, | |
| "learning_rate": 3.848576537200217e-06, | |
| "loss": 2.1282, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 37.26, | |
| "eval_runtime": 1.921, | |
| "eval_samples_per_second": 373.772, | |
| "eval_steps_per_second": 11.973, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 37.74, | |
| "learning_rate": 3.5923273143923885e-06, | |
| "loss": 2.1174, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 37.74, | |
| "eval_runtime": 1.9202, | |
| "eval_samples_per_second": 373.92, | |
| "eval_steps_per_second": 11.978, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 38.21, | |
| "learning_rate": 3.343026159706837e-06, | |
| "loss": 2.091, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 38.21, | |
| "eval_runtime": 1.9212, | |
| "eval_samples_per_second": 373.727, | |
| "eval_steps_per_second": 11.972, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 38.68, | |
| "learning_rate": 3.1009433992807925e-06, | |
| "loss": 2.0562, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 38.68, | |
| "eval_runtime": 1.9201, | |
| "eval_samples_per_second": 373.932, | |
| "eval_steps_per_second": 11.978, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 39.15, | |
| "learning_rate": 2.8709585976496825e-06, | |
| "loss": 2.0276, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 39.15, | |
| "eval_runtime": 1.9126, | |
| "eval_samples_per_second": 375.405, | |
| "eval_steps_per_second": 12.025, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 39.62, | |
| "learning_rate": 2.6439348660701634e-06, | |
| "loss": 2.0138, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 39.62, | |
| "eval_runtime": 1.918, | |
| "eval_samples_per_second": 374.351, | |
| "eval_steps_per_second": 11.992, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 40.09, | |
| "learning_rate": 2.424887578383799e-06, | |
| "loss": 1.9901, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 40.09, | |
| "eval_runtime": 2.0806, | |
| "eval_samples_per_second": 345.101, | |
| "eval_steps_per_second": 11.055, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 40.57, | |
| "learning_rate": 2.21405425538036e-06, | |
| "loss": 1.9743, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 40.57, | |
| "eval_runtime": 2.0154, | |
| "eval_samples_per_second": 356.253, | |
| "eval_steps_per_second": 11.412, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 41.04, | |
| "learning_rate": 2.011663511154628e-06, | |
| "loss": 1.9583, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 41.04, | |
| "eval_runtime": 2.7415, | |
| "eval_samples_per_second": 261.903, | |
| "eval_steps_per_second": 8.39, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 41.51, | |
| "learning_rate": 1.817934805211976e-06, | |
| "loss": 1.9442, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 41.51, | |
| "eval_runtime": 1.9318, | |
| "eval_samples_per_second": 371.67, | |
| "eval_steps_per_second": 11.906, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 41.98, | |
| "learning_rate": 1.6330782045006088e-06, | |
| "loss": 1.9376, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 41.98, | |
| "eval_runtime": 1.9289, | |
| "eval_samples_per_second": 372.225, | |
| "eval_steps_per_second": 11.924, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 42.45, | |
| "learning_rate": 1.457294155628457e-06, | |
| "loss": 1.919, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 42.45, | |
| "eval_runtime": 1.9343, | |
| "eval_samples_per_second": 371.203, | |
| "eval_steps_per_second": 11.891, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 42.92, | |
| "learning_rate": 1.2907732675117878e-06, | |
| "loss": 1.9033, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 42.92, | |
| "eval_runtime": 1.9245, | |
| "eval_samples_per_second": 373.093, | |
| "eval_steps_per_second": 11.951, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 43.4, | |
| "learning_rate": 1.1336961046911443e-06, | |
| "loss": 1.8907, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 43.4, | |
| "eval_runtime": 1.9188, | |
| "eval_samples_per_second": 374.186, | |
| "eval_steps_per_second": 11.986, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 43.87, | |
| "learning_rate": 9.862329915387669e-07, | |
| "loss": 1.8839, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 43.87, | |
| "eval_runtime": 1.9284, | |
| "eval_samples_per_second": 372.325, | |
| "eval_steps_per_second": 11.927, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 44.34, | |
| "learning_rate": 8.485438275698154e-07, | |
| "loss": 1.8705, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 44.34, | |
| "eval_runtime": 1.9257, | |
| "eval_samples_per_second": 372.846, | |
| "eval_steps_per_second": 11.944, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 44.81, | |
| "learning_rate": 7.207779140576066e-07, | |
| "loss": 1.8764, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 44.81, | |
| "eval_runtime": 1.9191, | |
| "eval_samples_per_second": 374.139, | |
| "eval_steps_per_second": 11.985, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 45.28, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 1.8671, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 45.28, | |
| "eval_runtime": 1.9182, | |
| "eval_samples_per_second": 374.313, | |
| "eval_steps_per_second": 11.991, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 45.75, | |
| "learning_rate": 4.955590925988896e-07, | |
| "loss": 1.8624, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 45.75, | |
| "eval_runtime": 1.9234, | |
| "eval_samples_per_second": 373.294, | |
| "eval_steps_per_second": 11.958, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 46.23, | |
| "learning_rate": 3.983503974564229e-07, | |
| "loss": 1.8639, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 46.23, | |
| "eval_runtime": 1.9237, | |
| "eval_samples_per_second": 373.238, | |
| "eval_steps_per_second": 11.956, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 46.7, | |
| "learning_rate": 3.115531135701155e-07, | |
| "loss": 1.8532, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 46.7, | |
| "eval_runtime": 1.9206, | |
| "eval_samples_per_second": 373.851, | |
| "eval_steps_per_second": 11.976, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 47.17, | |
| "learning_rate": 2.3526135833186527e-07, | |
| "loss": 1.849, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 47.17, | |
| "eval_runtime": 1.9156, | |
| "eval_samples_per_second": 374.809, | |
| "eval_steps_per_second": 12.006, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 47.64, | |
| "learning_rate": 1.6955785761400444e-07, | |
| "loss": 1.844, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 47.64, | |
| "eval_runtime": 1.9207, | |
| "eval_samples_per_second": 373.812, | |
| "eval_steps_per_second": 11.974, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 48.11, | |
| "learning_rate": 1.145138560667003e-07, | |
| "loss": 1.8538, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 48.11, | |
| "eval_runtime": 1.9259, | |
| "eval_samples_per_second": 372.809, | |
| "eval_steps_per_second": 11.942, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 48.58, | |
| "learning_rate": 7.018903986483083e-08, | |
| "loss": 1.8575, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 48.58, | |
| "eval_runtime": 1.9229, | |
| "eval_samples_per_second": 373.387, | |
| "eval_steps_per_second": 11.961, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 49.06, | |
| "learning_rate": 3.663147198813666e-08, | |
| "loss": 1.8503, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 49.06, | |
| "eval_runtime": 1.9176, | |
| "eval_samples_per_second": 374.432, | |
| "eval_steps_per_second": 11.994, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 49.53, | |
| "learning_rate": 1.3877540104818566e-08, | |
| "loss": 1.8424, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 49.53, | |
| "eval_runtime": 1.9158, | |
| "eval_samples_per_second": 374.782, | |
| "eval_steps_per_second": 12.006, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 1.951917115091684e-09, | |
| "loss": 1.8519, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_runtime": 1.9163, | |
| "eval_samples_per_second": 374.689, | |
| "eval_steps_per_second": 12.003, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 5300, | |
| "total_flos": 3.2498053856362496e+16, | |
| "train_loss": 7.9858926362811395, | |
| "train_runtime": 3481.491, | |
| "train_samples_per_second": 48.6, | |
| "train_steps_per_second": 1.522 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 5300, | |
| "num_train_epochs": 50, | |
| "save_steps": 50, | |
| "total_flos": 3.2498053856362496e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |