| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 56415, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.911371089249314e-05, | |
| "loss": 0.156, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 9.822742178498626e-05, | |
| "loss": 0.1245, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 9.73411326774794e-05, | |
| "loss": 0.1381, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 9.645484356997254e-05, | |
| "loss": 0.1251, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 9.556855446246566e-05, | |
| "loss": 0.123, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 9.468226535495879e-05, | |
| "loss": 0.1149, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 9.379597624745192e-05, | |
| "loss": 0.1299, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 9.290968713994506e-05, | |
| "loss": 0.1237, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 9.20233980324382e-05, | |
| "loss": 0.1249, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 9.113710892493131e-05, | |
| "loss": 0.1191, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 9.025081981742445e-05, | |
| "loss": 0.1135, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 8.936453070991759e-05, | |
| "loss": 0.114, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 8.84782416024107e-05, | |
| "loss": 0.1165, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 8.759195249490384e-05, | |
| "loss": 0.1191, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 8.670566338739697e-05, | |
| "loss": 0.1164, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 8.581937427989011e-05, | |
| "loss": 0.1211, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 8.493308517238323e-05, | |
| "loss": 0.1091, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 8.404679606487636e-05, | |
| "loss": 0.1096, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 8.31605069573695e-05, | |
| "loss": 0.1117, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 8.227421784986264e-05, | |
| "loss": 0.1125, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 8.138792874235575e-05, | |
| "loss": 0.1193, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 8.050163963484889e-05, | |
| "loss": 0.1094, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.10602948069572449, | |
| "eval_runtime": 418.0898, | |
| "eval_samples_per_second": 6.747, | |
| "eval_steps_per_second": 3.375, | |
| "step": 11283 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 7.961535052734203e-05, | |
| "loss": 0.0908, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 7.872906141983516e-05, | |
| "loss": 0.0683, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 7.784277231232828e-05, | |
| "loss": 0.0725, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 7.695648320482141e-05, | |
| "loss": 0.0723, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 7.607019409731455e-05, | |
| "loss": 0.0707, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 7.518390498980768e-05, | |
| "loss": 0.0731, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 7.42976158823008e-05, | |
| "loss": 0.0687, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 7.341132677479394e-05, | |
| "loss": 0.0686, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 7.252503766728708e-05, | |
| "loss": 0.0717, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 7.16387485597802e-05, | |
| "loss": 0.0703, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 7.075245945227334e-05, | |
| "loss": 0.0702, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 6.986617034476646e-05, | |
| "loss": 0.0687, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 6.89798812372596e-05, | |
| "loss": 0.0695, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 6.809359212975273e-05, | |
| "loss": 0.0667, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 6.720730302224585e-05, | |
| "loss": 0.0706, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 6.6321013914739e-05, | |
| "loss": 0.061, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 6.543472480723212e-05, | |
| "loss": 0.063, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 6.454843569972525e-05, | |
| "loss": 0.0692, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 6.366214659221839e-05, | |
| "loss": 0.0691, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 6.277585748471153e-05, | |
| "loss": 0.0649, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 6.188956837720464e-05, | |
| "loss": 0.0629, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 6.100327926969778e-05, | |
| "loss": 0.0678, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 6.011699016219091e-05, | |
| "loss": 0.0659, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.09332570433616638, | |
| "eval_runtime": 419.0801, | |
| "eval_samples_per_second": 6.731, | |
| "eval_steps_per_second": 3.367, | |
| "step": 22566 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 5.9230701054684045e-05, | |
| "loss": 0.0358, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 5.834441194717717e-05, | |
| "loss": 0.0333, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 5.7458122839670305e-05, | |
| "loss": 0.034, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 5.657183373216344e-05, | |
| "loss": 0.0331, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 5.568554462465656e-05, | |
| "loss": 0.0353, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 5.47992555171497e-05, | |
| "loss": 0.0362, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 5.391296640964283e-05, | |
| "loss": 0.0339, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 5.302667730213596e-05, | |
| "loss": 0.0332, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 5.214038819462909e-05, | |
| "loss": 0.0343, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 5.125409908712222e-05, | |
| "loss": 0.0337, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 5.0367809979615356e-05, | |
| "loss": 0.0324, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 4.948152087210848e-05, | |
| "loss": 0.0341, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 4.8595231764601615e-05, | |
| "loss": 0.0332, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 4.770894265709475e-05, | |
| "loss": 0.0346, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 4.682265354958788e-05, | |
| "loss": 0.0346, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 4.593636444208101e-05, | |
| "loss": 0.034, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 4.505007533457414e-05, | |
| "loss": 0.0371, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 4.4163786227067274e-05, | |
| "loss": 0.0335, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 4.32774971195604e-05, | |
| "loss": 0.0337, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 4.239120801205353e-05, | |
| "loss": 0.0332, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 4.1504918904546666e-05, | |
| "loss": 0.0325, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 4.06186297970398e-05, | |
| "loss": 0.0335, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.09837915003299713, | |
| "eval_runtime": 419.1038, | |
| "eval_samples_per_second": 6.731, | |
| "eval_steps_per_second": 3.367, | |
| "step": 33849 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 3.9732340689532926e-05, | |
| "loss": 0.028, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 3.884605158202606e-05, | |
| "loss": 0.0163, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 3.795976247451919e-05, | |
| "loss": 0.0153, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 3.7073473367012325e-05, | |
| "loss": 0.0165, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 3.618718425950545e-05, | |
| "loss": 0.0167, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 3.5300895151998584e-05, | |
| "loss": 0.0164, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 3.441460604449172e-05, | |
| "loss": 0.0166, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 3.3528316936984844e-05, | |
| "loss": 0.0185, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 3.264202782947798e-05, | |
| "loss": 0.0174, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 3.175573872197111e-05, | |
| "loss": 0.0176, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 3.086944961446424e-05, | |
| "loss": 0.0171, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 2.998316050695737e-05, | |
| "loss": 0.0167, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 2.9096871399450502e-05, | |
| "loss": 0.0177, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 2.8210582291943632e-05, | |
| "loss": 0.0173, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 2.7324293184436765e-05, | |
| "loss": 0.0178, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 2.6438004076929895e-05, | |
| "loss": 0.0168, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 2.555171496942303e-05, | |
| "loss": 0.0172, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 2.4665425861916157e-05, | |
| "loss": 0.0171, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 2.377913675440929e-05, | |
| "loss": 0.0173, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 2.289284764690242e-05, | |
| "loss": 0.0165, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 2.2006558539395553e-05, | |
| "loss": 0.0167, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 2.1120269431888683e-05, | |
| "loss": 0.0169, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 2.0233980324381816e-05, | |
| "loss": 0.0161, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.10828191787004471, | |
| "eval_runtime": 419.188, | |
| "eval_samples_per_second": 6.73, | |
| "eval_steps_per_second": 3.366, | |
| "step": 45132 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 1.9347691216874946e-05, | |
| "loss": 0.0116, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 1.8461402109368076e-05, | |
| "loss": 0.0098, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 1.757511300186121e-05, | |
| "loss": 0.0094, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 1.6688823894354338e-05, | |
| "loss": 0.01, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 1.580253478684747e-05, | |
| "loss": 0.0099, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 1.4916245679340601e-05, | |
| "loss": 0.0099, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 1.4029956571833732e-05, | |
| "loss": 0.0098, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 1.3143667464326864e-05, | |
| "loss": 0.0097, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 1.2257378356819995e-05, | |
| "loss": 0.0097, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "learning_rate": 1.1371089249313127e-05, | |
| "loss": 0.01, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 1.0484800141806258e-05, | |
| "loss": 0.0101, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 9.59851103429939e-06, | |
| "loss": 0.01, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 8.71222192679252e-06, | |
| "loss": 0.0141, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "learning_rate": 7.825932819285652e-06, | |
| "loss": 0.0097, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 6.939643711778783e-06, | |
| "loss": 0.0097, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 6.053354604271914e-06, | |
| "loss": 0.0111, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 5.1670654967650455e-06, | |
| "loss": 0.0097, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 4.280776389258176e-06, | |
| "loss": 0.0102, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 3.3944872817513074e-06, | |
| "loss": 0.0097, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 2.508198174244439e-06, | |
| "loss": 0.0097, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 1.62190906673757e-06, | |
| "loss": 0.0098, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 7.35619959230701e-07, | |
| "loss": 0.0095, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.11257254332304001, | |
| "eval_runtime": 419.1321, | |
| "eval_samples_per_second": 6.731, | |
| "eval_steps_per_second": 3.366, | |
| "step": 56415 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 56415, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 4.9105395843072e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |