| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 50, |
| "global_step": 1000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.005, |
| "grad_norm": 88.5, |
| "learning_rate": 2e-05, |
| "loss": 3.2215267181396485, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.01, |
| "grad_norm": 78.0, |
| "learning_rate": 4.5e-05, |
| "loss": 3.2558315277099608, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.015, |
| "grad_norm": 8.9375, |
| "learning_rate": 5e-05, |
| "loss": 3.2168857574462892, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.02, |
| "grad_norm": 65.0, |
| "learning_rate": 5e-05, |
| "loss": 3.222028350830078, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.025, |
| "grad_norm": 6.75, |
| "learning_rate": 5e-05, |
| "loss": 3.170660209655762, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.03, |
| "grad_norm": 9.625, |
| "learning_rate": 5e-05, |
| "loss": 3.2056911468505858, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.035, |
| "grad_norm": 7.5, |
| "learning_rate": 5e-05, |
| "loss": 3.1706233978271485, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.04, |
| "grad_norm": 20.5, |
| "learning_rate": 5e-05, |
| "loss": 3.1928180694580077, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.045, |
| "grad_norm": 30.0, |
| "learning_rate": 5e-05, |
| "loss": 3.1856739044189455, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.05, |
| "grad_norm": 3.171875, |
| "learning_rate": 5e-05, |
| "loss": 3.1777238845825195, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.05, |
| "eval_loss": 3.4269402027130127, |
| "eval_runtime": 6.7462, |
| "eval_samples_per_second": 5.485, |
| "eval_steps_per_second": 2.816, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.055, |
| "grad_norm": 9.5, |
| "learning_rate": 5e-05, |
| "loss": 3.150863265991211, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.06, |
| "grad_norm": 18.0, |
| "learning_rate": 5e-05, |
| "loss": 3.149982452392578, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.065, |
| "grad_norm": 836.0, |
| "learning_rate": 5e-05, |
| "loss": 3.147908592224121, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.07, |
| "grad_norm": 5.34375, |
| "learning_rate": 5e-05, |
| "loss": 3.1658950805664063, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.075, |
| "grad_norm": 4.65625, |
| "learning_rate": 5e-05, |
| "loss": 3.175393295288086, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.08, |
| "grad_norm": 19.875, |
| "learning_rate": 5e-05, |
| "loss": 3.1660781860351563, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.085, |
| "grad_norm": 3.46875, |
| "learning_rate": 5e-05, |
| "loss": 3.155829429626465, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.09, |
| "grad_norm": 9.9375, |
| "learning_rate": 5e-05, |
| "loss": 3.153760528564453, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.095, |
| "grad_norm": 298.0, |
| "learning_rate": 5e-05, |
| "loss": 3.131403160095215, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.1, |
| "grad_norm": 208.0, |
| "learning_rate": 5e-05, |
| "loss": 3.160065269470215, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.1, |
| "eval_loss": 3.3961074352264404, |
| "eval_runtime": 6.6964, |
| "eval_samples_per_second": 5.525, |
| "eval_steps_per_second": 2.837, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.105, |
| "grad_norm": 5.5625, |
| "learning_rate": 5e-05, |
| "loss": 3.15795955657959, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.11, |
| "grad_norm": 77.5, |
| "learning_rate": 5e-05, |
| "loss": 3.112112045288086, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.115, |
| "grad_norm": 16.25, |
| "learning_rate": 5e-05, |
| "loss": 3.1197088241577147, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.12, |
| "grad_norm": 14.25, |
| "learning_rate": 5e-05, |
| "loss": 3.1374229431152343, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.125, |
| "grad_norm": 28.75, |
| "learning_rate": 5e-05, |
| "loss": 3.1513505935668946, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.13, |
| "grad_norm": 6.15625, |
| "learning_rate": 5e-05, |
| "loss": 3.145720863342285, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.135, |
| "grad_norm": 8.125, |
| "learning_rate": 5e-05, |
| "loss": 3.1482744216918945, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.14, |
| "grad_norm": 12.8125, |
| "learning_rate": 5e-05, |
| "loss": 3.124919319152832, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.145, |
| "grad_norm": 432.0, |
| "learning_rate": 5e-05, |
| "loss": 3.122653579711914, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.15, |
| "grad_norm": 38.0, |
| "learning_rate": 5e-05, |
| "loss": 3.1383148193359376, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.15, |
| "eval_loss": 3.3837645053863525, |
| "eval_runtime": 6.6931, |
| "eval_samples_per_second": 5.528, |
| "eval_steps_per_second": 2.839, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.155, |
| "grad_norm": 20.5, |
| "learning_rate": 5e-05, |
| "loss": 3.1009441375732423, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.16, |
| "grad_norm": 7.0625, |
| "learning_rate": 5e-05, |
| "loss": 3.1632896423339845, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.165, |
| "grad_norm": 11.25, |
| "learning_rate": 5e-05, |
| "loss": 3.129340171813965, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.17, |
| "grad_norm": 7.5, |
| "learning_rate": 5e-05, |
| "loss": 3.0880687713623045, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.175, |
| "grad_norm": 264.0, |
| "learning_rate": 5e-05, |
| "loss": 3.1132593154907227, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.18, |
| "grad_norm": 32.5, |
| "learning_rate": 5e-05, |
| "loss": 3.096790313720703, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.185, |
| "grad_norm": 21.5, |
| "learning_rate": 5e-05, |
| "loss": 3.1038497924804687, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.19, |
| "grad_norm": 13.1875, |
| "learning_rate": 5e-05, |
| "loss": 3.1320869445800783, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.195, |
| "grad_norm": 19.0, |
| "learning_rate": 5e-05, |
| "loss": 3.0939069747924806, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.2, |
| "grad_norm": 3.90625, |
| "learning_rate": 5e-05, |
| "loss": 3.084807777404785, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.2, |
| "eval_loss": 3.3661341667175293, |
| "eval_runtime": 6.7101, |
| "eval_samples_per_second": 5.514, |
| "eval_steps_per_second": 2.832, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.205, |
| "grad_norm": 9.0625, |
| "learning_rate": 5e-05, |
| "loss": 3.09683895111084, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.21, |
| "grad_norm": 43.25, |
| "learning_rate": 5e-05, |
| "loss": 3.141134834289551, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.215, |
| "grad_norm": 2.4375, |
| "learning_rate": 5e-05, |
| "loss": 3.100748825073242, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.22, |
| "grad_norm": 12.5625, |
| "learning_rate": 5e-05, |
| "loss": 3.0931968688964844, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.225, |
| "grad_norm": 4.53125, |
| "learning_rate": 5e-05, |
| "loss": 3.0807849884033205, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.23, |
| "grad_norm": 5.03125, |
| "learning_rate": 5e-05, |
| "loss": 3.1373645782470705, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.235, |
| "grad_norm": 482.0, |
| "learning_rate": 5e-05, |
| "loss": 3.086713409423828, |
| "step": 235 |
| }, |
| { |
| "epoch": 0.24, |
| "grad_norm": 8.8125, |
| "learning_rate": 5e-05, |
| "loss": 3.107415199279785, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.245, |
| "grad_norm": 35.25, |
| "learning_rate": 5e-05, |
| "loss": 3.1350580215454102, |
| "step": 245 |
| }, |
| { |
| "epoch": 0.25, |
| "grad_norm": 10.8125, |
| "learning_rate": 5e-05, |
| "loss": 3.085038757324219, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.25, |
| "eval_loss": 3.35124135017395, |
| "eval_runtime": 6.7488, |
| "eval_samples_per_second": 5.482, |
| "eval_steps_per_second": 2.815, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.255, |
| "grad_norm": 7.78125, |
| "learning_rate": 5e-05, |
| "loss": 3.0779041290283202, |
| "step": 255 |
| }, |
| { |
| "epoch": 0.26, |
| "grad_norm": 8.9375, |
| "learning_rate": 5e-05, |
| "loss": 3.0426475524902346, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.265, |
| "grad_norm": 1.9765625, |
| "learning_rate": 5e-05, |
| "loss": 3.108402442932129, |
| "step": 265 |
| }, |
| { |
| "epoch": 0.27, |
| "grad_norm": 10.125, |
| "learning_rate": 5e-05, |
| "loss": 3.068155288696289, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.275, |
| "grad_norm": 25.625, |
| "learning_rate": 5e-05, |
| "loss": 3.0750944137573244, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.28, |
| "grad_norm": 1.609375, |
| "learning_rate": 5e-05, |
| "loss": 3.0479175567626955, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.285, |
| "grad_norm": 50.75, |
| "learning_rate": 5e-05, |
| "loss": 3.088734817504883, |
| "step": 285 |
| }, |
| { |
| "epoch": 0.29, |
| "grad_norm": 8.8125, |
| "learning_rate": 5e-05, |
| "loss": 3.0425872802734375, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.295, |
| "grad_norm": 18.5, |
| "learning_rate": 5e-05, |
| "loss": 3.0785484313964844, |
| "step": 295 |
| }, |
| { |
| "epoch": 0.3, |
| "grad_norm": 7.3125, |
| "learning_rate": 5e-05, |
| "loss": 3.067402648925781, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.3, |
| "eval_loss": 3.339944839477539, |
| "eval_runtime": 6.7462, |
| "eval_samples_per_second": 5.485, |
| "eval_steps_per_second": 2.816, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.305, |
| "grad_norm": 7.25, |
| "learning_rate": 5e-05, |
| "loss": 3.063138961791992, |
| "step": 305 |
| }, |
| { |
| "epoch": 0.31, |
| "grad_norm": 22.5, |
| "learning_rate": 5e-05, |
| "loss": 3.0437475204467774, |
| "step": 310 |
| }, |
| { |
| "epoch": 0.315, |
| "grad_norm": 9.75, |
| "learning_rate": 5e-05, |
| "loss": 3.0719404220581055, |
| "step": 315 |
| }, |
| { |
| "epoch": 0.32, |
| "grad_norm": 8.75, |
| "learning_rate": 5e-05, |
| "loss": 3.0688711166381837, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.325, |
| "grad_norm": 2.671875, |
| "learning_rate": 5e-05, |
| "loss": 3.0548662185668944, |
| "step": 325 |
| }, |
| { |
| "epoch": 0.33, |
| "grad_norm": 12.6875, |
| "learning_rate": 5e-05, |
| "loss": 3.057333564758301, |
| "step": 330 |
| }, |
| { |
| "epoch": 0.335, |
| "grad_norm": 298.0, |
| "learning_rate": 5e-05, |
| "loss": 3.0870180130004883, |
| "step": 335 |
| }, |
| { |
| "epoch": 0.34, |
| "grad_norm": 18.375, |
| "learning_rate": 5e-05, |
| "loss": 3.051515579223633, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.345, |
| "grad_norm": 44.0, |
| "learning_rate": 5e-05, |
| "loss": 3.0610748291015626, |
| "step": 345 |
| }, |
| { |
| "epoch": 0.35, |
| "grad_norm": 11.6875, |
| "learning_rate": 5e-05, |
| "loss": 3.02484016418457, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.35, |
| "eval_loss": 3.328155755996704, |
| "eval_runtime": 6.6732, |
| "eval_samples_per_second": 5.545, |
| "eval_steps_per_second": 2.847, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.355, |
| "grad_norm": 17.875, |
| "learning_rate": 5e-05, |
| "loss": 3.0564363479614256, |
| "step": 355 |
| }, |
| { |
| "epoch": 0.36, |
| "grad_norm": 2.6875, |
| "learning_rate": 5e-05, |
| "loss": 3.068992805480957, |
| "step": 360 |
| }, |
| { |
| "epoch": 0.365, |
| "grad_norm": 1.5078125, |
| "learning_rate": 5e-05, |
| "loss": 3.0664575576782225, |
| "step": 365 |
| }, |
| { |
| "epoch": 0.37, |
| "grad_norm": 204.0, |
| "learning_rate": 5e-05, |
| "loss": 3.015609931945801, |
| "step": 370 |
| }, |
| { |
| "epoch": 0.375, |
| "grad_norm": 296.0, |
| "learning_rate": 5e-05, |
| "loss": 3.0494104385375977, |
| "step": 375 |
| }, |
| { |
| "epoch": 0.38, |
| "grad_norm": 1024.0, |
| "learning_rate": 5e-05, |
| "loss": 3.0338293075561524, |
| "step": 380 |
| }, |
| { |
| "epoch": 0.385, |
| "grad_norm": 7.4375, |
| "learning_rate": 5e-05, |
| "loss": 3.037934684753418, |
| "step": 385 |
| }, |
| { |
| "epoch": 0.39, |
| "grad_norm": 3.28125, |
| "learning_rate": 5e-05, |
| "loss": 3.0603221893310546, |
| "step": 390 |
| }, |
| { |
| "epoch": 0.395, |
| "grad_norm": 17.125, |
| "learning_rate": 5e-05, |
| "loss": 3.0717981338500975, |
| "step": 395 |
| }, |
| { |
| "epoch": 0.4, |
| "grad_norm": 44.75, |
| "learning_rate": 5e-05, |
| "loss": 3.0237457275390627, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.4, |
| "eval_loss": 3.322870969772339, |
| "eval_runtime": 6.749, |
| "eval_samples_per_second": 5.482, |
| "eval_steps_per_second": 2.815, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.405, |
| "grad_norm": 24.5, |
| "learning_rate": 5e-05, |
| "loss": 3.0751102447509764, |
| "step": 405 |
| }, |
| { |
| "epoch": 0.41, |
| "grad_norm": 2.65625, |
| "learning_rate": 5e-05, |
| "loss": 3.0429092407226563, |
| "step": 410 |
| }, |
| { |
| "epoch": 0.415, |
| "grad_norm": 16.0, |
| "learning_rate": 5e-05, |
| "loss": 3.0210216522216795, |
| "step": 415 |
| }, |
| { |
| "epoch": 0.42, |
| "grad_norm": 7.53125, |
| "learning_rate": 5e-05, |
| "loss": 3.065318489074707, |
| "step": 420 |
| }, |
| { |
| "epoch": 0.425, |
| "grad_norm": 24.375, |
| "learning_rate": 5e-05, |
| "loss": 3.0230634689331053, |
| "step": 425 |
| }, |
| { |
| "epoch": 0.43, |
| "grad_norm": 5.625, |
| "learning_rate": 5e-05, |
| "loss": 3.02874698638916, |
| "step": 430 |
| }, |
| { |
| "epoch": 0.435, |
| "grad_norm": 16.5, |
| "learning_rate": 5e-05, |
| "loss": 3.052432632446289, |
| "step": 435 |
| }, |
| { |
| "epoch": 0.44, |
| "grad_norm": 15.9375, |
| "learning_rate": 5e-05, |
| "loss": 3.0222875595092775, |
| "step": 440 |
| }, |
| { |
| "epoch": 0.445, |
| "grad_norm": 19.0, |
| "learning_rate": 5e-05, |
| "loss": 3.053955841064453, |
| "step": 445 |
| }, |
| { |
| "epoch": 0.45, |
| "grad_norm": 86.5, |
| "learning_rate": 5e-05, |
| "loss": 3.0460411071777345, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.45, |
| "eval_loss": 3.317772626876831, |
| "eval_runtime": 6.7179, |
| "eval_samples_per_second": 5.508, |
| "eval_steps_per_second": 2.828, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.455, |
| "grad_norm": 1.84375, |
| "learning_rate": 5e-05, |
| "loss": 3.115060806274414, |
| "step": 455 |
| }, |
| { |
| "epoch": 0.46, |
| "grad_norm": 19.375, |
| "learning_rate": 5e-05, |
| "loss": 3.071373176574707, |
| "step": 460 |
| }, |
| { |
| "epoch": 0.465, |
| "grad_norm": 6.6875, |
| "learning_rate": 5e-05, |
| "loss": 3.0433387756347656, |
| "step": 465 |
| }, |
| { |
| "epoch": 0.47, |
| "grad_norm": 2.09375, |
| "learning_rate": 5e-05, |
| "loss": 3.049169921875, |
| "step": 470 |
| }, |
| { |
| "epoch": 0.475, |
| "grad_norm": 3.9375, |
| "learning_rate": 5e-05, |
| "loss": 3.079736328125, |
| "step": 475 |
| }, |
| { |
| "epoch": 0.48, |
| "grad_norm": 3.796875, |
| "learning_rate": 5e-05, |
| "loss": 3.0374773025512694, |
| "step": 480 |
| }, |
| { |
| "epoch": 0.485, |
| "grad_norm": 37.25, |
| "learning_rate": 5e-05, |
| "loss": 3.022678756713867, |
| "step": 485 |
| }, |
| { |
| "epoch": 0.49, |
| "grad_norm": 65.0, |
| "learning_rate": 5e-05, |
| "loss": 3.032781410217285, |
| "step": 490 |
| }, |
| { |
| "epoch": 0.495, |
| "grad_norm": 10.375, |
| "learning_rate": 5e-05, |
| "loss": 3.0296302795410157, |
| "step": 495 |
| }, |
| { |
| "epoch": 0.5, |
| "grad_norm": 7.9375, |
| "learning_rate": 5e-05, |
| "loss": 3.0553735733032226, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.5, |
| "eval_loss": 3.3104944229125977, |
| "eval_runtime": 6.7677, |
| "eval_samples_per_second": 5.467, |
| "eval_steps_per_second": 2.807, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.505, |
| "grad_norm": 2.71875, |
| "learning_rate": 5e-05, |
| "loss": 3.063258743286133, |
| "step": 505 |
| }, |
| { |
| "epoch": 0.51, |
| "grad_norm": 10.4375, |
| "learning_rate": 5e-05, |
| "loss": 3.0316259384155275, |
| "step": 510 |
| }, |
| { |
| "epoch": 0.515, |
| "grad_norm": 9.5, |
| "learning_rate": 5e-05, |
| "loss": 3.0448724746704103, |
| "step": 515 |
| }, |
| { |
| "epoch": 0.52, |
| "grad_norm": 2.109375, |
| "learning_rate": 5e-05, |
| "loss": 3.0403762817382813, |
| "step": 520 |
| }, |
| { |
| "epoch": 0.525, |
| "grad_norm": 3.609375, |
| "learning_rate": 5e-05, |
| "loss": 2.9868940353393554, |
| "step": 525 |
| }, |
| { |
| "epoch": 0.53, |
| "grad_norm": 3.875, |
| "learning_rate": 5e-05, |
| "loss": 3.010245704650879, |
| "step": 530 |
| }, |
| { |
| "epoch": 0.535, |
| "grad_norm": 1.5859375, |
| "learning_rate": 5e-05, |
| "loss": 3.039686393737793, |
| "step": 535 |
| }, |
| { |
| "epoch": 0.54, |
| "grad_norm": 25.5, |
| "learning_rate": 5e-05, |
| "loss": 3.041441535949707, |
| "step": 540 |
| }, |
| { |
| "epoch": 0.545, |
| "grad_norm": 8.0, |
| "learning_rate": 5e-05, |
| "loss": 3.0453310012817383, |
| "step": 545 |
| }, |
| { |
| "epoch": 0.55, |
| "grad_norm": 9.6875, |
| "learning_rate": 5e-05, |
| "loss": 3.0214563369750977, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.55, |
| "eval_loss": 3.3045036792755127, |
| "eval_runtime": 6.7272, |
| "eval_samples_per_second": 5.5, |
| "eval_steps_per_second": 2.824, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.555, |
| "grad_norm": 64.0, |
| "learning_rate": 5e-05, |
| "loss": 3.0599187850952148, |
| "step": 555 |
| }, |
| { |
| "epoch": 0.56, |
| "grad_norm": 5.5, |
| "learning_rate": 5e-05, |
| "loss": 3.0171770095825194, |
| "step": 560 |
| }, |
| { |
| "epoch": 0.565, |
| "grad_norm": 2.0625, |
| "learning_rate": 5e-05, |
| "loss": 3.0361049652099608, |
| "step": 565 |
| }, |
| { |
| "epoch": 0.57, |
| "grad_norm": 1.296875, |
| "learning_rate": 5e-05, |
| "loss": 3.0191267013549803, |
| "step": 570 |
| }, |
| { |
| "epoch": 0.575, |
| "grad_norm": 2.28125, |
| "learning_rate": 5e-05, |
| "loss": 3.0235448837280274, |
| "step": 575 |
| }, |
| { |
| "epoch": 0.58, |
| "grad_norm": 4.0625, |
| "learning_rate": 5e-05, |
| "loss": 3.0432418823242187, |
| "step": 580 |
| }, |
| { |
| "epoch": 0.585, |
| "grad_norm": 28.25, |
| "learning_rate": 5e-05, |
| "loss": 3.0085990905761717, |
| "step": 585 |
| }, |
| { |
| "epoch": 0.59, |
| "grad_norm": 1.546875, |
| "learning_rate": 5e-05, |
| "loss": 2.9956476211547853, |
| "step": 590 |
| }, |
| { |
| "epoch": 0.595, |
| "grad_norm": 28.125, |
| "learning_rate": 5e-05, |
| "loss": 3.0316390991210938, |
| "step": 595 |
| }, |
| { |
| "epoch": 0.6, |
| "grad_norm": 13.6875, |
| "learning_rate": 5e-05, |
| "loss": 3.040385627746582, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.6, |
| "eval_loss": 3.2979705333709717, |
| "eval_runtime": 6.7492, |
| "eval_samples_per_second": 5.482, |
| "eval_steps_per_second": 2.815, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.605, |
| "grad_norm": 26.125, |
| "learning_rate": 5e-05, |
| "loss": 3.0119483947753904, |
| "step": 605 |
| }, |
| { |
| "epoch": 0.61, |
| "grad_norm": 1.3515625, |
| "learning_rate": 5e-05, |
| "loss": 3.0221290588378906, |
| "step": 610 |
| }, |
| { |
| "epoch": 0.615, |
| "grad_norm": 1.328125, |
| "learning_rate": 5e-05, |
| "loss": 3.034591484069824, |
| "step": 615 |
| }, |
| { |
| "epoch": 0.62, |
| "grad_norm": 1.8359375, |
| "learning_rate": 5e-05, |
| "loss": 3.0486448287963865, |
| "step": 620 |
| }, |
| { |
| "epoch": 0.625, |
| "grad_norm": 6.125, |
| "learning_rate": 5e-05, |
| "loss": 3.0161321640014647, |
| "step": 625 |
| }, |
| { |
| "epoch": 0.63, |
| "grad_norm": 47.75, |
| "learning_rate": 5e-05, |
| "loss": 3.034286880493164, |
| "step": 630 |
| }, |
| { |
| "epoch": 0.635, |
| "grad_norm": 22.5, |
| "learning_rate": 5e-05, |
| "loss": 3.05667667388916, |
| "step": 635 |
| }, |
| { |
| "epoch": 0.64, |
| "grad_norm": 4.28125, |
| "learning_rate": 5e-05, |
| "loss": 3.0220699310302734, |
| "step": 640 |
| }, |
| { |
| "epoch": 0.645, |
| "grad_norm": 2.640625, |
| "learning_rate": 5e-05, |
| "loss": 2.995787811279297, |
| "step": 645 |
| }, |
| { |
| "epoch": 0.65, |
| "grad_norm": 3.140625, |
| "learning_rate": 5e-05, |
| "loss": 3.030243492126465, |
| "step": 650 |
| }, |
| { |
| "epoch": 0.65, |
| "eval_loss": 3.2928740978240967, |
| "eval_runtime": 6.768, |
| "eval_samples_per_second": 5.467, |
| "eval_steps_per_second": 2.807, |
| "step": 650 |
| }, |
| { |
| "epoch": 0.655, |
| "grad_norm": 1.7734375, |
| "learning_rate": 5e-05, |
| "loss": 2.9878942489624025, |
| "step": 655 |
| }, |
| { |
| "epoch": 0.66, |
| "grad_norm": 21.75, |
| "learning_rate": 5e-05, |
| "loss": 3.0131813049316407, |
| "step": 660 |
| }, |
| { |
| "epoch": 0.665, |
| "grad_norm": 2.0, |
| "learning_rate": 5e-05, |
| "loss": 2.988085174560547, |
| "step": 665 |
| }, |
| { |
| "epoch": 0.67, |
| "grad_norm": 1.4296875, |
| "learning_rate": 5e-05, |
| "loss": 3.022294044494629, |
| "step": 670 |
| }, |
| { |
| "epoch": 0.675, |
| "grad_norm": 6.875, |
| "learning_rate": 5e-05, |
| "loss": 3.0046091079711914, |
| "step": 675 |
| }, |
| { |
| "epoch": 0.68, |
| "grad_norm": 2.875, |
| "learning_rate": 5e-05, |
| "loss": 3.013047981262207, |
| "step": 680 |
| }, |
| { |
| "epoch": 0.685, |
| "grad_norm": 17.0, |
| "learning_rate": 5e-05, |
| "loss": 2.9954095840454102, |
| "step": 685 |
| }, |
| { |
| "epoch": 0.69, |
| "grad_norm": 7.03125, |
| "learning_rate": 5e-05, |
| "loss": 3.004934310913086, |
| "step": 690 |
| }, |
| { |
| "epoch": 0.695, |
| "grad_norm": 1.84375, |
| "learning_rate": 5e-05, |
| "loss": 3.037328910827637, |
| "step": 695 |
| }, |
| { |
| "epoch": 0.7, |
| "grad_norm": 11.25, |
| "learning_rate": 5e-05, |
| "loss": 3.022724914550781, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.7, |
| "eval_loss": 3.2860541343688965, |
| "eval_runtime": 6.7324, |
| "eval_samples_per_second": 5.496, |
| "eval_steps_per_second": 2.822, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.705, |
| "grad_norm": 1.7890625, |
| "learning_rate": 5e-05, |
| "loss": 2.9891803741455076, |
| "step": 705 |
| }, |
| { |
| "epoch": 0.71, |
| "grad_norm": 8.5, |
| "learning_rate": 5e-05, |
| "loss": 3.0228837966918944, |
| "step": 710 |
| }, |
| { |
| "epoch": 0.715, |
| "grad_norm": 2.171875, |
| "learning_rate": 5e-05, |
| "loss": 3.0380273818969727, |
| "step": 715 |
| }, |
| { |
| "epoch": 0.72, |
| "grad_norm": 1.6484375, |
| "learning_rate": 5e-05, |
| "loss": 3.013555335998535, |
| "step": 720 |
| }, |
| { |
| "epoch": 0.725, |
| "grad_norm": 1.3984375, |
| "learning_rate": 5e-05, |
| "loss": 2.9776601791381836, |
| "step": 725 |
| }, |
| { |
| "epoch": 0.73, |
| "grad_norm": 2.46875, |
| "learning_rate": 5e-05, |
| "loss": 3.0011289596557615, |
| "step": 730 |
| }, |
| { |
| "epoch": 0.735, |
| "grad_norm": 2.0625, |
| "learning_rate": 5e-05, |
| "loss": 2.99926872253418, |
| "step": 735 |
| }, |
| { |
| "epoch": 0.74, |
| "grad_norm": 3.453125, |
| "learning_rate": 5e-05, |
| "loss": 2.9941511154174805, |
| "step": 740 |
| }, |
| { |
| "epoch": 0.745, |
| "grad_norm": 6.625, |
| "learning_rate": 5e-05, |
| "loss": 3.0118370056152344, |
| "step": 745 |
| }, |
| { |
| "epoch": 0.75, |
| "grad_norm": 7.28125, |
| "learning_rate": 5e-05, |
| "loss": 2.9928028106689455, |
| "step": 750 |
| }, |
| { |
| "epoch": 0.75, |
| "eval_loss": 3.280454158782959, |
| "eval_runtime": 6.7672, |
| "eval_samples_per_second": 5.468, |
| "eval_steps_per_second": 2.808, |
| "step": 750 |
| }, |
| { |
| "epoch": 0.755, |
| "grad_norm": 4.34375, |
| "learning_rate": 5e-05, |
| "loss": 3.012207794189453, |
| "step": 755 |
| }, |
| { |
| "epoch": 0.76, |
| "grad_norm": 1.6796875, |
| "learning_rate": 5e-05, |
| "loss": 2.9685001373291016, |
| "step": 760 |
| }, |
| { |
| "epoch": 0.765, |
| "grad_norm": 0.95703125, |
| "learning_rate": 5e-05, |
| "loss": 3.032526397705078, |
| "step": 765 |
| }, |
| { |
| "epoch": 0.77, |
| "grad_norm": 2.0625, |
| "learning_rate": 5e-05, |
| "loss": 3.018193244934082, |
| "step": 770 |
| }, |
| { |
| "epoch": 0.775, |
| "grad_norm": 160.0, |
| "learning_rate": 5e-05, |
| "loss": 2.9998651504516602, |
| "step": 775 |
| }, |
| { |
| "epoch": 0.78, |
| "grad_norm": 182.0, |
| "learning_rate": 5e-05, |
| "loss": 3.0007757186889648, |
| "step": 780 |
| }, |
| { |
| "epoch": 0.785, |
| "grad_norm": 3.03125, |
| "learning_rate": 5e-05, |
| "loss": 2.9936758041381837, |
| "step": 785 |
| }, |
| { |
| "epoch": 0.79, |
| "grad_norm": 3.4375, |
| "learning_rate": 5e-05, |
| "loss": 2.9737762451171874, |
| "step": 790 |
| }, |
| { |
| "epoch": 0.795, |
| "grad_norm": 63.5, |
| "learning_rate": 5e-05, |
| "loss": 3.0091293334960936, |
| "step": 795 |
| }, |
| { |
| "epoch": 0.8, |
| "grad_norm": 46.0, |
| "learning_rate": 5e-05, |
| "loss": 3.021886444091797, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_loss": 3.2757456302642822, |
| "eval_runtime": 6.7253, |
| "eval_samples_per_second": 5.502, |
| "eval_steps_per_second": 2.825, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.805, |
| "grad_norm": 8.3125, |
| "learning_rate": 5e-05, |
| "loss": 2.9799240112304686, |
| "step": 805 |
| }, |
| { |
| "epoch": 0.81, |
| "grad_norm": 7.9375, |
| "learning_rate": 5e-05, |
| "loss": 3.0026453018188475, |
| "step": 810 |
| }, |
| { |
| "epoch": 0.815, |
| "grad_norm": 1.65625, |
| "learning_rate": 5e-05, |
| "loss": 2.999722480773926, |
| "step": 815 |
| }, |
| { |
| "epoch": 0.82, |
| "grad_norm": 1.265625, |
| "learning_rate": 5e-05, |
| "loss": 2.966134452819824, |
| "step": 820 |
| }, |
| { |
| "epoch": 0.825, |
| "grad_norm": 340.0, |
| "learning_rate": 5e-05, |
| "loss": 2.979732322692871, |
| "step": 825 |
| }, |
| { |
| "epoch": 0.83, |
| "grad_norm": 34.5, |
| "learning_rate": 5e-05, |
| "loss": 3.0184419631958006, |
| "step": 830 |
| }, |
| { |
| "epoch": 0.835, |
| "grad_norm": 4.71875, |
| "learning_rate": 5e-05, |
| "loss": 2.978903961181641, |
| "step": 835 |
| }, |
| { |
| "epoch": 0.84, |
| "grad_norm": 7.75, |
| "learning_rate": 5e-05, |
| "loss": 3.000539016723633, |
| "step": 840 |
| }, |
| { |
| "epoch": 0.845, |
| "grad_norm": 3.0, |
| "learning_rate": 5e-05, |
| "loss": 2.962353515625, |
| "step": 845 |
| }, |
| { |
| "epoch": 0.85, |
| "grad_norm": 2.171875, |
| "learning_rate": 5e-05, |
| "loss": 2.9898805618286133, |
| "step": 850 |
| }, |
| { |
| "epoch": 0.85, |
| "eval_loss": 3.2711822986602783, |
| "eval_runtime": 6.7266, |
| "eval_samples_per_second": 5.501, |
| "eval_steps_per_second": 2.825, |
| "step": 850 |
| }, |
| { |
| "epoch": 0.855, |
| "grad_norm": 4.34375, |
| "learning_rate": 5e-05, |
| "loss": 2.973104476928711, |
| "step": 855 |
| }, |
| { |
| "epoch": 0.86, |
| "grad_norm": 6.21875, |
| "learning_rate": 5e-05, |
| "loss": 3.00762996673584, |
| "step": 860 |
| }, |
| { |
| "epoch": 0.865, |
| "grad_norm": 3.96875, |
| "learning_rate": 5e-05, |
| "loss": 2.9680364608764647, |
| "step": 865 |
| }, |
| { |
| "epoch": 0.87, |
| "grad_norm": 3.546875, |
| "learning_rate": 5e-05, |
| "loss": 2.9908233642578126, |
| "step": 870 |
| }, |
| { |
| "epoch": 0.875, |
| "grad_norm": 5.25, |
| "learning_rate": 5e-05, |
| "loss": 2.9942108154296876, |
| "step": 875 |
| }, |
| { |
| "epoch": 0.88, |
| "grad_norm": 7.25, |
| "learning_rate": 5e-05, |
| "loss": 2.966433525085449, |
| "step": 880 |
| }, |
| { |
| "epoch": 0.885, |
| "grad_norm": 6.84375, |
| "learning_rate": 5e-05, |
| "loss": 2.9872104644775392, |
| "step": 885 |
| }, |
| { |
| "epoch": 0.89, |
| "grad_norm": 2.3125, |
| "learning_rate": 5e-05, |
| "loss": 2.9889734268188475, |
| "step": 890 |
| }, |
| { |
| "epoch": 0.895, |
| "grad_norm": 4.46875, |
| "learning_rate": 5e-05, |
| "loss": 2.9673152923583985, |
| "step": 895 |
| }, |
| { |
| "epoch": 0.9, |
| "grad_norm": 5.15625, |
| "learning_rate": 5e-05, |
| "loss": 2.992472267150879, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.9, |
| "eval_loss": 3.269010066986084, |
| "eval_runtime": 6.7827, |
| "eval_samples_per_second": 5.455, |
| "eval_steps_per_second": 2.801, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.905, |
| "grad_norm": 1.7890625, |
| "learning_rate": 5e-05, |
| "loss": 2.982642936706543, |
| "step": 905 |
| }, |
| { |
| "epoch": 0.91, |
| "grad_norm": 7.4375, |
| "learning_rate": 5e-05, |
| "loss": 3.015847396850586, |
| "step": 910 |
| }, |
| { |
| "epoch": 0.915, |
| "grad_norm": 5376.0, |
| "learning_rate": 5e-05, |
| "loss": 3.003455924987793, |
| "step": 915 |
| }, |
| { |
| "epoch": 0.92, |
| "grad_norm": 0.83984375, |
| "learning_rate": 5e-05, |
| "loss": 2.987352180480957, |
| "step": 920 |
| }, |
| { |
| "epoch": 0.925, |
| "grad_norm": 1.9921875, |
| "learning_rate": 5e-05, |
| "loss": 2.9890872955322267, |
| "step": 925 |
| }, |
| { |
| "epoch": 0.93, |
| "grad_norm": 352.0, |
| "learning_rate": 5e-05, |
| "loss": 2.990787315368652, |
| "step": 930 |
| }, |
| { |
| "epoch": 0.935, |
| "grad_norm": 9.5, |
| "learning_rate": 5e-05, |
| "loss": 2.950652313232422, |
| "step": 935 |
| }, |
| { |
| "epoch": 0.94, |
| "grad_norm": 84.5, |
| "learning_rate": 5e-05, |
| "loss": 2.978129768371582, |
| "step": 940 |
| }, |
| { |
| "epoch": 0.945, |
| "grad_norm": 900.0, |
| "learning_rate": 5e-05, |
| "loss": 2.9651300430297853, |
| "step": 945 |
| }, |
| { |
| "epoch": 0.95, |
| "grad_norm": 17.75, |
| "learning_rate": 5e-05, |
| "loss": 2.940041351318359, |
| "step": 950 |
| }, |
| { |
| "epoch": 0.95, |
| "eval_loss": 3.2654120922088623, |
| "eval_runtime": 6.7595, |
| "eval_samples_per_second": 5.474, |
| "eval_steps_per_second": 2.811, |
| "step": 950 |
| }, |
| { |
| "epoch": 0.955, |
| "grad_norm": 5.25, |
| "learning_rate": 5e-05, |
| "loss": 2.9906681060791014, |
| "step": 955 |
| }, |
| { |
| "epoch": 0.96, |
| "grad_norm": 1.65625, |
| "learning_rate": 5e-05, |
| "loss": 2.965926933288574, |
| "step": 960 |
| }, |
| { |
| "epoch": 0.965, |
| "grad_norm": 1.9609375, |
| "learning_rate": 5e-05, |
| "loss": 3.0010139465332033, |
| "step": 965 |
| }, |
| { |
| "epoch": 0.97, |
| "grad_norm": 24.625, |
| "learning_rate": 5e-05, |
| "loss": 2.9550275802612305, |
| "step": 970 |
| }, |
| { |
| "epoch": 0.975, |
| "grad_norm": 2.515625, |
| "learning_rate": 5e-05, |
| "loss": 2.970465660095215, |
| "step": 975 |
| }, |
| { |
| "epoch": 0.98, |
| "grad_norm": 3.4375, |
| "learning_rate": 5e-05, |
| "loss": 2.9523544311523438, |
| "step": 980 |
| }, |
| { |
| "epoch": 0.985, |
| "grad_norm": 14.625, |
| "learning_rate": 5e-05, |
| "loss": 3.002934455871582, |
| "step": 985 |
| }, |
| { |
| "epoch": 0.99, |
| "grad_norm": 2.921875, |
| "learning_rate": 5e-05, |
| "loss": 2.964326286315918, |
| "step": 990 |
| }, |
| { |
| "epoch": 0.995, |
| "grad_norm": 3.625, |
| "learning_rate": 5e-05, |
| "loss": 2.9944055557250975, |
| "step": 995 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 11.9375, |
| "learning_rate": 5e-05, |
| "loss": 2.9779850006103517, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 3.2615818977355957, |
| "eval_runtime": 6.7329, |
| "eval_samples_per_second": 5.495, |
| "eval_steps_per_second": 2.822, |
| "step": 1000 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 1000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 9223372036854775807, |
| "save_steps": 200, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.89915124318208e+17, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
