| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.384615384615385, | |
| "eval_steps": 10, | |
| "global_step": 1750, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.006153846153846154, | |
| "eval_loss": 1.489196538925171, | |
| "eval_runtime": 1.2747, | |
| "eval_samples_per_second": 89.432, | |
| "eval_steps_per_second": 4.707, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.03076923076923077, | |
| "grad_norm": 17.32219123840332, | |
| "learning_rate": 8.771929824561404e-07, | |
| "loss": 1.5722, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03076923076923077, | |
| "eval_loss": 1.3865292072296143, | |
| "eval_runtime": 1.1749, | |
| "eval_samples_per_second": 97.029, | |
| "eval_steps_per_second": 5.107, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06153846153846154, | |
| "grad_norm": 21.88077735900879, | |
| "learning_rate": 1.7543859649122807e-06, | |
| "loss": 1.3935, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06153846153846154, | |
| "eval_loss": 1.0996264219284058, | |
| "eval_runtime": 1.1706, | |
| "eval_samples_per_second": 97.386, | |
| "eval_steps_per_second": 5.126, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09230769230769231, | |
| "grad_norm": 17.395177841186523, | |
| "learning_rate": 2.631578947368421e-06, | |
| "loss": 1.0664, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.09230769230769231, | |
| "eval_loss": 0.8006665706634521, | |
| "eval_runtime": 1.1747, | |
| "eval_samples_per_second": 97.044, | |
| "eval_steps_per_second": 5.108, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12307692307692308, | |
| "grad_norm": 6.869284152984619, | |
| "learning_rate": 3.5087719298245615e-06, | |
| "loss": 0.7994, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.12307692307692308, | |
| "eval_loss": 0.708875298500061, | |
| "eval_runtime": 1.1742, | |
| "eval_samples_per_second": 97.087, | |
| "eval_steps_per_second": 5.11, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 6.4737162590026855, | |
| "learning_rate": 4.385964912280702e-06, | |
| "loss": 0.7751, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "eval_loss": 0.6778659224510193, | |
| "eval_runtime": 1.1722, | |
| "eval_samples_per_second": 97.257, | |
| "eval_steps_per_second": 5.119, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18461538461538463, | |
| "grad_norm": 5.418182373046875, | |
| "learning_rate": 5.263157894736842e-06, | |
| "loss": 0.6203, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.18461538461538463, | |
| "eval_loss": 0.6695793271064758, | |
| "eval_runtime": 1.1744, | |
| "eval_samples_per_second": 97.071, | |
| "eval_steps_per_second": 5.109, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2153846153846154, | |
| "grad_norm": 2.7423934936523438, | |
| "learning_rate": 6.140350877192983e-06, | |
| "loss": 0.767, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2153846153846154, | |
| "eval_loss": 0.6650205850601196, | |
| "eval_runtime": 1.1671, | |
| "eval_samples_per_second": 97.675, | |
| "eval_steps_per_second": 5.141, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.24615384615384617, | |
| "grad_norm": 2.572129726409912, | |
| "learning_rate": 7.017543859649123e-06, | |
| "loss": 0.6336, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.24615384615384617, | |
| "eval_loss": 0.6613836884498596, | |
| "eval_runtime": 1.1671, | |
| "eval_samples_per_second": 97.676, | |
| "eval_steps_per_second": 5.141, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.27692307692307694, | |
| "grad_norm": 4.622325420379639, | |
| "learning_rate": 7.894736842105265e-06, | |
| "loss": 0.631, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.27692307692307694, | |
| "eval_loss": 0.6617783904075623, | |
| "eval_runtime": 1.1744, | |
| "eval_samples_per_second": 97.071, | |
| "eval_steps_per_second": 5.109, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 3.24354887008667, | |
| "learning_rate": 8.771929824561405e-06, | |
| "loss": 0.6086, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "eval_loss": 0.662232518196106, | |
| "eval_runtime": 1.172, | |
| "eval_samples_per_second": 97.27, | |
| "eval_steps_per_second": 5.119, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3384615384615385, | |
| "grad_norm": 3.5868513584136963, | |
| "learning_rate": 9.649122807017545e-06, | |
| "loss": 0.7057, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3384615384615385, | |
| "eval_loss": 0.6619779467582703, | |
| "eval_runtime": 1.165, | |
| "eval_samples_per_second": 97.856, | |
| "eval_steps_per_second": 5.15, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.36923076923076925, | |
| "grad_norm": 3.4894511699676514, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 0.7385, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.36923076923076925, | |
| "eval_loss": 0.6673641800880432, | |
| "eval_runtime": 1.1682, | |
| "eval_samples_per_second": 97.588, | |
| "eval_steps_per_second": 5.136, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 5.117406845092773, | |
| "learning_rate": 1.1403508771929826e-05, | |
| "loss": 0.6533, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.6705042123794556, | |
| "eval_runtime": 1.1726, | |
| "eval_samples_per_second": 97.217, | |
| "eval_steps_per_second": 5.117, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.4307692307692308, | |
| "grad_norm": 3.9341437816619873, | |
| "learning_rate": 1.2280701754385966e-05, | |
| "loss": 0.7066, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.4307692307692308, | |
| "eval_loss": 0.6691470146179199, | |
| "eval_runtime": 1.1702, | |
| "eval_samples_per_second": 97.418, | |
| "eval_steps_per_second": 5.127, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "grad_norm": 3.4131579399108887, | |
| "learning_rate": 1.3157894736842108e-05, | |
| "loss": 0.6065, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "eval_loss": 0.6724759340286255, | |
| "eval_runtime": 1.1746, | |
| "eval_samples_per_second": 97.05, | |
| "eval_steps_per_second": 5.108, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.49230769230769234, | |
| "grad_norm": 3.820042848587036, | |
| "learning_rate": 1.4035087719298246e-05, | |
| "loss": 0.6474, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.49230769230769234, | |
| "eval_loss": 0.6704264879226685, | |
| "eval_runtime": 1.1772, | |
| "eval_samples_per_second": 96.838, | |
| "eval_steps_per_second": 5.097, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.5230769230769231, | |
| "grad_norm": 2.8097360134124756, | |
| "learning_rate": 1.4912280701754388e-05, | |
| "loss": 0.7022, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5230769230769231, | |
| "eval_loss": 0.6691609025001526, | |
| "eval_runtime": 1.1676, | |
| "eval_samples_per_second": 97.638, | |
| "eval_steps_per_second": 5.139, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5538461538461539, | |
| "grad_norm": 2.3641347885131836, | |
| "learning_rate": 1.578947368421053e-05, | |
| "loss": 0.6667, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.5538461538461539, | |
| "eval_loss": 0.6753159165382385, | |
| "eval_runtime": 1.1689, | |
| "eval_samples_per_second": 97.524, | |
| "eval_steps_per_second": 5.133, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.5846153846153846, | |
| "grad_norm": 6.359607219696045, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.6889, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.5846153846153846, | |
| "eval_loss": 0.6796454787254333, | |
| "eval_runtime": 1.1714, | |
| "eval_samples_per_second": 97.323, | |
| "eval_steps_per_second": 5.122, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 4.7979960441589355, | |
| "learning_rate": 1.754385964912281e-05, | |
| "loss": 0.6937, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "eval_loss": 0.6748442649841309, | |
| "eval_runtime": 1.1773, | |
| "eval_samples_per_second": 96.831, | |
| "eval_steps_per_second": 5.096, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6461538461538462, | |
| "grad_norm": 3.5100464820861816, | |
| "learning_rate": 1.8421052631578947e-05, | |
| "loss": 0.6613, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6461538461538462, | |
| "eval_loss": 0.6842787861824036, | |
| "eval_runtime": 1.1662, | |
| "eval_samples_per_second": 97.752, | |
| "eval_steps_per_second": 5.145, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.676923076923077, | |
| "grad_norm": 3.3538153171539307, | |
| "learning_rate": 1.929824561403509e-05, | |
| "loss": 0.6512, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.676923076923077, | |
| "eval_loss": 0.6873583197593689, | |
| "eval_runtime": 1.1667, | |
| "eval_samples_per_second": 97.712, | |
| "eval_steps_per_second": 5.143, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.7076923076923077, | |
| "grad_norm": 3.7595653533935547, | |
| "learning_rate": 1.9999952892103225e-05, | |
| "loss": 0.7656, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7076923076923077, | |
| "eval_loss": 0.6970181465148926, | |
| "eval_runtime": 1.1663, | |
| "eval_samples_per_second": 97.741, | |
| "eval_steps_per_second": 5.144, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7384615384615385, | |
| "grad_norm": 2.9787654876708984, | |
| "learning_rate": 1.999830416231782e-05, | |
| "loss": 0.6412, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.7384615384615385, | |
| "eval_loss": 0.7058220505714417, | |
| "eval_runtime": 1.1688, | |
| "eval_samples_per_second": 97.537, | |
| "eval_steps_per_second": 5.134, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 2.3455305099487305, | |
| "learning_rate": 1.9994300481505595e-05, | |
| "loss": 0.6148, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "eval_loss": 0.7070643305778503, | |
| "eval_runtime": 1.1728, | |
| "eval_samples_per_second": 97.202, | |
| "eval_steps_per_second": 5.116, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 4.057767868041992, | |
| "learning_rate": 1.998794279267369e-05, | |
| "loss": 0.6433, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.7127295136451721, | |
| "eval_runtime": 1.1689, | |
| "eval_samples_per_second": 97.529, | |
| "eval_steps_per_second": 5.133, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.8307692307692308, | |
| "grad_norm": 2.5686323642730713, | |
| "learning_rate": 1.9979232593280637e-05, | |
| "loss": 0.5972, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.8307692307692308, | |
| "eval_loss": 0.7106388211250305, | |
| "eval_runtime": 1.166, | |
| "eval_samples_per_second": 97.77, | |
| "eval_steps_per_second": 5.146, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.8615384615384616, | |
| "grad_norm": 1.2748245000839233, | |
| "learning_rate": 1.9968171934883647e-05, | |
| "loss": 0.6795, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.8615384615384616, | |
| "eval_loss": 0.7127251029014587, | |
| "eval_runtime": 1.1701, | |
| "eval_samples_per_second": 97.429, | |
| "eval_steps_per_second": 5.128, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.8923076923076924, | |
| "grad_norm": 3.1174163818359375, | |
| "learning_rate": 1.9954763422655396e-05, | |
| "loss": 0.6589, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.8923076923076924, | |
| "eval_loss": 0.7157883048057556, | |
| "eval_runtime": 1.1727, | |
| "eval_samples_per_second": 97.211, | |
| "eval_steps_per_second": 5.116, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 2.3197476863861084, | |
| "learning_rate": 1.9939010214770426e-05, | |
| "loss": 0.5539, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "eval_loss": 0.7173091173171997, | |
| "eval_runtime": 1.1712, | |
| "eval_samples_per_second": 97.332, | |
| "eval_steps_per_second": 5.123, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9538461538461539, | |
| "grad_norm": 2.7073607444763184, | |
| "learning_rate": 1.9920916021661277e-05, | |
| "loss": 0.7515, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.9538461538461539, | |
| "eval_loss": 0.720274806022644, | |
| "eval_runtime": 1.1673, | |
| "eval_samples_per_second": 97.665, | |
| "eval_steps_per_second": 5.14, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.9846153846153847, | |
| "grad_norm": 2.78871488571167, | |
| "learning_rate": 1.9900485105144544e-05, | |
| "loss": 0.6282, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.9846153846153847, | |
| "eval_loss": 0.7291054725646973, | |
| "eval_runtime": 1.1713, | |
| "eval_samples_per_second": 97.331, | |
| "eval_steps_per_second": 5.123, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0153846153846153, | |
| "grad_norm": 2.181795597076416, | |
| "learning_rate": 1.9877722277417085e-05, | |
| "loss": 0.4711, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.0153846153846153, | |
| "eval_loss": 0.7346097826957703, | |
| "eval_runtime": 1.1674, | |
| "eval_samples_per_second": 97.652, | |
| "eval_steps_per_second": 5.14, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.0461538461538462, | |
| "grad_norm": 6.42896842956543, | |
| "learning_rate": 1.985263289992256e-05, | |
| "loss": 0.4595, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.0461538461538462, | |
| "eval_loss": 0.7707703709602356, | |
| "eval_runtime": 1.1711, | |
| "eval_samples_per_second": 97.348, | |
| "eval_steps_per_second": 5.124, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.0769230769230769, | |
| "grad_norm": 6.206396102905273, | |
| "learning_rate": 1.9825222882088647e-05, | |
| "loss": 0.3704, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.0769230769230769, | |
| "eval_loss": 0.7343254685401917, | |
| "eval_runtime": 1.1717, | |
| "eval_samples_per_second": 97.298, | |
| "eval_steps_per_second": 5.121, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.1076923076923078, | |
| "grad_norm": 2.2610416412353516, | |
| "learning_rate": 1.9795498679935144e-05, | |
| "loss": 0.3845, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.1076923076923078, | |
| "eval_loss": 0.7747918963432312, | |
| "eval_runtime": 1.1652, | |
| "eval_samples_per_second": 97.838, | |
| "eval_steps_per_second": 5.149, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.1384615384615384, | |
| "grad_norm": 2.490269184112549, | |
| "learning_rate": 1.9763467294553364e-05, | |
| "loss": 0.3461, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.1384615384615384, | |
| "eval_loss": 0.7602437138557434, | |
| "eval_runtime": 1.1685, | |
| "eval_samples_per_second": 97.563, | |
| "eval_steps_per_second": 5.135, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.1692307692307693, | |
| "grad_norm": 2.435138702392578, | |
| "learning_rate": 1.9729136270457118e-05, | |
| "loss": 0.3871, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.1692307692307693, | |
| "eval_loss": 0.7585355043411255, | |
| "eval_runtime": 1.1702, | |
| "eval_samples_per_second": 97.419, | |
| "eval_steps_per_second": 5.127, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 1.649342656135559, | |
| "learning_rate": 1.9692513693805738e-05, | |
| "loss": 0.345, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 0.7817304730415344, | |
| "eval_runtime": 1.169, | |
| "eval_samples_per_second": 97.52, | |
| "eval_steps_per_second": 5.133, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 2.2906811237335205, | |
| "learning_rate": 1.965360819049948e-05, | |
| "loss": 0.4372, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "eval_loss": 0.7519503235816956, | |
| "eval_runtime": 1.1693, | |
| "eval_samples_per_second": 97.495, | |
| "eval_steps_per_second": 5.131, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2615384615384615, | |
| "grad_norm": 3.925938606262207, | |
| "learning_rate": 1.9612428924147842e-05, | |
| "loss": 0.4879, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.2615384615384615, | |
| "eval_loss": 0.7463933229446411, | |
| "eval_runtime": 1.1712, | |
| "eval_samples_per_second": 97.334, | |
| "eval_steps_per_second": 5.123, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.2923076923076924, | |
| "grad_norm": 1.4445178508758545, | |
| "learning_rate": 1.9568985593911206e-05, | |
| "loss": 0.4265, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.2923076923076924, | |
| "eval_loss": 0.7673818469047546, | |
| "eval_runtime": 1.1696, | |
| "eval_samples_per_second": 97.471, | |
| "eval_steps_per_second": 5.13, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.323076923076923, | |
| "grad_norm": 2.707318067550659, | |
| "learning_rate": 1.9523288432216333e-05, | |
| "loss": 0.4255, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.323076923076923, | |
| "eval_loss": 0.776730477809906, | |
| "eval_runtime": 1.1686, | |
| "eval_samples_per_second": 97.552, | |
| "eval_steps_per_second": 5.134, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.353846153846154, | |
| "grad_norm": 2.1543335914611816, | |
| "learning_rate": 1.9475348202346292e-05, | |
| "loss": 0.3992, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.353846153846154, | |
| "eval_loss": 0.7587007284164429, | |
| "eval_runtime": 1.1741, | |
| "eval_samples_per_second": 97.094, | |
| "eval_steps_per_second": 5.11, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.3846153846153846, | |
| "grad_norm": 2.1207425594329834, | |
| "learning_rate": 1.942517619590531e-05, | |
| "loss": 0.4197, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.3846153846153846, | |
| "eval_loss": 0.7629592418670654, | |
| "eval_runtime": 1.1714, | |
| "eval_samples_per_second": 97.323, | |
| "eval_steps_per_second": 5.122, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.4153846153846155, | |
| "grad_norm": 3.0015456676483154, | |
| "learning_rate": 1.9372784230159213e-05, | |
| "loss": 0.3963, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.4153846153846155, | |
| "eval_loss": 0.7871745824813843, | |
| "eval_runtime": 1.1671, | |
| "eval_samples_per_second": 97.676, | |
| "eval_steps_per_second": 5.141, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.4461538461538461, | |
| "grad_norm": 2.6770448684692383, | |
| "learning_rate": 1.9318184645252037e-05, | |
| "loss": 0.3689, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.4461538461538461, | |
| "eval_loss": 0.7728400826454163, | |
| "eval_runtime": 1.1721, | |
| "eval_samples_per_second": 97.262, | |
| "eval_steps_per_second": 5.119, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.476923076923077, | |
| "grad_norm": 2.6416432857513428, | |
| "learning_rate": 1.926139030129951e-05, | |
| "loss": 0.4004, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.476923076923077, | |
| "eval_loss": 0.7785875797271729, | |
| "eval_runtime": 1.1696, | |
| "eval_samples_per_second": 97.466, | |
| "eval_steps_per_second": 5.13, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.5076923076923077, | |
| "grad_norm": 1.8383666276931763, | |
| "learning_rate": 1.9202414575360024e-05, | |
| "loss": 0.4265, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.5076923076923077, | |
| "eval_loss": 0.7735582590103149, | |
| "eval_runtime": 1.168, | |
| "eval_samples_per_second": 97.604, | |
| "eval_steps_per_second": 5.137, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 1.905369758605957, | |
| "learning_rate": 1.9141271358283874e-05, | |
| "loss": 0.3675, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "eval_loss": 0.7821589708328247, | |
| "eval_runtime": 1.168, | |
| "eval_samples_per_second": 97.603, | |
| "eval_steps_per_second": 5.137, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5692307692307692, | |
| "grad_norm": 2.0887465476989746, | |
| "learning_rate": 1.9077975051441487e-05, | |
| "loss": 0.3769, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.5692307692307692, | |
| "eval_loss": 0.8046853542327881, | |
| "eval_runtime": 1.168, | |
| "eval_samples_per_second": 97.604, | |
| "eval_steps_per_second": 5.137, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 1.3952693939208984, | |
| "learning_rate": 1.9012540563331375e-05, | |
| "loss": 0.3842, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.7695035338401794, | |
| "eval_runtime": 1.1677, | |
| "eval_samples_per_second": 97.627, | |
| "eval_steps_per_second": 5.138, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.6307692307692307, | |
| "grad_norm": 2.4253690242767334, | |
| "learning_rate": 1.8944983306068683e-05, | |
| "loss": 0.372, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.6307692307692307, | |
| "eval_loss": 0.7812924385070801, | |
| "eval_runtime": 1.1688, | |
| "eval_samples_per_second": 97.535, | |
| "eval_steps_per_second": 5.133, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.6615384615384614, | |
| "grad_norm": 1.9009945392608643, | |
| "learning_rate": 1.8875319191755083e-05, | |
| "loss": 0.3825, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.6615384615384614, | |
| "eval_loss": 0.7782070636749268, | |
| "eval_runtime": 1.1713, | |
| "eval_samples_per_second": 97.328, | |
| "eval_steps_per_second": 5.123, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.6923076923076923, | |
| "grad_norm": 3.183929443359375, | |
| "learning_rate": 1.8803564628730916e-05, | |
| "loss": 0.396, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.6923076923076923, | |
| "eval_loss": 0.7839831709861755, | |
| "eval_runtime": 1.1706, | |
| "eval_samples_per_second": 97.388, | |
| "eval_steps_per_second": 5.126, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.7230769230769232, | |
| "grad_norm": 2.031071901321411, | |
| "learning_rate": 1.8729736517710454e-05, | |
| "loss": 0.3862, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.7230769230769232, | |
| "eval_loss": 0.7782466411590576, | |
| "eval_runtime": 1.1669, | |
| "eval_samples_per_second": 97.695, | |
| "eval_steps_per_second": 5.142, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.7538461538461538, | |
| "grad_norm": 1.2456035614013672, | |
| "learning_rate": 1.865385224780119e-05, | |
| "loss": 0.3909, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.7538461538461538, | |
| "eval_loss": 0.7737441062927246, | |
| "eval_runtime": 1.1644, | |
| "eval_samples_per_second": 97.908, | |
| "eval_steps_per_second": 5.153, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.7846153846153845, | |
| "grad_norm": 1.3416839838027954, | |
| "learning_rate": 1.8575929692408105e-05, | |
| "loss": 0.3585, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.7846153846153845, | |
| "eval_loss": 0.7757642269134521, | |
| "eval_runtime": 1.1707, | |
| "eval_samples_per_second": 97.376, | |
| "eval_steps_per_second": 5.125, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.8153846153846154, | |
| "grad_norm": 2.5810840129852295, | |
| "learning_rate": 1.8495987205023832e-05, | |
| "loss": 0.4018, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.8153846153846154, | |
| "eval_loss": 0.7671994566917419, | |
| "eval_runtime": 1.1677, | |
| "eval_samples_per_second": 97.627, | |
| "eval_steps_per_second": 5.138, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 2.186009407043457, | |
| "learning_rate": 1.8414043614905782e-05, | |
| "loss": 0.3936, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "eval_loss": 0.7763370871543884, | |
| "eval_runtime": 1.1743, | |
| "eval_samples_per_second": 97.078, | |
| "eval_steps_per_second": 5.109, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.876923076923077, | |
| "grad_norm": 2.5505406856536865, | |
| "learning_rate": 1.8330118222641192e-05, | |
| "loss": 0.4963, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.876923076923077, | |
| "eval_loss": 0.7642151713371277, | |
| "eval_runtime": 1.1648, | |
| "eval_samples_per_second": 97.867, | |
| "eval_steps_per_second": 5.151, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.9076923076923076, | |
| "grad_norm": 2.2109644412994385, | |
| "learning_rate": 1.824423079560116e-05, | |
| "loss": 0.4821, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.9076923076923076, | |
| "eval_loss": 0.7671634554862976, | |
| "eval_runtime": 1.1645, | |
| "eval_samples_per_second": 97.898, | |
| "eval_steps_per_second": 5.153, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.9384615384615385, | |
| "grad_norm": 2.702033519744873, | |
| "learning_rate": 1.8156401563284724e-05, | |
| "loss": 0.4216, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.9384615384615385, | |
| "eval_loss": 0.7776830792427063, | |
| "eval_runtime": 1.1704, | |
| "eval_samples_per_second": 97.399, | |
| "eval_steps_per_second": 5.126, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.9692307692307693, | |
| "grad_norm": 2.2329928874969482, | |
| "learning_rate": 1.8066651212554126e-05, | |
| "loss": 0.4422, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.9692307692307693, | |
| "eval_loss": 0.7678297162055969, | |
| "eval_runtime": 1.171, | |
| "eval_samples_per_second": 97.351, | |
| "eval_steps_per_second": 5.124, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.3216522932052612, | |
| "learning_rate": 1.797500088276232e-05, | |
| "loss": 0.3888, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.7601237297058105, | |
| "eval_runtime": 1.1955, | |
| "eval_samples_per_second": 95.354, | |
| "eval_steps_per_second": 5.019, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.0307692307692307, | |
| "grad_norm": 1.9767566919326782, | |
| "learning_rate": 1.7881472160773912e-05, | |
| "loss": 0.2158, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.0307692307692307, | |
| "eval_loss": 0.8567830920219421, | |
| "eval_runtime": 1.1711, | |
| "eval_samples_per_second": 97.344, | |
| "eval_steps_per_second": 5.123, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.0615384615384613, | |
| "grad_norm": 1.724826693534851, | |
| "learning_rate": 1.7786087075880698e-05, | |
| "loss": 0.1589, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.0615384615384613, | |
| "eval_loss": 0.8708633780479431, | |
| "eval_runtime": 1.1707, | |
| "eval_samples_per_second": 97.379, | |
| "eval_steps_per_second": 5.125, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.0923076923076924, | |
| "grad_norm": 1.9074368476867676, | |
| "learning_rate": 1.7688868094613e-05, | |
| "loss": 0.1974, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.0923076923076924, | |
| "eval_loss": 0.8557892441749573, | |
| "eval_runtime": 1.1707, | |
| "eval_samples_per_second": 97.382, | |
| "eval_steps_per_second": 5.125, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.123076923076923, | |
| "grad_norm": 1.2026904821395874, | |
| "learning_rate": 1.7589838115448005e-05, | |
| "loss": 0.1832, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.123076923076923, | |
| "eval_loss": 0.861406147480011, | |
| "eval_runtime": 1.1692, | |
| "eval_samples_per_second": 97.502, | |
| "eval_steps_per_second": 5.132, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "grad_norm": 1.541394829750061, | |
| "learning_rate": 1.748902046341637e-05, | |
| "loss": 0.1835, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "eval_loss": 0.8678261041641235, | |
| "eval_runtime": 1.1824, | |
| "eval_samples_per_second": 96.41, | |
| "eval_steps_per_second": 5.074, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.184615384615385, | |
| "grad_norm": 2.980255126953125, | |
| "learning_rate": 1.7386438884608366e-05, | |
| "loss": 0.1555, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.184615384615385, | |
| "eval_loss": 0.8728282451629639, | |
| "eval_runtime": 1.1698, | |
| "eval_samples_per_second": 97.454, | |
| "eval_steps_per_second": 5.129, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.2153846153846155, | |
| "grad_norm": 1.7097492218017578, | |
| "learning_rate": 1.7282117540580833e-05, | |
| "loss": 0.1789, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.2153846153846155, | |
| "eval_loss": 0.8826200366020203, | |
| "eval_runtime": 1.1687, | |
| "eval_samples_per_second": 97.541, | |
| "eval_steps_per_second": 5.134, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.246153846153846, | |
| "grad_norm": 2.0315046310424805, | |
| "learning_rate": 1.7176081002666295e-05, | |
| "loss": 0.1825, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.246153846153846, | |
| "eval_loss": 0.8737024068832397, | |
| "eval_runtime": 1.1727, | |
| "eval_samples_per_second": 97.211, | |
| "eval_steps_per_second": 5.116, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.276923076923077, | |
| "grad_norm": 1.5558000802993774, | |
| "learning_rate": 1.706835424618555e-05, | |
| "loss": 0.1906, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.276923076923077, | |
| "eval_loss": 0.8740295767784119, | |
| "eval_runtime": 1.1746, | |
| "eval_samples_per_second": 97.054, | |
| "eval_steps_per_second": 5.108, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 2.007802724838257, | |
| "learning_rate": 1.695896264456509e-05, | |
| "loss": 0.2171, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "eval_loss": 0.8841921091079712, | |
| "eval_runtime": 1.1694, | |
| "eval_samples_per_second": 97.487, | |
| "eval_steps_per_second": 5.131, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.3384615384615386, | |
| "grad_norm": 2.0817086696624756, | |
| "learning_rate": 1.6847931963360796e-05, | |
| "loss": 0.1993, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.3384615384615386, | |
| "eval_loss": 0.8875246644020081, | |
| "eval_runtime": 1.1683, | |
| "eval_samples_per_second": 97.576, | |
| "eval_steps_per_second": 5.136, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.3692307692307693, | |
| "grad_norm": 1.752740740776062, | |
| "learning_rate": 1.6735288354189225e-05, | |
| "loss": 0.1928, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.3692307692307693, | |
| "eval_loss": 0.8896489143371582, | |
| "eval_runtime": 1.1654, | |
| "eval_samples_per_second": 97.821, | |
| "eval_steps_per_second": 5.148, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 1.6042486429214478, | |
| "learning_rate": 1.6621058348568008e-05, | |
| "loss": 0.2059, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 0.8732131719589233, | |
| "eval_runtime": 1.1715, | |
| "eval_samples_per_second": 97.315, | |
| "eval_steps_per_second": 5.122, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.430769230769231, | |
| "grad_norm": 2.1651504039764404, | |
| "learning_rate": 1.6505268851666717e-05, | |
| "loss": 0.2101, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.430769230769231, | |
| "eval_loss": 0.8754842877388, | |
| "eval_runtime": 1.1681, | |
| "eval_samples_per_second": 97.595, | |
| "eval_steps_per_second": 5.137, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 2.9434351921081543, | |
| "learning_rate": 1.6387947135969796e-05, | |
| "loss": 0.2305, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "eval_loss": 0.9017049074172974, | |
| "eval_runtime": 1.1681, | |
| "eval_samples_per_second": 97.598, | |
| "eval_steps_per_second": 5.137, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.4923076923076923, | |
| "grad_norm": 1.2757177352905273, | |
| "learning_rate": 1.6269120834852892e-05, | |
| "loss": 0.1837, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.4923076923076923, | |
| "eval_loss": 0.8789340853691101, | |
| "eval_runtime": 1.1681, | |
| "eval_samples_per_second": 97.593, | |
| "eval_steps_per_second": 5.136, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.523076923076923, | |
| "grad_norm": 2.2374846935272217, | |
| "learning_rate": 1.6148817936074267e-05, | |
| "loss": 0.1846, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.523076923076923, | |
| "eval_loss": 0.8868066668510437, | |
| "eval_runtime": 1.1703, | |
| "eval_samples_per_second": 97.411, | |
| "eval_steps_per_second": 5.127, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.5538461538461537, | |
| "grad_norm": 2.112977981567383, | |
| "learning_rate": 1.6027066775182664e-05, | |
| "loss": 0.215, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.5538461538461537, | |
| "eval_loss": 0.8842012286186218, | |
| "eval_runtime": 1.168, | |
| "eval_samples_per_second": 97.604, | |
| "eval_steps_per_second": 5.137, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.5846153846153848, | |
| "grad_norm": 1.7254525423049927, | |
| "learning_rate": 1.5903896028843316e-05, | |
| "loss": 0.2021, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.5846153846153848, | |
| "eval_loss": 0.8752718567848206, | |
| "eval_runtime": 1.172, | |
| "eval_samples_per_second": 97.267, | |
| "eval_steps_per_second": 5.119, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.6153846153846154, | |
| "grad_norm": 2.7533833980560303, | |
| "learning_rate": 1.5779334708083585e-05, | |
| "loss": 0.2087, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.6153846153846154, | |
| "eval_loss": 0.881074070930481, | |
| "eval_runtime": 1.1742, | |
| "eval_samples_per_second": 97.086, | |
| "eval_steps_per_second": 5.11, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.646153846153846, | |
| "grad_norm": 1.364652395248413, | |
| "learning_rate": 1.565341215145983e-05, | |
| "loss": 0.205, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.646153846153846, | |
| "eval_loss": 0.8765241503715515, | |
| "eval_runtime": 1.1685, | |
| "eval_samples_per_second": 97.561, | |
| "eval_steps_per_second": 5.135, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.676923076923077, | |
| "grad_norm": 2.8074488639831543, | |
| "learning_rate": 1.5526158018147168e-05, | |
| "loss": 0.1872, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.676923076923077, | |
| "eval_loss": 0.8835176229476929, | |
| "eval_runtime": 1.1654, | |
| "eval_samples_per_second": 97.823, | |
| "eval_steps_per_second": 5.149, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.707692307692308, | |
| "grad_norm": 1.8859968185424805, | |
| "learning_rate": 1.5397602280953695e-05, | |
| "loss": 0.197, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.707692307692308, | |
| "eval_loss": 0.8719732761383057, | |
| "eval_runtime": 1.1708, | |
| "eval_samples_per_second": 97.368, | |
| "eval_steps_per_second": 5.125, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.7384615384615385, | |
| "grad_norm": 1.2429983615875244, | |
| "learning_rate": 1.526777521926084e-05, | |
| "loss": 0.1932, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.7384615384615385, | |
| "eval_loss": 0.8760843276977539, | |
| "eval_runtime": 1.1705, | |
| "eval_samples_per_second": 97.392, | |
| "eval_steps_per_second": 5.126, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": 1.7974516153335571, | |
| "learning_rate": 1.5136707411891483e-05, | |
| "loss": 0.2008, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "eval_loss": 0.8759164214134216, | |
| "eval_runtime": 1.1739, | |
| "eval_samples_per_second": 97.116, | |
| "eval_steps_per_second": 5.111, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 1.4502002000808716, | |
| "learning_rate": 1.5004429729907619e-05, | |
| "loss": 0.1998, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "eval_loss": 0.8729867935180664, | |
| "eval_runtime": 1.1659, | |
| "eval_samples_per_second": 97.775, | |
| "eval_steps_per_second": 5.146, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.830769230769231, | |
| "grad_norm": 1.3150732517242432, | |
| "learning_rate": 1.4870973329339112e-05, | |
| "loss": 0.1936, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.830769230769231, | |
| "eval_loss": 0.8798808455467224, | |
| "eval_runtime": 1.1664, | |
| "eval_samples_per_second": 97.737, | |
| "eval_steps_per_second": 5.144, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.8615384615384616, | |
| "grad_norm": 1.8317633867263794, | |
| "learning_rate": 1.4736369643845346e-05, | |
| "loss": 0.1951, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.8615384615384616, | |
| "eval_loss": 0.900560736656189, | |
| "eval_runtime": 1.1713, | |
| "eval_samples_per_second": 97.324, | |
| "eval_steps_per_second": 5.122, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.8923076923076922, | |
| "grad_norm": 2.284203290939331, | |
| "learning_rate": 1.4600650377311523e-05, | |
| "loss": 0.1884, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.8923076923076922, | |
| "eval_loss": 0.8905934691429138, | |
| "eval_runtime": 1.1697, | |
| "eval_samples_per_second": 97.462, | |
| "eval_steps_per_second": 5.13, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.9230769230769234, | |
| "grad_norm": 1.9384973049163818, | |
| "learning_rate": 1.446384749638128e-05, | |
| "loss": 0.1881, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.9230769230769234, | |
| "eval_loss": 0.8698312640190125, | |
| "eval_runtime": 1.1727, | |
| "eval_samples_per_second": 97.213, | |
| "eval_steps_per_second": 5.116, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.953846153846154, | |
| "grad_norm": 1.7943904399871826, | |
| "learning_rate": 1.4325993222927414e-05, | |
| "loss": 0.2166, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.953846153846154, | |
| "eval_loss": 0.8718281984329224, | |
| "eval_runtime": 1.1666, | |
| "eval_samples_per_second": 97.724, | |
| "eval_steps_per_second": 5.143, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.9846153846153847, | |
| "grad_norm": 2.0599663257598877, | |
| "learning_rate": 1.4187120026462508e-05, | |
| "loss": 0.2082, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 2.9846153846153847, | |
| "eval_loss": 0.8722580075263977, | |
| "eval_runtime": 1.1673, | |
| "eval_samples_per_second": 97.658, | |
| "eval_steps_per_second": 5.14, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 3.0153846153846153, | |
| "grad_norm": 1.2260268926620483, | |
| "learning_rate": 1.4047260616491225e-05, | |
| "loss": 0.1631, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 3.0153846153846153, | |
| "eval_loss": 0.8925275206565857, | |
| "eval_runtime": 1.1687, | |
| "eval_samples_per_second": 97.541, | |
| "eval_steps_per_second": 5.134, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 3.046153846153846, | |
| "grad_norm": 1.544405460357666, | |
| "learning_rate": 1.3906447934806074e-05, | |
| "loss": 0.1024, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.046153846153846, | |
| "eval_loss": 0.9651603102684021, | |
| "eval_runtime": 1.1687, | |
| "eval_samples_per_second": 97.54, | |
| "eval_steps_per_second": 5.134, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 1.8806118965148926, | |
| "learning_rate": 1.3764715147728451e-05, | |
| "loss": 0.1111, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "eval_loss": 0.9773316383361816, | |
| "eval_runtime": 1.1719, | |
| "eval_samples_per_second": 97.274, | |
| "eval_steps_per_second": 5.12, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.1076923076923078, | |
| "grad_norm": 1.2032195329666138, | |
| "learning_rate": 1.3622095638296827e-05, | |
| "loss": 0.1011, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 3.1076923076923078, | |
| "eval_loss": 0.9423761367797852, | |
| "eval_runtime": 1.1686, | |
| "eval_samples_per_second": 97.556, | |
| "eval_steps_per_second": 5.135, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 3.1384615384615384, | |
| "grad_norm": 1.2994767427444458, | |
| "learning_rate": 1.3478622998403861e-05, | |
| "loss": 0.1078, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 3.1384615384615384, | |
| "eval_loss": 0.9416558146476746, | |
| "eval_runtime": 1.1698, | |
| "eval_samples_per_second": 97.455, | |
| "eval_steps_per_second": 5.129, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 3.169230769230769, | |
| "grad_norm": 1.061914324760437, | |
| "learning_rate": 1.3334331020884328e-05, | |
| "loss": 0.0991, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 3.169230769230769, | |
| "eval_loss": 0.9638768434524536, | |
| "eval_runtime": 1.1664, | |
| "eval_samples_per_second": 97.741, | |
| "eval_steps_per_second": 5.144, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 1.4495635032653809, | |
| "learning_rate": 1.318925369155574e-05, | |
| "loss": 0.0979, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "eval_loss": 0.9724640846252441, | |
| "eval_runtime": 1.1718, | |
| "eval_samples_per_second": 97.282, | |
| "eval_steps_per_second": 5.12, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 3.230769230769231, | |
| "grad_norm": 1.5616148710250854, | |
| "learning_rate": 1.3043425181213471e-05, | |
| "loss": 0.1109, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.230769230769231, | |
| "eval_loss": 0.972594141960144, | |
| "eval_runtime": 1.169, | |
| "eval_samples_per_second": 97.515, | |
| "eval_steps_per_second": 5.132, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.2615384615384615, | |
| "grad_norm": 2.1167333126068115, | |
| "learning_rate": 1.2896879837582356e-05, | |
| "loss": 0.1047, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 3.2615384615384615, | |
| "eval_loss": 0.9514709711074829, | |
| "eval_runtime": 1.1679, | |
| "eval_samples_per_second": 97.612, | |
| "eval_steps_per_second": 5.137, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 3.292307692307692, | |
| "grad_norm": 1.5117723941802979, | |
| "learning_rate": 1.2749652177226592e-05, | |
| "loss": 0.1075, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 3.292307692307692, | |
| "eval_loss": 0.9520531296730042, | |
| "eval_runtime": 1.1746, | |
| "eval_samples_per_second": 97.058, | |
| "eval_steps_per_second": 5.108, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 3.3230769230769233, | |
| "grad_norm": 1.231341004371643, | |
| "learning_rate": 1.2601776877419876e-05, | |
| "loss": 0.1021, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 3.3230769230769233, | |
| "eval_loss": 0.9573836922645569, | |
| "eval_runtime": 1.1666, | |
| "eval_samples_per_second": 97.723, | |
| "eval_steps_per_second": 5.143, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 3.353846153846154, | |
| "grad_norm": 1.5957950353622437, | |
| "learning_rate": 1.2453288767977686e-05, | |
| "loss": 0.1069, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 3.353846153846154, | |
| "eval_loss": 0.9602900147438049, | |
| "eval_runtime": 1.169, | |
| "eval_samples_per_second": 97.519, | |
| "eval_steps_per_second": 5.133, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 3.3846153846153846, | |
| "grad_norm": 0.7704668641090393, | |
| "learning_rate": 1.2304222823053653e-05, | |
| "loss": 0.0969, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.3846153846153846, | |
| "eval_loss": 0.9564027190208435, | |
| "eval_runtime": 1.1725, | |
| "eval_samples_per_second": 97.232, | |
| "eval_steps_per_second": 5.117, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.4153846153846152, | |
| "grad_norm": 1.638713002204895, | |
| "learning_rate": 1.2154614152901916e-05, | |
| "loss": 0.1136, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 3.4153846153846152, | |
| "eval_loss": 0.9655954241752625, | |
| "eval_runtime": 1.1671, | |
| "eval_samples_per_second": 97.68, | |
| "eval_steps_per_second": 5.141, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 3.4461538461538463, | |
| "grad_norm": 1.2840876579284668, | |
| "learning_rate": 1.2004497995607415e-05, | |
| "loss": 0.1097, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 3.4461538461538463, | |
| "eval_loss": 0.9535898566246033, | |
| "eval_runtime": 1.1708, | |
| "eval_samples_per_second": 97.368, | |
| "eval_steps_per_second": 5.125, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 3.476923076923077, | |
| "grad_norm": 1.021537184715271, | |
| "learning_rate": 1.1853909708786111e-05, | |
| "loss": 0.1048, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 3.476923076923077, | |
| "eval_loss": 0.952057421207428, | |
| "eval_runtime": 1.1712, | |
| "eval_samples_per_second": 97.336, | |
| "eval_steps_per_second": 5.123, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 3.5076923076923077, | |
| "grad_norm": 2.217410087585449, | |
| "learning_rate": 1.1702884761257003e-05, | |
| "loss": 0.118, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 3.5076923076923077, | |
| "eval_loss": 0.9404497146606445, | |
| "eval_runtime": 1.1731, | |
| "eval_samples_per_second": 97.178, | |
| "eval_steps_per_second": 5.115, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 3.5384615384615383, | |
| "grad_norm": 1.5386874675750732, | |
| "learning_rate": 1.1551458724688e-05, | |
| "loss": 0.1053, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.5384615384615383, | |
| "eval_loss": 0.9536699056625366, | |
| "eval_runtime": 1.1718, | |
| "eval_samples_per_second": 97.284, | |
| "eval_steps_per_second": 5.12, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.569230769230769, | |
| "grad_norm": 0.8064364790916443, | |
| "learning_rate": 1.1399667265217522e-05, | |
| "loss": 0.1061, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 3.569230769230769, | |
| "eval_loss": 0.9544848799705505, | |
| "eval_runtime": 1.1669, | |
| "eval_samples_per_second": 97.693, | |
| "eval_steps_per_second": 5.142, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 0.917335569858551, | |
| "learning_rate": 1.1247546135053904e-05, | |
| "loss": 0.102, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "eval_loss": 0.9585210084915161, | |
| "eval_runtime": 1.17, | |
| "eval_samples_per_second": 97.434, | |
| "eval_steps_per_second": 5.128, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 3.6307692307692307, | |
| "grad_norm": 1.0991976261138916, | |
| "learning_rate": 1.1095131164054476e-05, | |
| "loss": 0.1072, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 3.6307692307692307, | |
| "eval_loss": 0.9673210978507996, | |
| "eval_runtime": 1.1692, | |
| "eval_samples_per_second": 97.5, | |
| "eval_steps_per_second": 5.132, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 3.6615384615384614, | |
| "grad_norm": 1.2590489387512207, | |
| "learning_rate": 1.0942458251286384e-05, | |
| "loss": 0.1007, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 3.6615384615384614, | |
| "eval_loss": 0.9589065313339233, | |
| "eval_runtime": 1.1693, | |
| "eval_samples_per_second": 97.491, | |
| "eval_steps_per_second": 5.131, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "grad_norm": 0.555060625076294, | |
| "learning_rate": 1.078956335657109e-05, | |
| "loss": 0.1045, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "eval_loss": 0.9449043273925781, | |
| "eval_runtime": 1.1697, | |
| "eval_samples_per_second": 97.457, | |
| "eval_steps_per_second": 5.129, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.723076923076923, | |
| "grad_norm": 1.4285695552825928, | |
| "learning_rate": 1.0636482492014603e-05, | |
| "loss": 0.1042, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 3.723076923076923, | |
| "eval_loss": 0.9505679607391357, | |
| "eval_runtime": 1.1646, | |
| "eval_samples_per_second": 97.886, | |
| "eval_steps_per_second": 5.152, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 3.753846153846154, | |
| "grad_norm": 1.0818232297897339, | |
| "learning_rate": 1.0483251713525335e-05, | |
| "loss": 0.1066, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 3.753846153846154, | |
| "eval_loss": 0.958114743232727, | |
| "eval_runtime": 1.1679, | |
| "eval_samples_per_second": 97.609, | |
| "eval_steps_per_second": 5.137, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 3.7846153846153845, | |
| "grad_norm": 1.027013897895813, | |
| "learning_rate": 1.0329907112321685e-05, | |
| "loss": 0.1103, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.7846153846153845, | |
| "eval_loss": 0.9584141969680786, | |
| "eval_runtime": 1.1717, | |
| "eval_samples_per_second": 97.292, | |
| "eval_steps_per_second": 5.121, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.815384615384615, | |
| "grad_norm": 1.2633482217788696, | |
| "learning_rate": 1.0176484806431288e-05, | |
| "loss": 0.1135, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 3.815384615384615, | |
| "eval_loss": 0.9456351399421692, | |
| "eval_runtime": 1.1713, | |
| "eval_samples_per_second": 97.325, | |
| "eval_steps_per_second": 5.122, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 1.5647493600845337, | |
| "learning_rate": 1.002302093218396e-05, | |
| "loss": 0.102, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "eval_loss": 0.9635660648345947, | |
| "eval_runtime": 1.1695, | |
| "eval_samples_per_second": 97.479, | |
| "eval_steps_per_second": 5.13, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.876923076923077, | |
| "grad_norm": 1.1846381425857544, | |
| "learning_rate": 9.869551635700321e-06, | |
| "loss": 0.0996, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 3.876923076923077, | |
| "eval_loss": 0.9652856588363647, | |
| "eval_runtime": 1.1667, | |
| "eval_samples_per_second": 97.709, | |
| "eval_steps_per_second": 5.143, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 3.9076923076923076, | |
| "grad_norm": 1.5169893503189087, | |
| "learning_rate": 9.716113064378113e-06, | |
| "loss": 0.0995, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 3.9076923076923076, | |
| "eval_loss": 0.9649413228034973, | |
| "eval_runtime": 1.1669, | |
| "eval_samples_per_second": 97.691, | |
| "eval_steps_per_second": 5.142, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 3.9384615384615387, | |
| "grad_norm": 1.013720154762268, | |
| "learning_rate": 9.562741358378239e-06, | |
| "loss": 0.1124, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 3.9384615384615387, | |
| "eval_loss": 0.9643102288246155, | |
| "eval_runtime": 1.1719, | |
| "eval_samples_per_second": 97.276, | |
| "eval_steps_per_second": 5.12, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 3.9692307692307693, | |
| "grad_norm": 1.1078283786773682, | |
| "learning_rate": 9.409472642112454e-06, | |
| "loss": 0.096, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 3.9692307692307693, | |
| "eval_loss": 0.9525210857391357, | |
| "eval_runtime": 1.1921, | |
| "eval_samples_per_second": 95.632, | |
| "eval_steps_per_second": 5.033, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.2566571235656738, | |
| "learning_rate": 9.256343015734842e-06, | |
| "loss": 0.1133, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.9479546546936035, | |
| "eval_runtime": 1.1715, | |
| "eval_samples_per_second": 97.308, | |
| "eval_steps_per_second": 5.121, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.030769230769231, | |
| "grad_norm": 1.1318272352218628, | |
| "learning_rate": 9.103388546638929e-06, | |
| "loss": 0.069, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 4.030769230769231, | |
| "eval_loss": 0.966812789440155, | |
| "eval_runtime": 1.1672, | |
| "eval_samples_per_second": 97.667, | |
| "eval_steps_per_second": 5.14, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 4.061538461538461, | |
| "grad_norm": 0.9845967292785645, | |
| "learning_rate": 8.950645260962572e-06, | |
| "loss": 0.0716, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 4.061538461538461, | |
| "eval_loss": 1.0023221969604492, | |
| "eval_runtime": 1.17, | |
| "eval_samples_per_second": 97.437, | |
| "eval_steps_per_second": 5.128, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 4.092307692307692, | |
| "grad_norm": 0.5891923904418945, | |
| "learning_rate": 8.798149135102528e-06, | |
| "loss": 0.0637, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 4.092307692307692, | |
| "eval_loss": 1.0226399898529053, | |
| "eval_runtime": 1.1693, | |
| "eval_samples_per_second": 97.493, | |
| "eval_steps_per_second": 5.131, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 4.123076923076923, | |
| "grad_norm": 0.9881710410118103, | |
| "learning_rate": 8.645936087240758e-06, | |
| "loss": 0.0695, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 4.123076923076923, | |
| "eval_loss": 1.0229477882385254, | |
| "eval_runtime": 1.1678, | |
| "eval_samples_per_second": 97.615, | |
| "eval_steps_per_second": 5.138, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 4.153846153846154, | |
| "grad_norm": 0.47998377680778503, | |
| "learning_rate": 8.494041968884423e-06, | |
| "loss": 0.0628, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 4.153846153846154, | |
| "eval_loss": 1.0193408727645874, | |
| "eval_runtime": 1.1721, | |
| "eval_samples_per_second": 97.261, | |
| "eval_steps_per_second": 5.119, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 4.184615384615385, | |
| "grad_norm": 0.6444534063339233, | |
| "learning_rate": 8.342502556421627e-06, | |
| "loss": 0.0711, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 4.184615384615385, | |
| "eval_loss": 1.0234897136688232, | |
| "eval_runtime": 1.1655, | |
| "eval_samples_per_second": 97.808, | |
| "eval_steps_per_second": 5.148, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 4.2153846153846155, | |
| "grad_norm": 1.1395206451416016, | |
| "learning_rate": 8.19135354269479e-06, | |
| "loss": 0.0662, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 4.2153846153846155, | |
| "eval_loss": 1.027916669845581, | |
| "eval_runtime": 1.1668, | |
| "eval_samples_per_second": 97.702, | |
| "eval_steps_per_second": 5.142, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 4.246153846153846, | |
| "grad_norm": 0.6894501447677612, | |
| "learning_rate": 8.040630528593753e-06, | |
| "loss": 0.0748, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 4.246153846153846, | |
| "eval_loss": 1.0266273021697998, | |
| "eval_runtime": 1.1674, | |
| "eval_samples_per_second": 97.65, | |
| "eval_steps_per_second": 5.139, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 4.276923076923077, | |
| "grad_norm": 0.6127138137817383, | |
| "learning_rate": 7.890369014670512e-06, | |
| "loss": 0.0571, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 4.276923076923077, | |
| "eval_loss": 1.0378963947296143, | |
| "eval_runtime": 1.169, | |
| "eval_samples_per_second": 97.523, | |
| "eval_steps_per_second": 5.133, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 4.3076923076923075, | |
| "grad_norm": 1.168372392654419, | |
| "learning_rate": 7.740604392777612e-06, | |
| "loss": 0.067, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.3076923076923075, | |
| "eval_loss": 1.0480735301971436, | |
| "eval_runtime": 1.1695, | |
| "eval_samples_per_second": 97.475, | |
| "eval_steps_per_second": 5.13, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.338461538461538, | |
| "grad_norm": 0.5983268022537231, | |
| "learning_rate": 7.591371937732091e-06, | |
| "loss": 0.0691, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 4.338461538461538, | |
| "eval_loss": 1.0469272136688232, | |
| "eval_runtime": 1.1663, | |
| "eval_samples_per_second": 97.748, | |
| "eval_steps_per_second": 5.145, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 4.36923076923077, | |
| "grad_norm": 0.7790025472640991, | |
| "learning_rate": 7.442706799007056e-06, | |
| "loss": 0.0711, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 4.36923076923077, | |
| "eval_loss": 1.0372542142868042, | |
| "eval_runtime": 1.1684, | |
| "eval_samples_per_second": 97.571, | |
| "eval_steps_per_second": 5.135, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 1.0238187313079834, | |
| "learning_rate": 7.294643992452735e-06, | |
| "loss": 0.0737, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "eval_loss": 1.03952157497406, | |
| "eval_runtime": 1.1679, | |
| "eval_samples_per_second": 97.615, | |
| "eval_steps_per_second": 5.138, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 4.430769230769231, | |
| "grad_norm": 1.3187087774276733, | |
| "learning_rate": 7.147218392049026e-06, | |
| "loss": 0.0673, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 4.430769230769231, | |
| "eval_loss": 1.036614179611206, | |
| "eval_runtime": 1.169, | |
| "eval_samples_per_second": 97.522, | |
| "eval_steps_per_second": 5.133, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 4.461538461538462, | |
| "grad_norm": 1.0066938400268555, | |
| "learning_rate": 7.000464721691438e-06, | |
| "loss": 0.0671, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 4.461538461538462, | |
| "eval_loss": 1.033462643623352, | |
| "eval_runtime": 1.1706, | |
| "eval_samples_per_second": 97.388, | |
| "eval_steps_per_second": 5.126, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 4.492307692307692, | |
| "grad_norm": 0.8664164543151855, | |
| "learning_rate": 6.854417547012415e-06, | |
| "loss": 0.0637, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 4.492307692307692, | |
| "eval_loss": 1.0385520458221436, | |
| "eval_runtime": 1.1667, | |
| "eval_samples_per_second": 97.708, | |
| "eval_steps_per_second": 5.143, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 4.523076923076923, | |
| "grad_norm": 0.41295212507247925, | |
| "learning_rate": 6.7091112672399e-06, | |
| "loss": 0.0626, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 4.523076923076923, | |
| "eval_loss": 1.0354348421096802, | |
| "eval_runtime": 1.1643, | |
| "eval_samples_per_second": 97.913, | |
| "eval_steps_per_second": 5.153, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 4.553846153846154, | |
| "grad_norm": 1.439942479133606, | |
| "learning_rate": 6.564580107095133e-06, | |
| "loss": 0.0675, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 4.553846153846154, | |
| "eval_loss": 1.0380290746688843, | |
| "eval_runtime": 1.1702, | |
| "eval_samples_per_second": 97.415, | |
| "eval_steps_per_second": 5.127, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 4.584615384615384, | |
| "grad_norm": 0.6881715059280396, | |
| "learning_rate": 6.4208581087315035e-06, | |
| "loss": 0.0662, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 4.584615384615384, | |
| "eval_loss": 1.029496192932129, | |
| "eval_runtime": 1.1716, | |
| "eval_samples_per_second": 97.305, | |
| "eval_steps_per_second": 5.121, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 0.5975565314292908, | |
| "learning_rate": 6.277979123716455e-06, | |
| "loss": 0.0628, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "eval_loss": 1.0207117795944214, | |
| "eval_runtime": 1.17, | |
| "eval_samples_per_second": 97.434, | |
| "eval_steps_per_second": 5.128, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.6461538461538465, | |
| "grad_norm": 0.7314721345901489, | |
| "learning_rate": 6.13597680505823e-06, | |
| "loss": 0.0748, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 4.6461538461538465, | |
| "eval_loss": 1.021004557609558, | |
| "eval_runtime": 1.1652, | |
| "eval_samples_per_second": 97.839, | |
| "eval_steps_per_second": 5.149, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 4.676923076923077, | |
| "grad_norm": 1.1508655548095703, | |
| "learning_rate": 5.994884599279443e-06, | |
| "loss": 0.0707, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 4.676923076923077, | |
| "eval_loss": 1.023290753364563, | |
| "eval_runtime": 1.1651, | |
| "eval_samples_per_second": 97.845, | |
| "eval_steps_per_second": 5.15, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 4.707692307692308, | |
| "grad_norm": 0.9571641087532043, | |
| "learning_rate": 5.854735738539203e-06, | |
| "loss": 0.0672, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 4.707692307692308, | |
| "eval_loss": 1.026742935180664, | |
| "eval_runtime": 1.1684, | |
| "eval_samples_per_second": 97.565, | |
| "eval_steps_per_second": 5.135, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 4.7384615384615385, | |
| "grad_norm": 0.8185418248176575, | |
| "learning_rate": 5.715563232805825e-06, | |
| "loss": 0.0606, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 4.7384615384615385, | |
| "eval_loss": 1.032530426979065, | |
| "eval_runtime": 1.1692, | |
| "eval_samples_per_second": 97.5, | |
| "eval_steps_per_second": 5.132, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 4.769230769230769, | |
| "grad_norm": 0.9479151368141174, | |
| "learning_rate": 5.577399862081789e-06, | |
| "loss": 0.0632, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 4.769230769230769, | |
| "eval_loss": 1.039183497428894, | |
| "eval_runtime": 1.1755, | |
| "eval_samples_per_second": 96.981, | |
| "eval_steps_per_second": 5.104, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.7193753123283386, | |
| "learning_rate": 5.4402781686829184e-06, | |
| "loss": 0.0709, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "eval_loss": 1.0410054922103882, | |
| "eval_runtime": 1.1653, | |
| "eval_samples_per_second": 97.831, | |
| "eval_steps_per_second": 5.149, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 4.8307692307692305, | |
| "grad_norm": 0.9305492639541626, | |
| "learning_rate": 5.304230449573523e-06, | |
| "loss": 0.0791, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 4.8307692307692305, | |
| "eval_loss": 1.0399616956710815, | |
| "eval_runtime": 1.17, | |
| "eval_samples_per_second": 97.434, | |
| "eval_steps_per_second": 5.128, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 4.861538461538462, | |
| "grad_norm": 0.687660276889801, | |
| "learning_rate": 5.169288748759327e-06, | |
| "loss": 0.0711, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 4.861538461538462, | |
| "eval_loss": 1.0336360931396484, | |
| "eval_runtime": 1.1698, | |
| "eval_samples_per_second": 97.451, | |
| "eval_steps_per_second": 5.129, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 4.892307692307693, | |
| "grad_norm": 0.8751097321510315, | |
| "learning_rate": 5.0354848497399865e-06, | |
| "loss": 0.0642, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 4.892307692307693, | |
| "eval_loss": 1.025221824645996, | |
| "eval_runtime": 1.1733, | |
| "eval_samples_per_second": 97.161, | |
| "eval_steps_per_second": 5.114, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "grad_norm": 0.4916076958179474, | |
| "learning_rate": 4.902850268022959e-06, | |
| "loss": 0.0686, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "eval_loss": 1.0251351594924927, | |
| "eval_runtime": 1.1702, | |
| "eval_samples_per_second": 97.42, | |
| "eval_steps_per_second": 5.127, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.953846153846154, | |
| "grad_norm": 0.9400696754455566, | |
| "learning_rate": 4.771416243700495e-06, | |
| "loss": 0.0733, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 4.953846153846154, | |
| "eval_loss": 1.0282999277114868, | |
| "eval_runtime": 1.1685, | |
| "eval_samples_per_second": 97.563, | |
| "eval_steps_per_second": 5.135, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 4.984615384615385, | |
| "grad_norm": 0.5670651197433472, | |
| "learning_rate": 4.641213734091507e-06, | |
| "loss": 0.0664, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 4.984615384615385, | |
| "eval_loss": 1.029799461364746, | |
| "eval_runtime": 1.1659, | |
| "eval_samples_per_second": 97.776, | |
| "eval_steps_per_second": 5.146, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 5.015384615384615, | |
| "grad_norm": 0.3696945607662201, | |
| "learning_rate": 4.5122734064500365e-06, | |
| "loss": 0.0547, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 5.015384615384615, | |
| "eval_loss": 1.037708044052124, | |
| "eval_runtime": 1.1669, | |
| "eval_samples_per_second": 97.696, | |
| "eval_steps_per_second": 5.142, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 5.046153846153846, | |
| "grad_norm": 0.6425753831863403, | |
| "learning_rate": 4.384625630742031e-06, | |
| "loss": 0.0465, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 5.046153846153846, | |
| "eval_loss": 1.0521013736724854, | |
| "eval_runtime": 1.1713, | |
| "eval_samples_per_second": 97.324, | |
| "eval_steps_per_second": 5.122, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 5.076923076923077, | |
| "grad_norm": 0.4301888346672058, | |
| "learning_rate": 4.258300472492165e-06, | |
| "loss": 0.0472, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 5.076923076923077, | |
| "eval_loss": 1.0649975538253784, | |
| "eval_runtime": 1.172, | |
| "eval_samples_per_second": 97.273, | |
| "eval_steps_per_second": 5.12, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 5.107692307692307, | |
| "grad_norm": 0.37904953956604004, | |
| "learning_rate": 4.1333276857023515e-06, | |
| "loss": 0.0499, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 5.107692307692307, | |
| "eval_loss": 1.072283148765564, | |
| "eval_runtime": 1.1667, | |
| "eval_samples_per_second": 97.708, | |
| "eval_steps_per_second": 5.143, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 5.138461538461539, | |
| "grad_norm": 0.5074180364608765, | |
| "learning_rate": 4.0097367058436156e-06, | |
| "loss": 0.0468, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 5.138461538461539, | |
| "eval_loss": 1.080543041229248, | |
| "eval_runtime": 1.1689, | |
| "eval_samples_per_second": 97.528, | |
| "eval_steps_per_second": 5.133, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 5.1692307692307695, | |
| "grad_norm": 0.5539130568504333, | |
| "learning_rate": 3.887556642923047e-06, | |
| "loss": 0.0481, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 5.1692307692307695, | |
| "eval_loss": 1.084172248840332, | |
| "eval_runtime": 1.1733, | |
| "eval_samples_per_second": 97.159, | |
| "eval_steps_per_second": 5.114, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "grad_norm": 0.3505866825580597, | |
| "learning_rate": 3.7668162746273283e-06, | |
| "loss": 0.046, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "eval_loss": 1.0881775617599487, | |
| "eval_runtime": 1.1713, | |
| "eval_samples_per_second": 97.327, | |
| "eval_steps_per_second": 5.122, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 5.230769230769231, | |
| "grad_norm": 0.438381552696228, | |
| "learning_rate": 3.647544039544615e-06, | |
| "loss": 0.0488, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 5.230769230769231, | |
| "eval_loss": 1.0943236351013184, | |
| "eval_runtime": 1.1694, | |
| "eval_samples_per_second": 97.489, | |
| "eval_steps_per_second": 5.131, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 5.2615384615384615, | |
| "grad_norm": 0.6292533278465271, | |
| "learning_rate": 3.5297680304662374e-06, | |
| "loss": 0.0514, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 5.2615384615384615, | |
| "eval_loss": 1.0977230072021484, | |
| "eval_runtime": 1.1685, | |
| "eval_samples_per_second": 97.562, | |
| "eval_steps_per_second": 5.135, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 5.292307692307692, | |
| "grad_norm": 1.0136582851409912, | |
| "learning_rate": 3.4135159877698633e-06, | |
| "loss": 0.0509, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 5.292307692307692, | |
| "eval_loss": 1.0975443124771118, | |
| "eval_runtime": 1.1684, | |
| "eval_samples_per_second": 97.57, | |
| "eval_steps_per_second": 5.135, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 5.323076923076923, | |
| "grad_norm": 0.5158536434173584, | |
| "learning_rate": 3.29881529288567e-06, | |
| "loss": 0.0421, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 5.323076923076923, | |
| "eval_loss": 1.0977866649627686, | |
| "eval_runtime": 1.1671, | |
| "eval_samples_per_second": 97.68, | |
| "eval_steps_per_second": 5.141, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 5.3538461538461535, | |
| "grad_norm": 0.39954596757888794, | |
| "learning_rate": 3.1856929618470635e-06, | |
| "loss": 0.0439, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 5.3538461538461535, | |
| "eval_loss": 1.1025538444519043, | |
| "eval_runtime": 1.1755, | |
| "eval_samples_per_second": 96.979, | |
| "eval_steps_per_second": 5.104, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 5.384615384615385, | |
| "grad_norm": 0.723860502243042, | |
| "learning_rate": 3.0741756389274325e-06, | |
| "loss": 0.056, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 5.384615384615385, | |
| "eval_loss": 1.102371096611023, | |
| "eval_runtime": 1.1735, | |
| "eval_samples_per_second": 97.148, | |
| "eval_steps_per_second": 5.113, | |
| "step": 1750 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2275, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 50, | |
| "total_flos": 1.0842145111932928e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |