| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 371940, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 3.125e-05, | |
| "loss": 6.2191, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 6.25e-05, | |
| "loss": 5.0107, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.375e-05, | |
| "loss": 4.6777, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.000125, | |
| "loss": 4.4612, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00015625, | |
| "loss": 4.2983, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0001875, | |
| "loss": 4.1806, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00021875, | |
| "loss": 4.0861, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00025, | |
| "loss": 3.9941, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00028125000000000003, | |
| "loss": 3.9122, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00031246875000000003, | |
| "loss": 3.8513, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00034371875, | |
| "loss": 3.8017, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0003749375, | |
| "loss": 3.7641, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00040615625, | |
| "loss": 3.7323, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00043740625, | |
| "loss": 3.6983, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.00046865625, | |
| "loss": 3.6707, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.000499875, | |
| "loss": 3.6495, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.000531125, | |
| "loss": 3.6264, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.00056234375, | |
| "loss": 3.6043, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.3595355462297898, | |
| "eval_loss": 3.7892966270446777, | |
| "eval_runtime": 151.5834, | |
| "eval_samples_per_second": 382.087, | |
| "eval_steps_per_second": 5.97, | |
| "step": 18597 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.00059359375, | |
| "loss": 3.5739, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.0006248437500000001, | |
| "loss": 3.5503, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.0006560625, | |
| "loss": 3.5388, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.0006873125, | |
| "loss": 3.5243, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0007185625, | |
| "loss": 3.5177, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.00074978125, | |
| "loss": 3.5038, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.0007810312499999999, | |
| "loss": 3.4971, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.00081225, | |
| "loss": 3.4858, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.00084346875, | |
| "loss": 3.4757, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.00087471875, | |
| "loss": 3.4682, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.00090596875, | |
| "loss": 3.4638, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.00093721875, | |
| "loss": 3.4533, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.0009684375, | |
| "loss": 3.4467, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.000999625, | |
| "loss": 3.4389, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 0.0009970936047537802, | |
| "loss": 3.4276, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 0.00099415190916044, | |
| "loss": 3.4181, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.0009912102135671002, | |
| "loss": 3.4015, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.00098826851797376, | |
| "loss": 3.3918, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.00098532682238042, | |
| "loss": 3.3863, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.3810844608834468, | |
| "eval_loss": 3.57956862449646, | |
| "eval_runtime": 153.178, | |
| "eval_samples_per_second": 378.109, | |
| "eval_steps_per_second": 5.908, | |
| "step": 37194 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.00098238512678708, | |
| "loss": 3.3371, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 0.0009794493145849268, | |
| "loss": 3.3294, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.0009765076189915868, | |
| "loss": 3.3237, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 0.0009735659233982468, | |
| "loss": 3.3196, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0009706242278049068, | |
| "loss": 3.3158, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 0.0009676854739071601, | |
| "loss": 3.3138, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 0.0009647437783138201, | |
| "loss": 3.3055, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 0.0009618020827204801, | |
| "loss": 3.2988, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 0.0009588603871271401, | |
| "loss": 3.298, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.0009559216332293934, | |
| "loss": 3.2918, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 0.0009529799376360534, | |
| "loss": 3.2842, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 0.0009500382420427134, | |
| "loss": 3.28, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.0009470994881449668, | |
| "loss": 3.2764, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 0.0009441577925516268, | |
| "loss": 3.2759, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.0009412190386538801, | |
| "loss": 3.2688, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 0.0009382773430605401, | |
| "loss": 3.2608, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 0.0009353356474672001, | |
| "loss": 3.2592, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 0.0009323939518738601, | |
| "loss": 3.2568, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.39334223092543424, | |
| "eval_loss": 3.4811289310455322, | |
| "eval_runtime": 153.2475, | |
| "eval_samples_per_second": 377.938, | |
| "eval_steps_per_second": 5.905, | |
| "step": 55791 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 0.0009294581396717068, | |
| "loss": 3.2389, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 0.0009265193857739601, | |
| "loss": 3.1872, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 0.0009235776901806201, | |
| "loss": 3.1931, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 0.0009206359945872801, | |
| "loss": 3.1976, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 0.0009176942989939401, | |
| "loss": 3.1911, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 0.0009147555450961935, | |
| "loss": 3.1961, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 0.0009118138495028535, | |
| "loss": 3.1993, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.0009088750956051068, | |
| "loss": 3.192, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 0.0009059334000117668, | |
| "loss": 3.1942, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 0.0009029917044184268, | |
| "loss": 3.1933, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 0.0009000529505206801, | |
| "loss": 3.1915, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 0.0008971112549273401, | |
| "loss": 3.1888, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 0.0008941695593340001, | |
| "loss": 3.1879, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 0.0008912308054362534, | |
| "loss": 3.1842, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 0.0008882891098429134, | |
| "loss": 3.1877, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 0.0008853474142495734, | |
| "loss": 3.182, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 0.0008824057186562334, | |
| "loss": 3.1792, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 0.0008794669647584868, | |
| "loss": 3.1796, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 0.0008765252691651468, | |
| "loss": 3.1802, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3991548899873604, | |
| "eval_loss": 3.431647777557373, | |
| "eval_runtime": 153.028, | |
| "eval_samples_per_second": 378.48, | |
| "eval_steps_per_second": 5.914, | |
| "step": 74388 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 0.0008735865152674002, | |
| "loss": 3.1347, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 0.0008706448196740601, | |
| "loss": 3.1147, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 0.0008677060657763135, | |
| "loss": 3.1163, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 0.0008647643701829735, | |
| "loss": 3.1204, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 0.0008618256162852268, | |
| "loss": 3.1204, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 0.0008588868623874801, | |
| "loss": 3.1244, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 0.0008559451667941401, | |
| "loss": 3.1271, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 0.0008530034712008001, | |
| "loss": 3.1253, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 0.0008500617756074601, | |
| "loss": 3.123, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 0.0008471230217097134, | |
| "loss": 3.1278, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 0.0008441813261163735, | |
| "loss": 3.1238, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 0.0008412425722186268, | |
| "loss": 3.1218, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 0.0008383008766252868, | |
| "loss": 3.1243, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 0.0008353591810319468, | |
| "loss": 3.1246, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 0.0008324204271342001, | |
| "loss": 3.1234, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 0.0008294787315408601, | |
| "loss": 3.1253, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 0.0008265399776431135, | |
| "loss": 3.1232, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 0.0008235982820497735, | |
| "loss": 3.1237, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.4032583702465165, | |
| "eval_loss": 3.3912699222564697, | |
| "eval_runtime": 152.8134, | |
| "eval_samples_per_second": 379.011, | |
| "eval_steps_per_second": 5.922, | |
| "step": 92985 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0008206565864564334, | |
| "loss": 3.1201, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 0.0008177178325586868, | |
| "loss": 3.0522, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 0.0008147761369653469, | |
| "loss": 3.0558, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 0.0008118373830676001, | |
| "loss": 3.0644, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 0.0008088956874742603, | |
| "loss": 3.0647, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 0.0008059569335765135, | |
| "loss": 3.072, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "learning_rate": 0.0008030152379831736, | |
| "loss": 3.0749, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 0.0008000735423898336, | |
| "loss": 3.0706, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 0.0007971347884920869, | |
| "loss": 3.0745, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "learning_rate": 0.0007941930928987469, | |
| "loss": 3.0719, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 0.0007912543390010001, | |
| "loss": 3.0772, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "learning_rate": 0.0007883126434076602, | |
| "loss": 3.0788, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 0.0007853709478143202, | |
| "loss": 3.0786, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 0.0007824321939165735, | |
| "loss": 3.0797, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "learning_rate": 0.0007794904983232335, | |
| "loss": 3.0815, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 0.0007765488027298935, | |
| "loss": 3.0788, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 0.0007736071071365536, | |
| "loss": 3.0783, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 5.91, | |
| "learning_rate": 0.0007706654115432136, | |
| "loss": 3.0784, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "learning_rate": 0.0007677266576454669, | |
| "loss": 3.0797, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.40422854638309924, | |
| "eval_loss": 3.4136133193969727, | |
| "eval_runtime": 153.3253, | |
| "eval_samples_per_second": 377.746, | |
| "eval_steps_per_second": 5.902, | |
| "step": 111582 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 0.0007647879037477203, | |
| "loss": 3.0463, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 0.0007618462081543802, | |
| "loss": 3.0101, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "learning_rate": 0.0007589045125610402, | |
| "loss": 3.017, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 6.18, | |
| "learning_rate": 0.0007559628169677002, | |
| "loss": 3.0259, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 0.0007530240630699535, | |
| "loss": 3.0265, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 0.0007500823674766135, | |
| "loss": 3.0314, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 0.0007471436135788669, | |
| "loss": 3.0286, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.0007442019179855268, | |
| "loss": 3.0334, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 0.0007412602223921868, | |
| "loss": 3.0326, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "learning_rate": 0.0007383214684944403, | |
| "loss": 3.034, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 0.0007353797729011003, | |
| "loss": 3.0373, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "learning_rate": 0.0007324439606989469, | |
| "loss": 3.0405, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 0.0007295022651056069, | |
| "loss": 3.0389, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 0.0007265605695122669, | |
| "loss": 3.0422, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 0.0007236188739189269, | |
| "loss": 3.0432, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 0.0007206801200211802, | |
| "loss": 3.0428, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "learning_rate": 0.0007177384244278402, | |
| "loss": 3.0404, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 0.0007147996705300935, | |
| "loss": 3.0411, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 0.0007118579749367535, | |
| "loss": 3.0447, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.40579288066329655, | |
| "eval_loss": 3.3948402404785156, | |
| "eval_runtime": 153.079, | |
| "eval_samples_per_second": 378.354, | |
| "eval_steps_per_second": 5.912, | |
| "step": 130179 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 0.0007089162793434135, | |
| "loss": 2.9826, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 0.0007059775254456669, | |
| "loss": 2.9794, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 7.15, | |
| "learning_rate": 0.0007030358298523269, | |
| "loss": 2.9853, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "learning_rate": 0.0007000941342589869, | |
| "loss": 2.9905, | |
| "step": 134000 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 0.0006971553803612403, | |
| "loss": 2.9933, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "learning_rate": 0.0006942136847679002, | |
| "loss": 2.9951, | |
| "step": 136000 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 0.0006912719891745602, | |
| "loss": 2.9955, | |
| "step": 137000 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 0.0006883302935812202, | |
| "loss": 2.9979, | |
| "step": 138000 | |
| }, | |
| { | |
| "epoch": 7.47, | |
| "learning_rate": 0.0006853944813790669, | |
| "loss": 3.005, | |
| "step": 139000 | |
| }, | |
| { | |
| "epoch": 7.53, | |
| "learning_rate": 0.0006824527857857269, | |
| "loss": 3.0049, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "learning_rate": 0.0006795140318879802, | |
| "loss": 3.0031, | |
| "step": 141000 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "learning_rate": 0.0006765723362946402, | |
| "loss": 3.0013, | |
| "step": 142000 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 0.0006736306407013002, | |
| "loss": 3.0099, | |
| "step": 143000 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 0.0006706918868035535, | |
| "loss": 3.0063, | |
| "step": 144000 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "learning_rate": 0.0006677501912102136, | |
| "loss": 3.0065, | |
| "step": 145000 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "learning_rate": 0.0006648084956168736, | |
| "loss": 3.0124, | |
| "step": 146000 | |
| }, | |
| { | |
| "epoch": 7.9, | |
| "learning_rate": 0.0006618697417191269, | |
| "loss": 3.0132, | |
| "step": 147000 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "learning_rate": 0.0006589280461257869, | |
| "loss": 3.0084, | |
| "step": 148000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.40786423795455884, | |
| "eval_loss": 3.377242088317871, | |
| "eval_runtime": 153.122, | |
| "eval_samples_per_second": 378.247, | |
| "eval_steps_per_second": 5.91, | |
| "step": 148776 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 0.0006559863505324469, | |
| "loss": 2.9931, | |
| "step": 149000 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 0.0006530446549391069, | |
| "loss": 2.943, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 0.0006501059010413602, | |
| "loss": 2.9528, | |
| "step": 151000 | |
| }, | |
| { | |
| "epoch": 8.17, | |
| "learning_rate": 0.0006471642054480202, | |
| "loss": 2.9563, | |
| "step": 152000 | |
| }, | |
| { | |
| "epoch": 8.23, | |
| "learning_rate": 0.0006442254515502735, | |
| "loss": 2.9565, | |
| "step": 153000 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "learning_rate": 0.0006412837559569335, | |
| "loss": 2.9659, | |
| "step": 154000 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 0.0006383420603635935, | |
| "loss": 2.9644, | |
| "step": 155000 | |
| }, | |
| { | |
| "epoch": 8.39, | |
| "learning_rate": 0.0006354033064658468, | |
| "loss": 2.9698, | |
| "step": 156000 | |
| }, | |
| { | |
| "epoch": 8.44, | |
| "learning_rate": 0.0006324616108725069, | |
| "loss": 2.9681, | |
| "step": 157000 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "learning_rate": 0.000629519915279167, | |
| "loss": 2.9729, | |
| "step": 158000 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 0.0006265811613814202, | |
| "loss": 2.9799, | |
| "step": 159000 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "learning_rate": 0.0006236394657880802, | |
| "loss": 2.9751, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "learning_rate": 0.0006207007118903336, | |
| "loss": 2.9799, | |
| "step": 161000 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "learning_rate": 0.0006177590162969937, | |
| "loss": 2.9774, | |
| "step": 162000 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 0.0006148202623992469, | |
| "loss": 2.9757, | |
| "step": 163000 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "learning_rate": 0.000611878566805907, | |
| "loss": 2.9785, | |
| "step": 164000 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "learning_rate": 0.000608936871212567, | |
| "loss": 2.9794, | |
| "step": 165000 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "learning_rate": 0.0006059951756192269, | |
| "loss": 2.9827, | |
| "step": 166000 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "learning_rate": 0.0006030564217214803, | |
| "loss": 2.985, | |
| "step": 167000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.4100755354779728, | |
| "eval_loss": 3.358936071395874, | |
| "eval_runtime": 152.9763, | |
| "eval_samples_per_second": 378.608, | |
| "eval_steps_per_second": 5.916, | |
| "step": 167373 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 0.0006001147261281403, | |
| "loss": 2.9399, | |
| "step": 168000 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 0.0005971730305348003, | |
| "loss": 2.9191, | |
| "step": 169000 | |
| }, | |
| { | |
| "epoch": 9.14, | |
| "learning_rate": 0.0005942313349414603, | |
| "loss": 2.9268, | |
| "step": 170000 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "learning_rate": 0.000591295522739307, | |
| "loss": 2.9329, | |
| "step": 171000 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "learning_rate": 0.0005883567688415604, | |
| "loss": 2.9352, | |
| "step": 172000 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 0.0005854150732482203, | |
| "loss": 2.9368, | |
| "step": 173000 | |
| }, | |
| { | |
| "epoch": 9.36, | |
| "learning_rate": 0.0005824733776548803, | |
| "loss": 2.9406, | |
| "step": 174000 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "learning_rate": 0.0005795316820615403, | |
| "loss": 2.9468, | |
| "step": 175000 | |
| }, | |
| { | |
| "epoch": 9.46, | |
| "learning_rate": 0.0005765929281637936, | |
| "loss": 2.9437, | |
| "step": 176000 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 0.0005736512325704536, | |
| "loss": 2.9463, | |
| "step": 177000 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "learning_rate": 0.000570712478672707, | |
| "loss": 2.945, | |
| "step": 178000 | |
| }, | |
| { | |
| "epoch": 9.63, | |
| "learning_rate": 0.000567770783079367, | |
| "loss": 2.9485, | |
| "step": 179000 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "learning_rate": 0.0005648290874860269, | |
| "loss": 2.9481, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "learning_rate": 0.000561887391892687, | |
| "loss": 2.9523, | |
| "step": 181000 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 0.0005589486379949404, | |
| "loss": 2.9571, | |
| "step": 182000 | |
| }, | |
| { | |
| "epoch": 9.84, | |
| "learning_rate": 0.0005560069424016003, | |
| "loss": 2.9573, | |
| "step": 183000 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "learning_rate": 0.0005530652468082603, | |
| "loss": 2.9588, | |
| "step": 184000 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "learning_rate": 0.0005501264929105137, | |
| "loss": 2.9555, | |
| "step": 185000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.4095514482235249, | |
| "eval_loss": 3.3776910305023193, | |
| "eval_runtime": 153.0291, | |
| "eval_samples_per_second": 378.477, | |
| "eval_steps_per_second": 5.914, | |
| "step": 185970 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.0005471847973171736, | |
| "loss": 2.9541, | |
| "step": 186000 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 0.0005442431017238336, | |
| "loss": 2.8943, | |
| "step": 187000 | |
| }, | |
| { | |
| "epoch": 10.11, | |
| "learning_rate": 0.000541304347826087, | |
| "loss": 2.8957, | |
| "step": 188000 | |
| }, | |
| { | |
| "epoch": 10.16, | |
| "learning_rate": 0.0005383626522327469, | |
| "loss": 2.9049, | |
| "step": 189000 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "learning_rate": 0.0005354209566394069, | |
| "loss": 2.9062, | |
| "step": 190000 | |
| }, | |
| { | |
| "epoch": 10.27, | |
| "learning_rate": 0.0005324822027416603, | |
| "loss": 2.9085, | |
| "step": 191000 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "learning_rate": 0.0005295405071483203, | |
| "loss": 2.9143, | |
| "step": 192000 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "learning_rate": 0.0005266017532505737, | |
| "loss": 2.9149, | |
| "step": 193000 | |
| }, | |
| { | |
| "epoch": 10.43, | |
| "learning_rate": 0.0005236600576572337, | |
| "loss": 2.9209, | |
| "step": 194000 | |
| }, | |
| { | |
| "epoch": 10.49, | |
| "learning_rate": 0.0005207183620638937, | |
| "loss": 2.9225, | |
| "step": 195000 | |
| }, | |
| { | |
| "epoch": 10.54, | |
| "learning_rate": 0.0005177766664705536, | |
| "loss": 2.925, | |
| "step": 196000 | |
| }, | |
| { | |
| "epoch": 10.59, | |
| "learning_rate": 0.0005148408542684003, | |
| "loss": 2.9207, | |
| "step": 197000 | |
| }, | |
| { | |
| "epoch": 10.65, | |
| "learning_rate": 0.0005118991586750603, | |
| "loss": 2.9256, | |
| "step": 198000 | |
| }, | |
| { | |
| "epoch": 10.7, | |
| "learning_rate": 0.0005089604047773137, | |
| "loss": 2.9319, | |
| "step": 199000 | |
| }, | |
| { | |
| "epoch": 10.75, | |
| "learning_rate": 0.0005060187091839736, | |
| "loss": 2.9318, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 10.81, | |
| "learning_rate": 0.000503079955286227, | |
| "loss": 2.9347, | |
| "step": 201000 | |
| }, | |
| { | |
| "epoch": 10.86, | |
| "learning_rate": 0.000500138259692887, | |
| "loss": 2.9302, | |
| "step": 202000 | |
| }, | |
| { | |
| "epoch": 10.92, | |
| "learning_rate": 0.000497196564099547, | |
| "loss": 2.9316, | |
| "step": 203000 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "learning_rate": 0.0004942578102018003, | |
| "loss": 2.9324, | |
| "step": 204000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.4109965343462238, | |
| "eval_loss": 3.3606457710266113, | |
| "eval_runtime": 153.1523, | |
| "eval_samples_per_second": 378.173, | |
| "eval_steps_per_second": 5.909, | |
| "step": 204567 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 0.0004913190563040537, | |
| "loss": 2.9005, | |
| "step": 205000 | |
| }, | |
| { | |
| "epoch": 11.08, | |
| "learning_rate": 0.0004883773607107136, | |
| "loss": 2.8737, | |
| "step": 206000 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "learning_rate": 0.00048543566511737366, | |
| "loss": 2.8788, | |
| "step": 207000 | |
| }, | |
| { | |
| "epoch": 11.18, | |
| "learning_rate": 0.00048249396952403366, | |
| "loss": 2.8849, | |
| "step": 208000 | |
| }, | |
| { | |
| "epoch": 11.24, | |
| "learning_rate": 0.00047955227393069367, | |
| "loss": 2.8838, | |
| "step": 209000 | |
| }, | |
| { | |
| "epoch": 11.29, | |
| "learning_rate": 0.0004766105783373537, | |
| "loss": 2.8952, | |
| "step": 210000 | |
| }, | |
| { | |
| "epoch": 11.35, | |
| "learning_rate": 0.00047367182443960697, | |
| "loss": 2.8926, | |
| "step": 211000 | |
| }, | |
| { | |
| "epoch": 11.4, | |
| "learning_rate": 0.000470730128846267, | |
| "loss": 2.9012, | |
| "step": 212000 | |
| }, | |
| { | |
| "epoch": 11.45, | |
| "learning_rate": 0.0004677913749485203, | |
| "loss": 2.8991, | |
| "step": 213000 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "learning_rate": 0.0004648496793551804, | |
| "loss": 2.8997, | |
| "step": 214000 | |
| }, | |
| { | |
| "epoch": 11.56, | |
| "learning_rate": 0.00046191092545743367, | |
| "loss": 2.9006, | |
| "step": 215000 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "learning_rate": 0.000458972171559687, | |
| "loss": 2.9041, | |
| "step": 216000 | |
| }, | |
| { | |
| "epoch": 11.67, | |
| "learning_rate": 0.000456030475966347, | |
| "loss": 2.9044, | |
| "step": 217000 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "learning_rate": 0.000453088780373007, | |
| "loss": 2.9056, | |
| "step": 218000 | |
| }, | |
| { | |
| "epoch": 11.78, | |
| "learning_rate": 0.00045015002647526037, | |
| "loss": 2.9076, | |
| "step": 219000 | |
| }, | |
| { | |
| "epoch": 11.83, | |
| "learning_rate": 0.0004472083308819204, | |
| "loss": 2.9088, | |
| "step": 220000 | |
| }, | |
| { | |
| "epoch": 11.88, | |
| "learning_rate": 0.00044426663528858033, | |
| "loss": 2.9049, | |
| "step": 221000 | |
| }, | |
| { | |
| "epoch": 11.94, | |
| "learning_rate": 0.00044132493969524034, | |
| "loss": 2.9106, | |
| "step": 222000 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "learning_rate": 0.00043838324410190035, | |
| "loss": 2.9092, | |
| "step": 223000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.4111943855145803, | |
| "eval_loss": 3.3721723556518555, | |
| "eval_runtime": 153.0663, | |
| "eval_samples_per_second": 378.385, | |
| "eval_steps_per_second": 5.912, | |
| "step": 223164 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 0.0004354415485085603, | |
| "loss": 2.8591, | |
| "step": 224000 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "learning_rate": 0.00043250279461081364, | |
| "loss": 2.8554, | |
| "step": 225000 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "learning_rate": 0.00042956404071306704, | |
| "loss": 2.8627, | |
| "step": 226000 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "learning_rate": 0.00042662234511972705, | |
| "loss": 2.8607, | |
| "step": 227000 | |
| }, | |
| { | |
| "epoch": 12.26, | |
| "learning_rate": 0.000423680649526387, | |
| "loss": 2.8658, | |
| "step": 228000 | |
| }, | |
| { | |
| "epoch": 12.31, | |
| "learning_rate": 0.000420738953933047, | |
| "loss": 2.871, | |
| "step": 229000 | |
| }, | |
| { | |
| "epoch": 12.37, | |
| "learning_rate": 0.000417797258339707, | |
| "loss": 2.8725, | |
| "step": 230000 | |
| }, | |
| { | |
| "epoch": 12.42, | |
| "learning_rate": 0.000414855562746367, | |
| "loss": 2.8749, | |
| "step": 231000 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "learning_rate": 0.00041191680884862037, | |
| "loss": 2.8814, | |
| "step": 232000 | |
| }, | |
| { | |
| "epoch": 12.53, | |
| "learning_rate": 0.0004089751132552804, | |
| "loss": 2.874, | |
| "step": 233000 | |
| }, | |
| { | |
| "epoch": 12.58, | |
| "learning_rate": 0.00040603635935753367, | |
| "loss": 2.8853, | |
| "step": 234000 | |
| }, | |
| { | |
| "epoch": 12.64, | |
| "learning_rate": 0.0004030946637641937, | |
| "loss": 2.8811, | |
| "step": 235000 | |
| }, | |
| { | |
| "epoch": 12.69, | |
| "learning_rate": 0.0004001529681708537, | |
| "loss": 2.8845, | |
| "step": 236000 | |
| }, | |
| { | |
| "epoch": 12.74, | |
| "learning_rate": 0.00039721127257751364, | |
| "loss": 2.8837, | |
| "step": 237000 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 0.00039427251867976703, | |
| "loss": 2.8851, | |
| "step": 238000 | |
| }, | |
| { | |
| "epoch": 12.85, | |
| "learning_rate": 0.00039133082308642704, | |
| "loss": 2.8924, | |
| "step": 239000 | |
| }, | |
| { | |
| "epoch": 12.91, | |
| "learning_rate": 0.00038838912749308705, | |
| "loss": 2.8865, | |
| "step": 240000 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "learning_rate": 0.00038545331529093367, | |
| "loss": 2.89, | |
| "step": 241000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.41136455767398505, | |
| "eval_loss": 3.3736608028411865, | |
| "eval_runtime": 153.2544, | |
| "eval_samples_per_second": 377.921, | |
| "eval_steps_per_second": 5.905, | |
| "step": 241761 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 0.0003825116196975937, | |
| "loss": 2.875, | |
| "step": 242000 | |
| }, | |
| { | |
| "epoch": 13.07, | |
| "learning_rate": 0.0003795699241042537, | |
| "loss": 2.8309, | |
| "step": 243000 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "learning_rate": 0.0003766282285109137, | |
| "loss": 2.8374, | |
| "step": 244000 | |
| }, | |
| { | |
| "epoch": 13.17, | |
| "learning_rate": 0.00037368947461316704, | |
| "loss": 2.843, | |
| "step": 245000 | |
| }, | |
| { | |
| "epoch": 13.23, | |
| "learning_rate": 0.00037074777901982705, | |
| "loss": 2.843, | |
| "step": 246000 | |
| }, | |
| { | |
| "epoch": 13.28, | |
| "learning_rate": 0.0003678090251220804, | |
| "loss": 2.8496, | |
| "step": 247000 | |
| }, | |
| { | |
| "epoch": 13.34, | |
| "learning_rate": 0.00036486732952874034, | |
| "loss": 2.8547, | |
| "step": 248000 | |
| }, | |
| { | |
| "epoch": 13.39, | |
| "learning_rate": 0.00036192563393540035, | |
| "loss": 2.8517, | |
| "step": 249000 | |
| }, | |
| { | |
| "epoch": 13.44, | |
| "learning_rate": 0.00035898688003765375, | |
| "loss": 2.8575, | |
| "step": 250000 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "learning_rate": 0.0003560451844443137, | |
| "loss": 2.8576, | |
| "step": 251000 | |
| }, | |
| { | |
| "epoch": 13.55, | |
| "learning_rate": 0.00035310643054656704, | |
| "loss": 2.8592, | |
| "step": 252000 | |
| }, | |
| { | |
| "epoch": 13.6, | |
| "learning_rate": 0.00035016473495322705, | |
| "loss": 2.8627, | |
| "step": 253000 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "learning_rate": 0.00034722598105548034, | |
| "loss": 2.862, | |
| "step": 254000 | |
| }, | |
| { | |
| "epoch": 13.71, | |
| "learning_rate": 0.0003442842854621404, | |
| "loss": 2.8622, | |
| "step": 255000 | |
| }, | |
| { | |
| "epoch": 13.77, | |
| "learning_rate": 0.00034134553156439374, | |
| "loss": 2.8674, | |
| "step": 256000 | |
| }, | |
| { | |
| "epoch": 13.82, | |
| "learning_rate": 0.00033840383597105375, | |
| "loss": 2.8654, | |
| "step": 257000 | |
| }, | |
| { | |
| "epoch": 13.87, | |
| "learning_rate": 0.0003354621403777137, | |
| "loss": 2.871, | |
| "step": 258000 | |
| }, | |
| { | |
| "epoch": 13.93, | |
| "learning_rate": 0.00033252338647996705, | |
| "loss": 2.8669, | |
| "step": 259000 | |
| }, | |
| { | |
| "epoch": 13.98, | |
| "learning_rate": 0.00032958169088662706, | |
| "loss": 2.8651, | |
| "step": 260000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.4109751032689044, | |
| "eval_loss": 3.3934197425842285, | |
| "eval_runtime": 153.0719, | |
| "eval_samples_per_second": 378.371, | |
| "eval_steps_per_second": 5.912, | |
| "step": 260358 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "learning_rate": 0.00032663999529328707, | |
| "loss": 2.8337, | |
| "step": 261000 | |
| }, | |
| { | |
| "epoch": 14.09, | |
| "learning_rate": 0.0003237012413955404, | |
| "loss": 2.8203, | |
| "step": 262000 | |
| }, | |
| { | |
| "epoch": 14.14, | |
| "learning_rate": 0.0003207595458022004, | |
| "loss": 2.8229, | |
| "step": 263000 | |
| }, | |
| { | |
| "epoch": 14.2, | |
| "learning_rate": 0.00031782079190445376, | |
| "loss": 2.8286, | |
| "step": 264000 | |
| }, | |
| { | |
| "epoch": 14.25, | |
| "learning_rate": 0.0003148790963111137, | |
| "loss": 2.8263, | |
| "step": 265000 | |
| }, | |
| { | |
| "epoch": 14.3, | |
| "learning_rate": 0.0003119374007177737, | |
| "loss": 2.8291, | |
| "step": 266000 | |
| }, | |
| { | |
| "epoch": 14.36, | |
| "learning_rate": 0.0003089986468200271, | |
| "loss": 2.8315, | |
| "step": 267000 | |
| }, | |
| { | |
| "epoch": 14.41, | |
| "learning_rate": 0.00030605695122668707, | |
| "loss": 2.84, | |
| "step": 268000 | |
| }, | |
| { | |
| "epoch": 14.46, | |
| "learning_rate": 0.0003031152556333471, | |
| "loss": 2.8378, | |
| "step": 269000 | |
| }, | |
| { | |
| "epoch": 14.52, | |
| "learning_rate": 0.0003001735600400071, | |
| "loss": 2.8364, | |
| "step": 270000 | |
| }, | |
| { | |
| "epoch": 14.57, | |
| "learning_rate": 0.0002972348061422604, | |
| "loss": 2.839, | |
| "step": 271000 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "learning_rate": 0.0002942931105489204, | |
| "loss": 2.8425, | |
| "step": 272000 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "learning_rate": 0.00029135141495558045, | |
| "loss": 2.8419, | |
| "step": 273000 | |
| }, | |
| { | |
| "epoch": 14.73, | |
| "learning_rate": 0.00028841266105783373, | |
| "loss": 2.8472, | |
| "step": 274000 | |
| }, | |
| { | |
| "epoch": 14.79, | |
| "learning_rate": 0.00028547096546449374, | |
| "loss": 2.846, | |
| "step": 275000 | |
| }, | |
| { | |
| "epoch": 14.84, | |
| "learning_rate": 0.0002825351532623404, | |
| "loss": 2.8429, | |
| "step": 276000 | |
| }, | |
| { | |
| "epoch": 14.89, | |
| "learning_rate": 0.0002795934576690004, | |
| "loss": 2.8454, | |
| "step": 277000 | |
| }, | |
| { | |
| "epoch": 14.95, | |
| "learning_rate": 0.0002766517620756604, | |
| "loss": 2.8499, | |
| "step": 278000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.41160291962486073, | |
| "eval_loss": 3.3911032676696777, | |
| "eval_runtime": 153.0958, | |
| "eval_samples_per_second": 378.312, | |
| "eval_steps_per_second": 5.911, | |
| "step": 278955 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 0.0002737130081779138, | |
| "loss": 2.8485, | |
| "step": 279000 | |
| }, | |
| { | |
| "epoch": 15.06, | |
| "learning_rate": 0.0002707713125845738, | |
| "loss": 2.798, | |
| "step": 280000 | |
| }, | |
| { | |
| "epoch": 15.11, | |
| "learning_rate": 0.00026782961699123374, | |
| "loss": 2.8027, | |
| "step": 281000 | |
| }, | |
| { | |
| "epoch": 15.16, | |
| "learning_rate": 0.00026488792139789375, | |
| "loss": 2.805, | |
| "step": 282000 | |
| }, | |
| { | |
| "epoch": 15.22, | |
| "learning_rate": 0.00026194622580455375, | |
| "loss": 2.8109, | |
| "step": 283000 | |
| }, | |
| { | |
| "epoch": 15.27, | |
| "learning_rate": 0.00025900747190680704, | |
| "loss": 2.8134, | |
| "step": 284000 | |
| }, | |
| { | |
| "epoch": 15.33, | |
| "learning_rate": 0.0002560657763134671, | |
| "loss": 2.8149, | |
| "step": 285000 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 0.00025312702241572045, | |
| "loss": 2.8167, | |
| "step": 286000 | |
| }, | |
| { | |
| "epoch": 15.43, | |
| "learning_rate": 0.00025018532682238046, | |
| "loss": 2.8182, | |
| "step": 287000 | |
| }, | |
| { | |
| "epoch": 15.49, | |
| "learning_rate": 0.0002472436312290404, | |
| "loss": 2.8201, | |
| "step": 288000 | |
| }, | |
| { | |
| "epoch": 15.54, | |
| "learning_rate": 0.00024430487733129375, | |
| "loss": 2.8221, | |
| "step": 289000 | |
| }, | |
| { | |
| "epoch": 15.59, | |
| "learning_rate": 0.00024136318173795376, | |
| "loss": 2.8212, | |
| "step": 290000 | |
| }, | |
| { | |
| "epoch": 15.65, | |
| "learning_rate": 0.0002384244278402071, | |
| "loss": 2.8247, | |
| "step": 291000 | |
| }, | |
| { | |
| "epoch": 15.7, | |
| "learning_rate": 0.00023548273224686708, | |
| "loss": 2.8229, | |
| "step": 292000 | |
| }, | |
| { | |
| "epoch": 15.76, | |
| "learning_rate": 0.0002325410366535271, | |
| "loss": 2.8239, | |
| "step": 293000 | |
| }, | |
| { | |
| "epoch": 15.81, | |
| "learning_rate": 0.0002295993410601871, | |
| "loss": 2.827, | |
| "step": 294000 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "learning_rate": 0.00022666058716244042, | |
| "loss": 2.8271, | |
| "step": 295000 | |
| }, | |
| { | |
| "epoch": 15.92, | |
| "learning_rate": 0.00022371889156910045, | |
| "loss": 2.8278, | |
| "step": 296000 | |
| }, | |
| { | |
| "epoch": 15.97, | |
| "learning_rate": 0.00022077719597576043, | |
| "loss": 2.8292, | |
| "step": 297000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.41136260939422875, | |
| "eval_loss": 3.394244909286499, | |
| "eval_runtime": 153.2088, | |
| "eval_samples_per_second": 378.033, | |
| "eval_steps_per_second": 5.907, | |
| "step": 297552 | |
| }, | |
| { | |
| "epoch": 16.02, | |
| "learning_rate": 0.00021783844207801375, | |
| "loss": 2.8099, | |
| "step": 298000 | |
| }, | |
| { | |
| "epoch": 16.08, | |
| "learning_rate": 0.00021489674648467378, | |
| "loss": 2.7837, | |
| "step": 299000 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "learning_rate": 0.0002119579925869271, | |
| "loss": 2.7927, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 16.19, | |
| "learning_rate": 0.0002090162969935871, | |
| "loss": 2.7916, | |
| "step": 301000 | |
| }, | |
| { | |
| "epoch": 16.24, | |
| "learning_rate": 0.0002060746014002471, | |
| "loss": 2.7954, | |
| "step": 302000 | |
| }, | |
| { | |
| "epoch": 16.29, | |
| "learning_rate": 0.00020313584750250046, | |
| "loss": 2.7948, | |
| "step": 303000 | |
| }, | |
| { | |
| "epoch": 16.35, | |
| "learning_rate": 0.00020019415190916044, | |
| "loss": 2.7956, | |
| "step": 304000 | |
| }, | |
| { | |
| "epoch": 16.4, | |
| "learning_rate": 0.00019725245631582044, | |
| "loss": 2.7982, | |
| "step": 305000 | |
| }, | |
| { | |
| "epoch": 16.45, | |
| "learning_rate": 0.0001943137024180738, | |
| "loss": 2.8014, | |
| "step": 306000 | |
| }, | |
| { | |
| "epoch": 16.51, | |
| "learning_rate": 0.00019137200682473377, | |
| "loss": 2.8014, | |
| "step": 307000 | |
| }, | |
| { | |
| "epoch": 16.56, | |
| "learning_rate": 0.00018843031123139378, | |
| "loss": 2.8045, | |
| "step": 308000 | |
| }, | |
| { | |
| "epoch": 16.62, | |
| "learning_rate": 0.00018549155733364712, | |
| "loss": 2.8021, | |
| "step": 309000 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "learning_rate": 0.0001825498617403071, | |
| "loss": 2.8044, | |
| "step": 310000 | |
| }, | |
| { | |
| "epoch": 16.72, | |
| "learning_rate": 0.00017961110784256047, | |
| "loss": 2.8052, | |
| "step": 311000 | |
| }, | |
| { | |
| "epoch": 16.78, | |
| "learning_rate": 0.00017666941224922045, | |
| "loss": 2.8076, | |
| "step": 312000 | |
| }, | |
| { | |
| "epoch": 16.83, | |
| "learning_rate": 0.0001737306583514738, | |
| "loss": 2.8107, | |
| "step": 313000 | |
| }, | |
| { | |
| "epoch": 16.88, | |
| "learning_rate": 0.00017079190445372713, | |
| "loss": 2.8091, | |
| "step": 314000 | |
| }, | |
| { | |
| "epoch": 16.94, | |
| "learning_rate": 0.00016785020886038712, | |
| "loss": 2.8081, | |
| "step": 315000 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "learning_rate": 0.00016491145496264046, | |
| "loss": 2.8105, | |
| "step": 316000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.4112535057278753, | |
| "eval_loss": 3.4117093086242676, | |
| "eval_runtime": 152.9687, | |
| "eval_samples_per_second": 378.627, | |
| "eval_steps_per_second": 5.916, | |
| "step": 316149 | |
| }, | |
| { | |
| "epoch": 17.05, | |
| "learning_rate": 0.00016196975936930047, | |
| "loss": 2.7783, | |
| "step": 317000 | |
| }, | |
| { | |
| "epoch": 17.1, | |
| "learning_rate": 0.00015902806377596047, | |
| "loss": 2.7714, | |
| "step": 318000 | |
| }, | |
| { | |
| "epoch": 17.15, | |
| "learning_rate": 0.00015608636818262046, | |
| "loss": 2.7767, | |
| "step": 319000 | |
| }, | |
| { | |
| "epoch": 17.21, | |
| "learning_rate": 0.00015314467258928046, | |
| "loss": 2.7784, | |
| "step": 320000 | |
| }, | |
| { | |
| "epoch": 17.26, | |
| "learning_rate": 0.0001502059186915338, | |
| "loss": 2.774, | |
| "step": 321000 | |
| }, | |
| { | |
| "epoch": 17.31, | |
| "learning_rate": 0.00014726716479378712, | |
| "loss": 2.781, | |
| "step": 322000 | |
| }, | |
| { | |
| "epoch": 17.37, | |
| "learning_rate": 0.00014432546920044716, | |
| "loss": 2.7785, | |
| "step": 323000 | |
| }, | |
| { | |
| "epoch": 17.42, | |
| "learning_rate": 0.00014138671530270047, | |
| "loss": 2.7872, | |
| "step": 324000 | |
| }, | |
| { | |
| "epoch": 17.48, | |
| "learning_rate": 0.00013844501970936045, | |
| "loss": 2.7856, | |
| "step": 325000 | |
| }, | |
| { | |
| "epoch": 17.53, | |
| "learning_rate": 0.0001355033241160205, | |
| "loss": 2.7862, | |
| "step": 326000 | |
| }, | |
| { | |
| "epoch": 17.58, | |
| "learning_rate": 0.00013256162852268047, | |
| "loss": 2.787, | |
| "step": 327000 | |
| }, | |
| { | |
| "epoch": 17.64, | |
| "learning_rate": 0.0001296228746249338, | |
| "loss": 2.7876, | |
| "step": 328000 | |
| }, | |
| { | |
| "epoch": 17.69, | |
| "learning_rate": 0.00012668117903159382, | |
| "loss": 2.7864, | |
| "step": 329000 | |
| }, | |
| { | |
| "epoch": 17.74, | |
| "learning_rate": 0.00012374242513384713, | |
| "loss": 2.7893, | |
| "step": 330000 | |
| }, | |
| { | |
| "epoch": 17.8, | |
| "learning_rate": 0.00012080367123610049, | |
| "loss": 2.7895, | |
| "step": 331000 | |
| }, | |
| { | |
| "epoch": 17.85, | |
| "learning_rate": 0.0001178619756427605, | |
| "loss": 2.7923, | |
| "step": 332000 | |
| }, | |
| { | |
| "epoch": 17.91, | |
| "learning_rate": 0.00011492028004942048, | |
| "loss": 2.7886, | |
| "step": 333000 | |
| }, | |
| { | |
| "epoch": 17.96, | |
| "learning_rate": 0.00011198152615167382, | |
| "loss": 2.7877, | |
| "step": 334000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.41155508599774027, | |
| "eval_loss": 3.40728497505188, | |
| "eval_runtime": 153.0938, | |
| "eval_samples_per_second": 378.317, | |
| "eval_steps_per_second": 5.911, | |
| "step": 334746 | |
| }, | |
| { | |
| "epoch": 18.01, | |
| "learning_rate": 0.00010903983055833383, | |
| "loss": 2.7816, | |
| "step": 335000 | |
| }, | |
| { | |
| "epoch": 18.07, | |
| "learning_rate": 0.00010609813496499382, | |
| "loss": 2.759, | |
| "step": 336000 | |
| }, | |
| { | |
| "epoch": 18.12, | |
| "learning_rate": 0.00010315643937165382, | |
| "loss": 2.7611, | |
| "step": 337000 | |
| }, | |
| { | |
| "epoch": 18.17, | |
| "learning_rate": 0.00010021474377831383, | |
| "loss": 2.769, | |
| "step": 338000 | |
| }, | |
| { | |
| "epoch": 18.23, | |
| "learning_rate": 9.727598988056716e-05, | |
| "loss": 2.7636, | |
| "step": 339000 | |
| }, | |
| { | |
| "epoch": 18.28, | |
| "learning_rate": 9.433429428722716e-05, | |
| "loss": 2.7624, | |
| "step": 340000 | |
| }, | |
| { | |
| "epoch": 18.34, | |
| "learning_rate": 9.139259869388716e-05, | |
| "loss": 2.7664, | |
| "step": 341000 | |
| }, | |
| { | |
| "epoch": 18.39, | |
| "learning_rate": 8.845384479614049e-05, | |
| "loss": 2.7625, | |
| "step": 342000 | |
| }, | |
| { | |
| "epoch": 18.44, | |
| "learning_rate": 8.55121492028005e-05, | |
| "loss": 2.7686, | |
| "step": 343000 | |
| }, | |
| { | |
| "epoch": 18.5, | |
| "learning_rate": 8.25704536094605e-05, | |
| "loss": 2.7686, | |
| "step": 344000 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "learning_rate": 7.963169971171383e-05, | |
| "loss": 2.7683, | |
| "step": 345000 | |
| }, | |
| { | |
| "epoch": 18.61, | |
| "learning_rate": 7.669000411837383e-05, | |
| "loss": 2.7683, | |
| "step": 346000 | |
| }, | |
| { | |
| "epoch": 18.66, | |
| "learning_rate": 7.374830852503384e-05, | |
| "loss": 2.7696, | |
| "step": 347000 | |
| }, | |
| { | |
| "epoch": 18.71, | |
| "learning_rate": 7.080955462728716e-05, | |
| "loss": 2.7716, | |
| "step": 348000 | |
| }, | |
| { | |
| "epoch": 18.77, | |
| "learning_rate": 6.786785903394717e-05, | |
| "loss": 2.7684, | |
| "step": 349000 | |
| }, | |
| { | |
| "epoch": 18.82, | |
| "learning_rate": 6.493204683179385e-05, | |
| "loss": 2.7767, | |
| "step": 350000 | |
| }, | |
| { | |
| "epoch": 18.87, | |
| "learning_rate": 6.199035123845384e-05, | |
| "loss": 2.7682, | |
| "step": 351000 | |
| }, | |
| { | |
| "epoch": 18.93, | |
| "learning_rate": 5.9048655645113845e-05, | |
| "loss": 2.7754, | |
| "step": 352000 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "learning_rate": 5.610696005177384e-05, | |
| "loss": 2.773, | |
| "step": 353000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.411468219593433, | |
| "eval_loss": 3.4168522357940674, | |
| "eval_runtime": 153.1617, | |
| "eval_samples_per_second": 378.149, | |
| "eval_steps_per_second": 5.909, | |
| "step": 353343 | |
| }, | |
| { | |
| "epoch": 19.04, | |
| "learning_rate": 5.316526445843385e-05, | |
| "loss": 2.756, | |
| "step": 354000 | |
| }, | |
| { | |
| "epoch": 19.09, | |
| "learning_rate": 5.022356886509384e-05, | |
| "loss": 2.7486, | |
| "step": 355000 | |
| }, | |
| { | |
| "epoch": 19.14, | |
| "learning_rate": 4.728481496734718e-05, | |
| "loss": 2.7548, | |
| "step": 356000 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "learning_rate": 4.434900276519386e-05, | |
| "loss": 2.7531, | |
| "step": 357000 | |
| }, | |
| { | |
| "epoch": 19.25, | |
| "learning_rate": 4.1407307171853856e-05, | |
| "loss": 2.7497, | |
| "step": 358000 | |
| }, | |
| { | |
| "epoch": 19.3, | |
| "learning_rate": 3.846561157851386e-05, | |
| "loss": 2.7474, | |
| "step": 359000 | |
| }, | |
| { | |
| "epoch": 19.36, | |
| "learning_rate": 3.552391598517386e-05, | |
| "loss": 2.7525, | |
| "step": 360000 | |
| }, | |
| { | |
| "epoch": 19.41, | |
| "learning_rate": 3.258222039183385e-05, | |
| "loss": 2.757, | |
| "step": 361000 | |
| }, | |
| { | |
| "epoch": 19.47, | |
| "learning_rate": 2.9640524798493854e-05, | |
| "loss": 2.7523, | |
| "step": 362000 | |
| }, | |
| { | |
| "epoch": 19.52, | |
| "learning_rate": 2.669882920515385e-05, | |
| "loss": 2.7568, | |
| "step": 363000 | |
| }, | |
| { | |
| "epoch": 19.57, | |
| "learning_rate": 2.3760075307407187e-05, | |
| "loss": 2.7589, | |
| "step": 364000 | |
| }, | |
| { | |
| "epoch": 19.63, | |
| "learning_rate": 2.081837971406719e-05, | |
| "loss": 2.7527, | |
| "step": 365000 | |
| }, | |
| { | |
| "epoch": 19.68, | |
| "learning_rate": 1.7879625816320528e-05, | |
| "loss": 2.7531, | |
| "step": 366000 | |
| }, | |
| { | |
| "epoch": 19.73, | |
| "learning_rate": 1.4937930222980527e-05, | |
| "loss": 2.7539, | |
| "step": 367000 | |
| }, | |
| { | |
| "epoch": 19.79, | |
| "learning_rate": 1.1999176325233865e-05, | |
| "loss": 2.7548, | |
| "step": 368000 | |
| }, | |
| { | |
| "epoch": 19.84, | |
| "learning_rate": 9.057480731893863e-06, | |
| "loss": 2.7532, | |
| "step": 369000 | |
| }, | |
| { | |
| "epoch": 19.9, | |
| "learning_rate": 6.115785138553862e-06, | |
| "loss": 2.755, | |
| "step": 370000 | |
| }, | |
| { | |
| "epoch": 19.95, | |
| "learning_rate": 3.1770312408072013e-06, | |
| "loss": 2.7535, | |
| "step": 371000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.4116416836738053, | |
| "eval_loss": 3.419250011444092, | |
| "eval_runtime": 153.0852, | |
| "eval_samples_per_second": 378.338, | |
| "eval_steps_per_second": 5.912, | |
| "step": 371940 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 371940, | |
| "total_flos": 1.56702845389824e+18, | |
| "train_loss": 3.025299089003771, | |
| "train_runtime": 80778.7854, | |
| "train_samples_per_second": 147.334, | |
| "train_steps_per_second": 4.604 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 371940, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 5000, | |
| "total_flos": 1.56702845389824e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |