|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 20.0, |
|
"eval_steps": 500, |
|
"global_step": 371900, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.05, |
|
"grad_norm": 1.1065584421157837, |
|
"learning_rate": 3.125e-06, |
|
"loss": 7.5986, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"grad_norm": 0.6919564604759216, |
|
"learning_rate": 6.25e-06, |
|
"loss": 5.8259, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"grad_norm": 0.7977421283721924, |
|
"learning_rate": 9.375000000000001e-06, |
|
"loss": 5.3849, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"grad_norm": 1.386985421180725, |
|
"learning_rate": 1.25e-05, |
|
"loss": 5.1627, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"grad_norm": 0.9054310917854309, |
|
"learning_rate": 1.5625e-05, |
|
"loss": 5.005, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"grad_norm": 1.0550600290298462, |
|
"learning_rate": 1.8750000000000002e-05, |
|
"loss": 4.8627, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"grad_norm": 1.0669718980789185, |
|
"learning_rate": 2.1875e-05, |
|
"loss": 4.7454, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"grad_norm": 1.0990982055664062, |
|
"learning_rate": 2.5e-05, |
|
"loss": 4.6486, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"grad_norm": 1.0567301511764526, |
|
"learning_rate": 2.8125000000000003e-05, |
|
"loss": 4.5585, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"grad_norm": 1.1454998254776, |
|
"learning_rate": 3.125e-05, |
|
"loss": 4.4828, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 0.59, |
|
"grad_norm": 1.0922003984451294, |
|
"learning_rate": 3.4375e-05, |
|
"loss": 4.4099, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 0.65, |
|
"grad_norm": 1.0760829448699951, |
|
"learning_rate": 3.7500000000000003e-05, |
|
"loss": 4.3435, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"grad_norm": 1.0779520273208618, |
|
"learning_rate": 4.0621875e-05, |
|
"loss": 4.2871, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"grad_norm": 1.0053024291992188, |
|
"learning_rate": 4.374375e-05, |
|
"loss": 4.2362, |
|
"step": 14000 |
|
}, |
|
{ |
|
"epoch": 0.81, |
|
"grad_norm": 1.0250402688980103, |
|
"learning_rate": 4.686875e-05, |
|
"loss": 4.1841, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"grad_norm": 1.025078296661377, |
|
"learning_rate": 4.999375e-05, |
|
"loss": 4.1362, |
|
"step": 16000 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"grad_norm": 1.020418643951416, |
|
"learning_rate": 5.311875000000001e-05, |
|
"loss": 4.091, |
|
"step": 17000 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"grad_norm": 1.023269772529602, |
|
"learning_rate": 5.6240625e-05, |
|
"loss": 4.0524, |
|
"step": 18000 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.30844237989493534, |
|
"eval_loss": 4.28109073638916, |
|
"eval_runtime": 151.382, |
|
"eval_samples_per_second": 382.595, |
|
"eval_steps_per_second": 5.978, |
|
"step": 18595 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"grad_norm": 1.0309756994247437, |
|
"learning_rate": 5.9365625e-05, |
|
"loss": 4.015, |
|
"step": 19000 |
|
}, |
|
{ |
|
"epoch": 1.08, |
|
"grad_norm": 0.9628090262413025, |
|
"learning_rate": 6.24875e-05, |
|
"loss": 3.969, |
|
"step": 20000 |
|
}, |
|
{ |
|
"epoch": 1.13, |
|
"grad_norm": 1.0119906663894653, |
|
"learning_rate": 6.560937500000001e-05, |
|
"loss": 3.933, |
|
"step": 21000 |
|
}, |
|
{ |
|
"epoch": 1.18, |
|
"grad_norm": 0.9792044758796692, |
|
"learning_rate": 6.8734375e-05, |
|
"loss": 3.8956, |
|
"step": 22000 |
|
}, |
|
{ |
|
"epoch": 1.24, |
|
"grad_norm": 1.0101176500320435, |
|
"learning_rate": 7.185312499999999e-05, |
|
"loss": 3.8586, |
|
"step": 23000 |
|
}, |
|
{ |
|
"epoch": 1.29, |
|
"grad_norm": 0.9678108096122742, |
|
"learning_rate": 7.4978125e-05, |
|
"loss": 3.8302, |
|
"step": 24000 |
|
}, |
|
{ |
|
"epoch": 1.34, |
|
"grad_norm": 0.9615768194198608, |
|
"learning_rate": 7.8103125e-05, |
|
"loss": 3.801, |
|
"step": 25000 |
|
}, |
|
{ |
|
"epoch": 1.4, |
|
"grad_norm": 0.9395524859428406, |
|
"learning_rate": 8.1221875e-05, |
|
"loss": 3.7804, |
|
"step": 26000 |
|
}, |
|
{ |
|
"epoch": 1.45, |
|
"grad_norm": 0.8997763395309448, |
|
"learning_rate": 8.434687500000001e-05, |
|
"loss": 3.7585, |
|
"step": 27000 |
|
}, |
|
{ |
|
"epoch": 1.51, |
|
"grad_norm": 0.9433351159095764, |
|
"learning_rate": 8.746875e-05, |
|
"loss": 3.732, |
|
"step": 28000 |
|
}, |
|
{ |
|
"epoch": 1.56, |
|
"grad_norm": 0.8686715960502625, |
|
"learning_rate": 9.059375e-05, |
|
"loss": 3.7131, |
|
"step": 29000 |
|
}, |
|
{ |
|
"epoch": 1.61, |
|
"grad_norm": 0.9142166376113892, |
|
"learning_rate": 9.3715625e-05, |
|
"loss": 3.6941, |
|
"step": 30000 |
|
}, |
|
{ |
|
"epoch": 1.67, |
|
"grad_norm": 0.9287761449813843, |
|
"learning_rate": 9.684062500000001e-05, |
|
"loss": 3.6698, |
|
"step": 31000 |
|
}, |
|
{ |
|
"epoch": 1.72, |
|
"grad_norm": 0.8525364995002747, |
|
"learning_rate": 9.99625e-05, |
|
"loss": 3.6574, |
|
"step": 32000 |
|
}, |
|
{ |
|
"epoch": 1.77, |
|
"grad_norm": 0.8378422260284424, |
|
"learning_rate": 9.970932627243307e-05, |
|
"loss": 3.6384, |
|
"step": 33000 |
|
}, |
|
{ |
|
"epoch": 1.83, |
|
"grad_norm": 0.8161050081253052, |
|
"learning_rate": 9.941512209473375e-05, |
|
"loss": 3.6184, |
|
"step": 34000 |
|
}, |
|
{ |
|
"epoch": 1.88, |
|
"grad_norm": 0.8268817067146301, |
|
"learning_rate": 9.912121212121213e-05, |
|
"loss": 3.5973, |
|
"step": 35000 |
|
}, |
|
{ |
|
"epoch": 1.94, |
|
"grad_norm": 0.7921673059463501, |
|
"learning_rate": 9.882700794351281e-05, |
|
"loss": 3.5878, |
|
"step": 36000 |
|
}, |
|
{ |
|
"epoch": 1.99, |
|
"grad_norm": 0.7795915007591248, |
|
"learning_rate": 9.853309796999118e-05, |
|
"loss": 3.5716, |
|
"step": 37000 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.36151083317444777, |
|
"eval_loss": 3.7749860286712646, |
|
"eval_runtime": 153.0787, |
|
"eval_samples_per_second": 378.354, |
|
"eval_steps_per_second": 5.912, |
|
"step": 37190 |
|
}, |
|
{ |
|
"epoch": 2.04, |
|
"grad_norm": 0.816111147403717, |
|
"learning_rate": 9.823889379229185e-05, |
|
"loss": 3.5338, |
|
"step": 38000 |
|
}, |
|
{ |
|
"epoch": 2.1, |
|
"grad_norm": 0.8356756567955017, |
|
"learning_rate": 9.794498381877023e-05, |
|
"loss": 3.5187, |
|
"step": 39000 |
|
}, |
|
{ |
|
"epoch": 2.15, |
|
"grad_norm": 0.8443105220794678, |
|
"learning_rate": 9.765077964107092e-05, |
|
"loss": 3.5061, |
|
"step": 40000 |
|
}, |
|
{ |
|
"epoch": 2.2, |
|
"grad_norm": 0.7713498473167419, |
|
"learning_rate": 9.735686966754929e-05, |
|
"loss": 3.4944, |
|
"step": 41000 |
|
}, |
|
{ |
|
"epoch": 2.26, |
|
"grad_norm": 0.795734167098999, |
|
"learning_rate": 9.706266548984995e-05, |
|
"loss": 3.496, |
|
"step": 42000 |
|
}, |
|
{ |
|
"epoch": 2.31, |
|
"grad_norm": 0.8273535370826721, |
|
"learning_rate": 9.676875551632834e-05, |
|
"loss": 3.4833, |
|
"step": 43000 |
|
}, |
|
{ |
|
"epoch": 2.37, |
|
"grad_norm": 0.7914260625839233, |
|
"learning_rate": 9.647455133862902e-05, |
|
"loss": 3.4718, |
|
"step": 44000 |
|
}, |
|
{ |
|
"epoch": 2.42, |
|
"grad_norm": 0.782301664352417, |
|
"learning_rate": 9.618064136510739e-05, |
|
"loss": 3.46, |
|
"step": 45000 |
|
}, |
|
{ |
|
"epoch": 2.47, |
|
"grad_norm": 0.7781464457511902, |
|
"learning_rate": 9.588643718740806e-05, |
|
"loss": 3.4537, |
|
"step": 46000 |
|
}, |
|
{ |
|
"epoch": 2.53, |
|
"grad_norm": 0.774398148059845, |
|
"learning_rate": 9.559223300970874e-05, |
|
"loss": 3.4478, |
|
"step": 47000 |
|
}, |
|
{ |
|
"epoch": 2.58, |
|
"grad_norm": 0.7705032229423523, |
|
"learning_rate": 9.529832303618712e-05, |
|
"loss": 3.4398, |
|
"step": 48000 |
|
}, |
|
{ |
|
"epoch": 2.64, |
|
"grad_norm": 0.773110568523407, |
|
"learning_rate": 9.50041188584878e-05, |
|
"loss": 3.4366, |
|
"step": 49000 |
|
}, |
|
{ |
|
"epoch": 2.69, |
|
"grad_norm": 0.7716487646102905, |
|
"learning_rate": 9.471050308914386e-05, |
|
"loss": 3.4256, |
|
"step": 50000 |
|
}, |
|
{ |
|
"epoch": 2.74, |
|
"grad_norm": 0.7782188653945923, |
|
"learning_rate": 9.441629891144455e-05, |
|
"loss": 3.417, |
|
"step": 51000 |
|
}, |
|
{ |
|
"epoch": 2.8, |
|
"grad_norm": 0.7639435529708862, |
|
"learning_rate": 9.412209473374523e-05, |
|
"loss": 3.4123, |
|
"step": 52000 |
|
}, |
|
{ |
|
"epoch": 2.85, |
|
"grad_norm": 0.7292437553405762, |
|
"learning_rate": 9.382789055604591e-05, |
|
"loss": 3.405, |
|
"step": 53000 |
|
}, |
|
{ |
|
"epoch": 2.9, |
|
"grad_norm": 0.7682722210884094, |
|
"learning_rate": 9.353368637834658e-05, |
|
"loss": 3.399, |
|
"step": 54000 |
|
}, |
|
{ |
|
"epoch": 2.96, |
|
"grad_norm": 0.7343226671218872, |
|
"learning_rate": 9.323977640482495e-05, |
|
"loss": 3.3913, |
|
"step": 55000 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.3779394670823355, |
|
"eval_loss": 3.600841760635376, |
|
"eval_runtime": 153.5925, |
|
"eval_samples_per_second": 377.089, |
|
"eval_steps_per_second": 5.892, |
|
"step": 55785 |
|
}, |
|
{ |
|
"epoch": 3.01, |
|
"grad_norm": 0.7667288184165955, |
|
"learning_rate": 9.294557222712563e-05, |
|
"loss": 3.3741, |
|
"step": 56000 |
|
}, |
|
{ |
|
"epoch": 3.07, |
|
"grad_norm": 0.791042685508728, |
|
"learning_rate": 9.265166225360401e-05, |
|
"loss": 3.3318, |
|
"step": 57000 |
|
}, |
|
{ |
|
"epoch": 3.12, |
|
"grad_norm": 0.7432328462600708, |
|
"learning_rate": 9.235745807590468e-05, |
|
"loss": 3.3377, |
|
"step": 58000 |
|
}, |
|
{ |
|
"epoch": 3.17, |
|
"grad_norm": 0.7701253294944763, |
|
"learning_rate": 9.206325389820536e-05, |
|
"loss": 3.3324, |
|
"step": 59000 |
|
}, |
|
{ |
|
"epoch": 3.23, |
|
"grad_norm": 0.7645742297172546, |
|
"learning_rate": 9.176963812886143e-05, |
|
"loss": 3.327, |
|
"step": 60000 |
|
}, |
|
{ |
|
"epoch": 3.28, |
|
"grad_norm": 0.7636393904685974, |
|
"learning_rate": 9.147543395116211e-05, |
|
"loss": 3.3306, |
|
"step": 61000 |
|
}, |
|
{ |
|
"epoch": 3.33, |
|
"grad_norm": 0.7321722507476807, |
|
"learning_rate": 9.118122977346278e-05, |
|
"loss": 3.3303, |
|
"step": 62000 |
|
}, |
|
{ |
|
"epoch": 3.39, |
|
"grad_norm": 0.790723443031311, |
|
"learning_rate": 9.088702559576346e-05, |
|
"loss": 3.3243, |
|
"step": 63000 |
|
}, |
|
{ |
|
"epoch": 3.44, |
|
"grad_norm": 0.7484601140022278, |
|
"learning_rate": 9.059311562224185e-05, |
|
"loss": 3.3244, |
|
"step": 64000 |
|
}, |
|
{ |
|
"epoch": 3.5, |
|
"grad_norm": 0.746504008769989, |
|
"learning_rate": 9.029891144454253e-05, |
|
"loss": 3.3176, |
|
"step": 65000 |
|
}, |
|
{ |
|
"epoch": 3.55, |
|
"grad_norm": 0.7723560929298401, |
|
"learning_rate": 9.000500147102089e-05, |
|
"loss": 3.3123, |
|
"step": 66000 |
|
}, |
|
{ |
|
"epoch": 3.6, |
|
"grad_norm": 0.7322754859924316, |
|
"learning_rate": 8.971079729332157e-05, |
|
"loss": 3.3116, |
|
"step": 67000 |
|
}, |
|
{ |
|
"epoch": 3.66, |
|
"grad_norm": 0.7716720104217529, |
|
"learning_rate": 8.941659311562225e-05, |
|
"loss": 3.3073, |
|
"step": 68000 |
|
}, |
|
{ |
|
"epoch": 3.71, |
|
"grad_norm": 0.7449876666069031, |
|
"learning_rate": 8.912268314210063e-05, |
|
"loss": 3.3009, |
|
"step": 69000 |
|
}, |
|
{ |
|
"epoch": 3.76, |
|
"grad_norm": 0.7449329495429993, |
|
"learning_rate": 8.88284789644013e-05, |
|
"loss": 3.306, |
|
"step": 70000 |
|
}, |
|
{ |
|
"epoch": 3.82, |
|
"grad_norm": 0.701579213142395, |
|
"learning_rate": 8.853427478670198e-05, |
|
"loss": 3.2996, |
|
"step": 71000 |
|
}, |
|
{ |
|
"epoch": 3.87, |
|
"grad_norm": 0.7269728183746338, |
|
"learning_rate": 8.824036481318035e-05, |
|
"loss": 3.294, |
|
"step": 72000 |
|
}, |
|
{ |
|
"epoch": 3.93, |
|
"grad_norm": 0.7257309556007385, |
|
"learning_rate": 8.794616063548102e-05, |
|
"loss": 3.2955, |
|
"step": 73000 |
|
}, |
|
{ |
|
"epoch": 3.98, |
|
"grad_norm": 0.6977525353431702, |
|
"learning_rate": 8.76522506619594e-05, |
|
"loss": 3.2921, |
|
"step": 74000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.3883133849640905, |
|
"eval_loss": 3.5102479457855225, |
|
"eval_runtime": 153.2986, |
|
"eval_samples_per_second": 377.812, |
|
"eval_steps_per_second": 5.904, |
|
"step": 74380 |
|
}, |
|
{ |
|
"epoch": 4.03, |
|
"grad_norm": 0.7187395691871643, |
|
"learning_rate": 8.735804648426008e-05, |
|
"loss": 3.2537, |
|
"step": 75000 |
|
}, |
|
{ |
|
"epoch": 4.09, |
|
"grad_norm": 0.7566428184509277, |
|
"learning_rate": 8.706413651073846e-05, |
|
"loss": 3.2417, |
|
"step": 76000 |
|
}, |
|
{ |
|
"epoch": 4.14, |
|
"grad_norm": 0.7315700650215149, |
|
"learning_rate": 8.676993233303912e-05, |
|
"loss": 3.2394, |
|
"step": 77000 |
|
}, |
|
{ |
|
"epoch": 4.19, |
|
"grad_norm": 0.7529075741767883, |
|
"learning_rate": 8.64757281553398e-05, |
|
"loss": 3.2415, |
|
"step": 78000 |
|
}, |
|
{ |
|
"epoch": 4.25, |
|
"grad_norm": 0.7442134022712708, |
|
"learning_rate": 8.618152397764049e-05, |
|
"loss": 3.2399, |
|
"step": 79000 |
|
}, |
|
{ |
|
"epoch": 4.3, |
|
"grad_norm": 0.7772579193115234, |
|
"learning_rate": 8.588790820829656e-05, |
|
"loss": 3.2352, |
|
"step": 80000 |
|
}, |
|
{ |
|
"epoch": 4.36, |
|
"grad_norm": 0.7373689413070679, |
|
"learning_rate": 8.559370403059724e-05, |
|
"loss": 3.2348, |
|
"step": 81000 |
|
}, |
|
{ |
|
"epoch": 4.41, |
|
"grad_norm": 0.7681490778923035, |
|
"learning_rate": 8.529949985289791e-05, |
|
"loss": 3.2282, |
|
"step": 82000 |
|
}, |
|
{ |
|
"epoch": 4.46, |
|
"grad_norm": 0.7327556014060974, |
|
"learning_rate": 8.500529567519859e-05, |
|
"loss": 3.2349, |
|
"step": 83000 |
|
}, |
|
{ |
|
"epoch": 4.52, |
|
"grad_norm": 0.7320055961608887, |
|
"learning_rate": 8.471109149749927e-05, |
|
"loss": 3.2342, |
|
"step": 84000 |
|
}, |
|
{ |
|
"epoch": 4.57, |
|
"grad_norm": 0.7526534199714661, |
|
"learning_rate": 8.441718152397764e-05, |
|
"loss": 3.2329, |
|
"step": 85000 |
|
}, |
|
{ |
|
"epoch": 4.62, |
|
"grad_norm": 0.7839016914367676, |
|
"learning_rate": 8.412297734627832e-05, |
|
"loss": 3.2371, |
|
"step": 86000 |
|
}, |
|
{ |
|
"epoch": 4.68, |
|
"grad_norm": 0.7079972624778748, |
|
"learning_rate": 8.382906737275669e-05, |
|
"loss": 3.2293, |
|
"step": 87000 |
|
}, |
|
{ |
|
"epoch": 4.73, |
|
"grad_norm": 0.7379385232925415, |
|
"learning_rate": 8.353486319505737e-05, |
|
"loss": 3.2315, |
|
"step": 88000 |
|
}, |
|
{ |
|
"epoch": 4.79, |
|
"grad_norm": 0.7435966730117798, |
|
"learning_rate": 8.324065901735804e-05, |
|
"loss": 3.2285, |
|
"step": 89000 |
|
}, |
|
{ |
|
"epoch": 4.84, |
|
"grad_norm": 0.7578746676445007, |
|
"learning_rate": 8.294645483965872e-05, |
|
"loss": 3.2233, |
|
"step": 90000 |
|
}, |
|
{ |
|
"epoch": 4.89, |
|
"grad_norm": 0.7228344678878784, |
|
"learning_rate": 8.265254486613711e-05, |
|
"loss": 3.2261, |
|
"step": 91000 |
|
}, |
|
{ |
|
"epoch": 4.95, |
|
"grad_norm": 0.7377974987030029, |
|
"learning_rate": 8.235863489261548e-05, |
|
"loss": 3.2283, |
|
"step": 92000 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.3929738045053096, |
|
"eval_loss": 3.478940010070801, |
|
"eval_runtime": 154.0106, |
|
"eval_samples_per_second": 376.065, |
|
"eval_steps_per_second": 5.876, |
|
"step": 92975 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"grad_norm": 0.7944506406784058, |
|
"learning_rate": 8.206443071491615e-05, |
|
"loss": 3.2153, |
|
"step": 93000 |
|
}, |
|
{ |
|
"epoch": 5.06, |
|
"grad_norm": 0.752309262752533, |
|
"learning_rate": 8.177022653721683e-05, |
|
"loss": 3.16, |
|
"step": 94000 |
|
}, |
|
{ |
|
"epoch": 5.11, |
|
"grad_norm": 0.7849130630493164, |
|
"learning_rate": 8.147631656369521e-05, |
|
"loss": 3.172, |
|
"step": 95000 |
|
}, |
|
{ |
|
"epoch": 5.16, |
|
"grad_norm": 0.7549823522567749, |
|
"learning_rate": 8.118211238599589e-05, |
|
"loss": 3.1717, |
|
"step": 96000 |
|
}, |
|
{ |
|
"epoch": 5.22, |
|
"grad_norm": 0.761482834815979, |
|
"learning_rate": 8.088790820829656e-05, |
|
"loss": 3.1693, |
|
"step": 97000 |
|
}, |
|
{ |
|
"epoch": 5.27, |
|
"grad_norm": 0.7370353937149048, |
|
"learning_rate": 8.059399823477494e-05, |
|
"loss": 3.1749, |
|
"step": 98000 |
|
}, |
|
{ |
|
"epoch": 5.32, |
|
"grad_norm": 0.7203201651573181, |
|
"learning_rate": 8.029979405707561e-05, |
|
"loss": 3.1721, |
|
"step": 99000 |
|
}, |
|
{ |
|
"epoch": 5.38, |
|
"grad_norm": 0.7289775609970093, |
|
"learning_rate": 8.0005884083554e-05, |
|
"loss": 3.1809, |
|
"step": 100000 |
|
}, |
|
{ |
|
"epoch": 5.43, |
|
"grad_norm": 0.7284764051437378, |
|
"learning_rate": 7.971167990585466e-05, |
|
"loss": 3.1722, |
|
"step": 101000 |
|
}, |
|
{ |
|
"epoch": 5.49, |
|
"grad_norm": 0.7419421672821045, |
|
"learning_rate": 7.941747572815534e-05, |
|
"loss": 3.178, |
|
"step": 102000 |
|
}, |
|
{ |
|
"epoch": 5.54, |
|
"grad_norm": 0.7295576930046082, |
|
"learning_rate": 7.912356575463373e-05, |
|
"loss": 3.1777, |
|
"step": 103000 |
|
}, |
|
{ |
|
"epoch": 5.59, |
|
"grad_norm": 0.7482835054397583, |
|
"learning_rate": 7.88293615769344e-05, |
|
"loss": 3.178, |
|
"step": 104000 |
|
}, |
|
{ |
|
"epoch": 5.65, |
|
"grad_norm": 0.778439998626709, |
|
"learning_rate": 7.853515739923508e-05, |
|
"loss": 3.1735, |
|
"step": 105000 |
|
}, |
|
{ |
|
"epoch": 5.7, |
|
"grad_norm": 0.7204304337501526, |
|
"learning_rate": 7.824124742571345e-05, |
|
"loss": 3.1719, |
|
"step": 106000 |
|
}, |
|
{ |
|
"epoch": 5.75, |
|
"grad_norm": 0.7353498339653015, |
|
"learning_rate": 7.794704324801413e-05, |
|
"loss": 3.1761, |
|
"step": 107000 |
|
}, |
|
{ |
|
"epoch": 5.81, |
|
"grad_norm": 0.7248005867004395, |
|
"learning_rate": 7.76528390703148e-05, |
|
"loss": 3.1752, |
|
"step": 108000 |
|
}, |
|
{ |
|
"epoch": 5.86, |
|
"grad_norm": 0.7356109619140625, |
|
"learning_rate": 7.735863489261548e-05, |
|
"loss": 3.1754, |
|
"step": 109000 |
|
}, |
|
{ |
|
"epoch": 5.92, |
|
"grad_norm": 0.708113968372345, |
|
"learning_rate": 7.706472491909386e-05, |
|
"loss": 3.1744, |
|
"step": 110000 |
|
}, |
|
{ |
|
"epoch": 5.97, |
|
"grad_norm": 0.7147455215454102, |
|
"learning_rate": 7.677052074139453e-05, |
|
"loss": 3.1673, |
|
"step": 111000 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.3961948483989776, |
|
"eval_loss": 3.4379465579986572, |
|
"eval_runtime": 153.8126, |
|
"eval_samples_per_second": 376.549, |
|
"eval_steps_per_second": 5.884, |
|
"step": 111570 |
|
}, |
|
{ |
|
"epoch": 6.02, |
|
"grad_norm": 0.7331594824790955, |
|
"learning_rate": 7.64766107678729e-05, |
|
"loss": 3.1401, |
|
"step": 112000 |
|
}, |
|
{ |
|
"epoch": 6.08, |
|
"grad_norm": 0.7715177536010742, |
|
"learning_rate": 7.618240659017358e-05, |
|
"loss": 3.1162, |
|
"step": 113000 |
|
}, |
|
{ |
|
"epoch": 6.13, |
|
"grad_norm": 0.7466005682945251, |
|
"learning_rate": 7.588820241247426e-05, |
|
"loss": 3.1136, |
|
"step": 114000 |
|
}, |
|
{ |
|
"epoch": 6.18, |
|
"grad_norm": 0.734748899936676, |
|
"learning_rate": 7.559399823477493e-05, |
|
"loss": 3.1227, |
|
"step": 115000 |
|
}, |
|
{ |
|
"epoch": 6.24, |
|
"grad_norm": 0.7688254714012146, |
|
"learning_rate": 7.530008826125331e-05, |
|
"loss": 3.1252, |
|
"step": 116000 |
|
}, |
|
{ |
|
"epoch": 6.29, |
|
"grad_norm": 0.7620559930801392, |
|
"learning_rate": 7.5005884083554e-05, |
|
"loss": 3.1243, |
|
"step": 117000 |
|
}, |
|
{ |
|
"epoch": 6.35, |
|
"grad_norm": 0.7437037825584412, |
|
"learning_rate": 7.471167990585468e-05, |
|
"loss": 3.1286, |
|
"step": 118000 |
|
}, |
|
{ |
|
"epoch": 6.4, |
|
"grad_norm": 0.7445530295372009, |
|
"learning_rate": 7.441747572815534e-05, |
|
"loss": 3.1279, |
|
"step": 119000 |
|
}, |
|
{ |
|
"epoch": 6.45, |
|
"grad_norm": 0.7443315386772156, |
|
"learning_rate": 7.412356575463371e-05, |
|
"loss": 3.1314, |
|
"step": 120000 |
|
}, |
|
{ |
|
"epoch": 6.51, |
|
"grad_norm": 0.7500354647636414, |
|
"learning_rate": 7.38293615769344e-05, |
|
"loss": 3.1288, |
|
"step": 121000 |
|
}, |
|
{ |
|
"epoch": 6.56, |
|
"grad_norm": 0.7664729952812195, |
|
"learning_rate": 7.353545160341278e-05, |
|
"loss": 3.1307, |
|
"step": 122000 |
|
}, |
|
{ |
|
"epoch": 6.61, |
|
"grad_norm": 0.7711361646652222, |
|
"learning_rate": 7.324154162989115e-05, |
|
"loss": 3.132, |
|
"step": 123000 |
|
}, |
|
{ |
|
"epoch": 6.67, |
|
"grad_norm": 0.7488456964492798, |
|
"learning_rate": 7.294733745219182e-05, |
|
"loss": 3.1303, |
|
"step": 124000 |
|
}, |
|
{ |
|
"epoch": 6.72, |
|
"grad_norm": 0.7189005017280579, |
|
"learning_rate": 7.26531332744925e-05, |
|
"loss": 3.1285, |
|
"step": 125000 |
|
}, |
|
{ |
|
"epoch": 6.78, |
|
"grad_norm": 0.7578772902488708, |
|
"learning_rate": 7.235922330097088e-05, |
|
"loss": 3.1281, |
|
"step": 126000 |
|
}, |
|
{ |
|
"epoch": 6.83, |
|
"grad_norm": 0.7351284027099609, |
|
"learning_rate": 7.206501912327155e-05, |
|
"loss": 3.1305, |
|
"step": 127000 |
|
}, |
|
{ |
|
"epoch": 6.88, |
|
"grad_norm": 0.7461217641830444, |
|
"learning_rate": 7.177081494557223e-05, |
|
"loss": 3.1327, |
|
"step": 128000 |
|
}, |
|
{ |
|
"epoch": 6.94, |
|
"grad_norm": 0.7394551038742065, |
|
"learning_rate": 7.14769049720506e-05, |
|
"loss": 3.133, |
|
"step": 129000 |
|
}, |
|
{ |
|
"epoch": 6.99, |
|
"grad_norm": 0.7317260503768921, |
|
"learning_rate": 7.118270079435128e-05, |
|
"loss": 3.127, |
|
"step": 130000 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.39883100527338866, |
|
"eval_loss": 3.41751766204834, |
|
"eval_runtime": 153.2748, |
|
"eval_samples_per_second": 377.87, |
|
"eval_steps_per_second": 5.904, |
|
"step": 130165 |
|
}, |
|
{ |
|
"epoch": 7.04, |
|
"grad_norm": 0.7801479697227478, |
|
"learning_rate": 7.088879082082967e-05, |
|
"loss": 3.0813, |
|
"step": 131000 |
|
}, |
|
{ |
|
"epoch": 7.1, |
|
"grad_norm": 0.7538677453994751, |
|
"learning_rate": 7.059458664313034e-05, |
|
"loss": 3.077, |
|
"step": 132000 |
|
}, |
|
{ |
|
"epoch": 7.15, |
|
"grad_norm": 0.7728538513183594, |
|
"learning_rate": 7.030038246543102e-05, |
|
"loss": 3.0807, |
|
"step": 133000 |
|
}, |
|
{ |
|
"epoch": 7.21, |
|
"grad_norm": 0.7852578163146973, |
|
"learning_rate": 7.00061782877317e-05, |
|
"loss": 3.0811, |
|
"step": 134000 |
|
}, |
|
{ |
|
"epoch": 7.26, |
|
"grad_norm": 0.7735338807106018, |
|
"learning_rate": 6.971226831421006e-05, |
|
"loss": 3.0861, |
|
"step": 135000 |
|
}, |
|
{ |
|
"epoch": 7.31, |
|
"grad_norm": 0.7543830871582031, |
|
"learning_rate": 6.941835834068844e-05, |
|
"loss": 3.0846, |
|
"step": 136000 |
|
}, |
|
{ |
|
"epoch": 7.37, |
|
"grad_norm": 0.7399442195892334, |
|
"learning_rate": 6.912415416298912e-05, |
|
"loss": 3.0866, |
|
"step": 137000 |
|
}, |
|
{ |
|
"epoch": 7.42, |
|
"grad_norm": 0.7640860676765442, |
|
"learning_rate": 6.88299499852898e-05, |
|
"loss": 3.0893, |
|
"step": 138000 |
|
}, |
|
{ |
|
"epoch": 7.48, |
|
"grad_norm": 0.7997984886169434, |
|
"learning_rate": 6.853574580759047e-05, |
|
"loss": 3.0909, |
|
"step": 139000 |
|
}, |
|
{ |
|
"epoch": 7.53, |
|
"grad_norm": 0.7792380452156067, |
|
"learning_rate": 6.824183583406884e-05, |
|
"loss": 3.0939, |
|
"step": 140000 |
|
}, |
|
{ |
|
"epoch": 7.58, |
|
"grad_norm": 0.7403313517570496, |
|
"learning_rate": 6.794763165636952e-05, |
|
"loss": 3.0906, |
|
"step": 141000 |
|
}, |
|
{ |
|
"epoch": 7.64, |
|
"grad_norm": 0.7356191873550415, |
|
"learning_rate": 6.765342747867019e-05, |
|
"loss": 3.0909, |
|
"step": 142000 |
|
}, |
|
{ |
|
"epoch": 7.69, |
|
"grad_norm": 0.7950366735458374, |
|
"learning_rate": 6.735951750514857e-05, |
|
"loss": 3.0965, |
|
"step": 143000 |
|
}, |
|
{ |
|
"epoch": 7.74, |
|
"grad_norm": 0.7580872774124146, |
|
"learning_rate": 6.706531332744925e-05, |
|
"loss": 3.0922, |
|
"step": 144000 |
|
}, |
|
{ |
|
"epoch": 7.8, |
|
"grad_norm": 0.7285676002502441, |
|
"learning_rate": 6.677140335392763e-05, |
|
"loss": 3.0952, |
|
"step": 145000 |
|
}, |
|
{ |
|
"epoch": 7.85, |
|
"grad_norm": 0.730719268321991, |
|
"learning_rate": 6.64771991762283e-05, |
|
"loss": 3.0943, |
|
"step": 146000 |
|
}, |
|
{ |
|
"epoch": 7.91, |
|
"grad_norm": 0.7722078561782837, |
|
"learning_rate": 6.618299499852897e-05, |
|
"loss": 3.0957, |
|
"step": 147000 |
|
}, |
|
{ |
|
"epoch": 7.96, |
|
"grad_norm": 0.7444345951080322, |
|
"learning_rate": 6.588879082082966e-05, |
|
"loss": 3.0935, |
|
"step": 148000 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.39977229312392953, |
|
"eval_loss": 3.4271562099456787, |
|
"eval_runtime": 154.0268, |
|
"eval_samples_per_second": 376.025, |
|
"eval_steps_per_second": 5.876, |
|
"step": 148760 |
|
}, |
|
{ |
|
"epoch": 8.01, |
|
"grad_norm": 0.7648381590843201, |
|
"learning_rate": 6.559488084730804e-05, |
|
"loss": 3.0798, |
|
"step": 149000 |
|
}, |
|
{ |
|
"epoch": 8.07, |
|
"grad_norm": 0.746784508228302, |
|
"learning_rate": 6.530067666960871e-05, |
|
"loss": 3.0327, |
|
"step": 150000 |
|
}, |
|
{ |
|
"epoch": 8.12, |
|
"grad_norm": 0.7783890962600708, |
|
"learning_rate": 6.500647249190939e-05, |
|
"loss": 3.0398, |
|
"step": 151000 |
|
}, |
|
{ |
|
"epoch": 8.17, |
|
"grad_norm": 0.7425867915153503, |
|
"learning_rate": 6.471226831421007e-05, |
|
"loss": 3.0455, |
|
"step": 152000 |
|
}, |
|
{ |
|
"epoch": 8.23, |
|
"grad_norm": 0.7634230852127075, |
|
"learning_rate": 6.441835834068844e-05, |
|
"loss": 3.0529, |
|
"step": 153000 |
|
}, |
|
{ |
|
"epoch": 8.28, |
|
"grad_norm": 0.747742235660553, |
|
"learning_rate": 6.412415416298911e-05, |
|
"loss": 3.0487, |
|
"step": 154000 |
|
}, |
|
{ |
|
"epoch": 8.34, |
|
"grad_norm": 0.7804195284843445, |
|
"learning_rate": 6.383024418946749e-05, |
|
"loss": 3.0545, |
|
"step": 155000 |
|
}, |
|
{ |
|
"epoch": 8.39, |
|
"grad_norm": 0.7713324427604675, |
|
"learning_rate": 6.353604001176817e-05, |
|
"loss": 3.06, |
|
"step": 156000 |
|
}, |
|
{ |
|
"epoch": 8.44, |
|
"grad_norm": 0.7601629495620728, |
|
"learning_rate": 6.324183583406885e-05, |
|
"loss": 3.0537, |
|
"step": 157000 |
|
}, |
|
{ |
|
"epoch": 8.5, |
|
"grad_norm": 0.7745895385742188, |
|
"learning_rate": 6.294792586054723e-05, |
|
"loss": 3.0605, |
|
"step": 158000 |
|
}, |
|
{ |
|
"epoch": 8.55, |
|
"grad_norm": 0.7613347172737122, |
|
"learning_rate": 6.265372168284789e-05, |
|
"loss": 3.0573, |
|
"step": 159000 |
|
}, |
|
{ |
|
"epoch": 8.6, |
|
"grad_norm": 0.7753309607505798, |
|
"learning_rate": 6.235981170932628e-05, |
|
"loss": 3.0627, |
|
"step": 160000 |
|
}, |
|
{ |
|
"epoch": 8.66, |
|
"grad_norm": 0.7721776962280273, |
|
"learning_rate": 6.206560753162696e-05, |
|
"loss": 3.0612, |
|
"step": 161000 |
|
}, |
|
{ |
|
"epoch": 8.71, |
|
"grad_norm": 0.8062002062797546, |
|
"learning_rate": 6.177140335392763e-05, |
|
"loss": 3.0622, |
|
"step": 162000 |
|
}, |
|
{ |
|
"epoch": 8.77, |
|
"grad_norm": 0.7599053978919983, |
|
"learning_rate": 6.147719917622831e-05, |
|
"loss": 3.061, |
|
"step": 163000 |
|
}, |
|
{ |
|
"epoch": 8.82, |
|
"grad_norm": 0.7855945229530334, |
|
"learning_rate": 6.118328920270668e-05, |
|
"loss": 3.0638, |
|
"step": 164000 |
|
}, |
|
{ |
|
"epoch": 8.87, |
|
"grad_norm": 0.7659214735031128, |
|
"learning_rate": 6.088908502500735e-05, |
|
"loss": 3.0637, |
|
"step": 165000 |
|
}, |
|
{ |
|
"epoch": 8.93, |
|
"grad_norm": 0.7432619333267212, |
|
"learning_rate": 6.059517505148573e-05, |
|
"loss": 3.0661, |
|
"step": 166000 |
|
}, |
|
{ |
|
"epoch": 8.98, |
|
"grad_norm": 0.7428516745567322, |
|
"learning_rate": 6.030097087378641e-05, |
|
"loss": 3.0679, |
|
"step": 167000 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_accuracy": 0.4010507677364335, |
|
"eval_loss": 3.4149951934814453, |
|
"eval_runtime": 153.7947, |
|
"eval_samples_per_second": 376.593, |
|
"eval_steps_per_second": 5.884, |
|
"step": 167355 |
|
}, |
|
{ |
|
"epoch": 9.03, |
|
"grad_norm": 0.7959059476852417, |
|
"learning_rate": 6.000676669608709e-05, |
|
"loss": 3.0238, |
|
"step": 168000 |
|
}, |
|
{ |
|
"epoch": 9.09, |
|
"grad_norm": 0.7902527451515198, |
|
"learning_rate": 5.971256251838776e-05, |
|
"loss": 3.007, |
|
"step": 169000 |
|
}, |
|
{ |
|
"epoch": 9.14, |
|
"grad_norm": 0.7812909483909607, |
|
"learning_rate": 5.941865254486614e-05, |
|
"loss": 3.0162, |
|
"step": 170000 |
|
}, |
|
{ |
|
"epoch": 9.2, |
|
"grad_norm": 0.7669194936752319, |
|
"learning_rate": 5.9124742571344514e-05, |
|
"loss": 3.0162, |
|
"step": 171000 |
|
}, |
|
{ |
|
"epoch": 9.25, |
|
"grad_norm": 0.7774065136909485, |
|
"learning_rate": 5.8830538393645195e-05, |
|
"loss": 3.0219, |
|
"step": 172000 |
|
}, |
|
{ |
|
"epoch": 9.3, |
|
"grad_norm": 0.7773292064666748, |
|
"learning_rate": 5.853662842012357e-05, |
|
"loss": 3.0238, |
|
"step": 173000 |
|
}, |
|
{ |
|
"epoch": 9.36, |
|
"grad_norm": 0.7975197434425354, |
|
"learning_rate": 5.824242424242424e-05, |
|
"loss": 3.0246, |
|
"step": 174000 |
|
}, |
|
{ |
|
"epoch": 9.41, |
|
"grad_norm": 0.7579612731933594, |
|
"learning_rate": 5.794822006472492e-05, |
|
"loss": 3.0259, |
|
"step": 175000 |
|
}, |
|
{ |
|
"epoch": 9.46, |
|
"grad_norm": 0.7753346562385559, |
|
"learning_rate": 5.76540158870256e-05, |
|
"loss": 3.0287, |
|
"step": 176000 |
|
}, |
|
{ |
|
"epoch": 9.52, |
|
"grad_norm": 0.7583608031272888, |
|
"learning_rate": 5.7360105913503966e-05, |
|
"loss": 3.0281, |
|
"step": 177000 |
|
}, |
|
{ |
|
"epoch": 9.57, |
|
"grad_norm": 0.7782021760940552, |
|
"learning_rate": 5.706590173580465e-05, |
|
"loss": 3.0298, |
|
"step": 178000 |
|
}, |
|
{ |
|
"epoch": 9.63, |
|
"grad_norm": 0.7898275852203369, |
|
"learning_rate": 5.677169755810533e-05, |
|
"loss": 3.0384, |
|
"step": 179000 |
|
}, |
|
{ |
|
"epoch": 9.68, |
|
"grad_norm": 0.7844170928001404, |
|
"learning_rate": 5.6477787584583706e-05, |
|
"loss": 3.0341, |
|
"step": 180000 |
|
}, |
|
{ |
|
"epoch": 9.73, |
|
"grad_norm": 0.7562811374664307, |
|
"learning_rate": 5.618387761106208e-05, |
|
"loss": 3.0346, |
|
"step": 181000 |
|
}, |
|
{ |
|
"epoch": 9.79, |
|
"grad_norm": 0.7964505553245544, |
|
"learning_rate": 5.588967343336276e-05, |
|
"loss": 3.0364, |
|
"step": 182000 |
|
}, |
|
{ |
|
"epoch": 9.84, |
|
"grad_norm": 0.7780364751815796, |
|
"learning_rate": 5.5595763459841135e-05, |
|
"loss": 3.0376, |
|
"step": 183000 |
|
}, |
|
{ |
|
"epoch": 9.9, |
|
"grad_norm": 0.7689926028251648, |
|
"learning_rate": 5.530155928214181e-05, |
|
"loss": 3.0343, |
|
"step": 184000 |
|
}, |
|
{ |
|
"epoch": 9.95, |
|
"grad_norm": 0.7878050804138184, |
|
"learning_rate": 5.5007355104442484e-05, |
|
"loss": 3.0402, |
|
"step": 185000 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_accuracy": 0.4023324670878444, |
|
"eval_loss": 3.414785146713257, |
|
"eval_runtime": 153.4202, |
|
"eval_samples_per_second": 377.512, |
|
"eval_steps_per_second": 5.899, |
|
"step": 185950 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"grad_norm": 0.7869421243667603, |
|
"learning_rate": 5.471344513092086e-05, |
|
"loss": 3.0355, |
|
"step": 186000 |
|
}, |
|
{ |
|
"epoch": 10.06, |
|
"grad_norm": 0.7897782325744629, |
|
"learning_rate": 5.441924095322154e-05, |
|
"loss": 2.9791, |
|
"step": 187000 |
|
}, |
|
{ |
|
"epoch": 10.11, |
|
"grad_norm": 0.810604989528656, |
|
"learning_rate": 5.412503677552222e-05, |
|
"loss": 2.9856, |
|
"step": 188000 |
|
}, |
|
{ |
|
"epoch": 10.16, |
|
"grad_norm": 0.802144467830658, |
|
"learning_rate": 5.383112680200059e-05, |
|
"loss": 2.9921, |
|
"step": 189000 |
|
}, |
|
{ |
|
"epoch": 10.22, |
|
"grad_norm": 0.8142128586769104, |
|
"learning_rate": 5.353692262430127e-05, |
|
"loss": 2.9952, |
|
"step": 190000 |
|
}, |
|
{ |
|
"epoch": 10.27, |
|
"grad_norm": 0.792819082736969, |
|
"learning_rate": 5.3243012650779646e-05, |
|
"loss": 2.9997, |
|
"step": 191000 |
|
}, |
|
{ |
|
"epoch": 10.33, |
|
"grad_norm": 0.7813856601715088, |
|
"learning_rate": 5.294880847308033e-05, |
|
"loss": 3.0017, |
|
"step": 192000 |
|
}, |
|
{ |
|
"epoch": 10.38, |
|
"grad_norm": 0.8088542222976685, |
|
"learning_rate": 5.2654604295380995e-05, |
|
"loss": 3.0005, |
|
"step": 193000 |
|
}, |
|
{ |
|
"epoch": 10.43, |
|
"grad_norm": 0.7641571164131165, |
|
"learning_rate": 5.2360400117681677e-05, |
|
"loss": 3.0007, |
|
"step": 194000 |
|
}, |
|
{ |
|
"epoch": 10.49, |
|
"grad_norm": 0.7936010956764221, |
|
"learning_rate": 5.2066490144160054e-05, |
|
"loss": 3.0038, |
|
"step": 195000 |
|
}, |
|
{ |
|
"epoch": 10.54, |
|
"grad_norm": 0.8503419756889343, |
|
"learning_rate": 5.177258017063843e-05, |
|
"loss": 3.0049, |
|
"step": 196000 |
|
}, |
|
{ |
|
"epoch": 10.59, |
|
"grad_norm": 0.7761277556419373, |
|
"learning_rate": 5.14783759929391e-05, |
|
"loss": 3.0079, |
|
"step": 197000 |
|
}, |
|
{ |
|
"epoch": 10.65, |
|
"grad_norm": 0.7853693962097168, |
|
"learning_rate": 5.118417181523978e-05, |
|
"loss": 3.0091, |
|
"step": 198000 |
|
}, |
|
{ |
|
"epoch": 10.7, |
|
"grad_norm": 0.7898029685020447, |
|
"learning_rate": 5.089026184171816e-05, |
|
"loss": 3.0047, |
|
"step": 199000 |
|
}, |
|
{ |
|
"epoch": 10.76, |
|
"grad_norm": 0.7760916948318481, |
|
"learning_rate": 5.0596351868196535e-05, |
|
"loss": 3.01, |
|
"step": 200000 |
|
}, |
|
{ |
|
"epoch": 10.81, |
|
"grad_norm": 0.8151899576187134, |
|
"learning_rate": 5.03021476904972e-05, |
|
"loss": 3.0079, |
|
"step": 201000 |
|
}, |
|
{ |
|
"epoch": 10.86, |
|
"grad_norm": 0.837149977684021, |
|
"learning_rate": 5.0007943512797884e-05, |
|
"loss": 3.0115, |
|
"step": 202000 |
|
}, |
|
{ |
|
"epoch": 10.92, |
|
"grad_norm": 0.7857736945152283, |
|
"learning_rate": 4.971373933509856e-05, |
|
"loss": 3.0123, |
|
"step": 203000 |
|
}, |
|
{ |
|
"epoch": 10.97, |
|
"grad_norm": 0.7611953020095825, |
|
"learning_rate": 4.9419829361576935e-05, |
|
"loss": 3.0082, |
|
"step": 204000 |
|
}, |
|
{ |
|
"epoch": 11.0, |
|
"eval_accuracy": 0.4029490640396869, |
|
"eval_loss": 3.4169130325317383, |
|
"eval_runtime": 153.2931, |
|
"eval_samples_per_second": 377.825, |
|
"eval_steps_per_second": 5.904, |
|
"step": 204545 |
|
}, |
|
{ |
|
"epoch": 11.02, |
|
"grad_norm": 0.816586971282959, |
|
"learning_rate": 4.9125625183877617e-05, |
|
"loss": 2.9829, |
|
"step": 205000 |
|
}, |
|
{ |
|
"epoch": 11.08, |
|
"grad_norm": 0.7916204929351807, |
|
"learning_rate": 4.883171521035599e-05, |
|
"loss": 2.9574, |
|
"step": 206000 |
|
}, |
|
{ |
|
"epoch": 11.13, |
|
"grad_norm": 0.8013209104537964, |
|
"learning_rate": 4.853751103265667e-05, |
|
"loss": 2.9652, |
|
"step": 207000 |
|
}, |
|
{ |
|
"epoch": 11.19, |
|
"grad_norm": 0.7968335747718811, |
|
"learning_rate": 4.824330685495734e-05, |
|
"loss": 2.9686, |
|
"step": 208000 |
|
}, |
|
{ |
|
"epoch": 11.24, |
|
"grad_norm": 0.8586769104003906, |
|
"learning_rate": 4.794910267725802e-05, |
|
"loss": 2.9722, |
|
"step": 209000 |
|
}, |
|
{ |
|
"epoch": 11.29, |
|
"grad_norm": 0.8471382260322571, |
|
"learning_rate": 4.7655192703736395e-05, |
|
"loss": 2.9747, |
|
"step": 210000 |
|
}, |
|
{ |
|
"epoch": 11.35, |
|
"grad_norm": 0.8427879810333252, |
|
"learning_rate": 4.736098852603707e-05, |
|
"loss": 2.9741, |
|
"step": 211000 |
|
}, |
|
{ |
|
"epoch": 11.4, |
|
"grad_norm": 0.8404529094696045, |
|
"learning_rate": 4.706678434833775e-05, |
|
"loss": 2.9773, |
|
"step": 212000 |
|
}, |
|
{ |
|
"epoch": 11.45, |
|
"grad_norm": 0.8058509826660156, |
|
"learning_rate": 4.677287437481612e-05, |
|
"loss": 2.9749, |
|
"step": 213000 |
|
}, |
|
{ |
|
"epoch": 11.51, |
|
"grad_norm": 0.8321877717971802, |
|
"learning_rate": 4.64786701971168e-05, |
|
"loss": 2.9794, |
|
"step": 214000 |
|
}, |
|
{ |
|
"epoch": 11.56, |
|
"grad_norm": 0.8273854851722717, |
|
"learning_rate": 4.618476022359517e-05, |
|
"loss": 2.9855, |
|
"step": 215000 |
|
}, |
|
{ |
|
"epoch": 11.62, |
|
"grad_norm": 0.8307694792747498, |
|
"learning_rate": 4.589085025007355e-05, |
|
"loss": 2.9862, |
|
"step": 216000 |
|
}, |
|
{ |
|
"epoch": 11.67, |
|
"grad_norm": 0.8184703588485718, |
|
"learning_rate": 4.559694027655193e-05, |
|
"loss": 2.9836, |
|
"step": 217000 |
|
}, |
|
{ |
|
"epoch": 11.72, |
|
"grad_norm": 0.7970172762870789, |
|
"learning_rate": 4.530273609885261e-05, |
|
"loss": 2.9843, |
|
"step": 218000 |
|
}, |
|
{ |
|
"epoch": 11.78, |
|
"grad_norm": 0.8030002117156982, |
|
"learning_rate": 4.500853192115328e-05, |
|
"loss": 2.9891, |
|
"step": 219000 |
|
}, |
|
{ |
|
"epoch": 11.83, |
|
"grad_norm": 0.795408308506012, |
|
"learning_rate": 4.471432774345396e-05, |
|
"loss": 2.987, |
|
"step": 220000 |
|
}, |
|
{ |
|
"epoch": 11.88, |
|
"grad_norm": 0.8084612488746643, |
|
"learning_rate": 4.4420417769932335e-05, |
|
"loss": 2.9898, |
|
"step": 221000 |
|
}, |
|
{ |
|
"epoch": 11.94, |
|
"grad_norm": 0.8061679005622864, |
|
"learning_rate": 4.412650779641071e-05, |
|
"loss": 2.9941, |
|
"step": 222000 |
|
}, |
|
{ |
|
"epoch": 11.99, |
|
"grad_norm": 0.8087071776390076, |
|
"learning_rate": 4.383230361871139e-05, |
|
"loss": 2.988, |
|
"step": 223000 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"eval_accuracy": 0.4035294498608861, |
|
"eval_loss": 3.412954568862915, |
|
"eval_runtime": 153.0197, |
|
"eval_samples_per_second": 378.5, |
|
"eval_steps_per_second": 5.914, |
|
"step": 223140 |
|
}, |
|
{ |
|
"epoch": 12.05, |
|
"grad_norm": 0.850972056388855, |
|
"learning_rate": 4.353809944101207e-05, |
|
"loss": 2.9439, |
|
"step": 224000 |
|
}, |
|
{ |
|
"epoch": 12.1, |
|
"grad_norm": 0.8421295285224915, |
|
"learning_rate": 4.324389526331274e-05, |
|
"loss": 2.9402, |
|
"step": 225000 |
|
}, |
|
{ |
|
"epoch": 12.15, |
|
"grad_norm": 0.8112582564353943, |
|
"learning_rate": 4.294998528979112e-05, |
|
"loss": 2.9473, |
|
"step": 226000 |
|
}, |
|
{ |
|
"epoch": 12.21, |
|
"grad_norm": 0.8302417397499084, |
|
"learning_rate": 4.2655781112091794e-05, |
|
"loss": 2.9491, |
|
"step": 227000 |
|
}, |
|
{ |
|
"epoch": 12.26, |
|
"grad_norm": 0.8474385738372803, |
|
"learning_rate": 4.2361576934392475e-05, |
|
"loss": 2.9484, |
|
"step": 228000 |
|
}, |
|
{ |
|
"epoch": 12.32, |
|
"grad_norm": 0.8351003527641296, |
|
"learning_rate": 4.206737275669315e-05, |
|
"loss": 2.9529, |
|
"step": 229000 |
|
}, |
|
{ |
|
"epoch": 12.37, |
|
"grad_norm": 0.85118168592453, |
|
"learning_rate": 4.177346278317153e-05, |
|
"loss": 2.9562, |
|
"step": 230000 |
|
}, |
|
{ |
|
"epoch": 12.42, |
|
"grad_norm": 0.8224354982376099, |
|
"learning_rate": 4.14792586054722e-05, |
|
"loss": 2.9539, |
|
"step": 231000 |
|
}, |
|
{ |
|
"epoch": 12.48, |
|
"grad_norm": 0.8389358520507812, |
|
"learning_rate": 4.1185054427772876e-05, |
|
"loss": 2.9581, |
|
"step": 232000 |
|
}, |
|
{ |
|
"epoch": 12.53, |
|
"grad_norm": 0.8302854895591736, |
|
"learning_rate": 4.089114445425125e-05, |
|
"loss": 2.9604, |
|
"step": 233000 |
|
}, |
|
{ |
|
"epoch": 12.58, |
|
"grad_norm": 0.8315780758857727, |
|
"learning_rate": 4.0596940276551934e-05, |
|
"loss": 2.9655, |
|
"step": 234000 |
|
}, |
|
{ |
|
"epoch": 12.64, |
|
"grad_norm": 0.8219565749168396, |
|
"learning_rate": 4.030273609885261e-05, |
|
"loss": 2.9651, |
|
"step": 235000 |
|
}, |
|
{ |
|
"epoch": 12.69, |
|
"grad_norm": 0.8112459778785706, |
|
"learning_rate": 4.0008826125330986e-05, |
|
"loss": 2.9647, |
|
"step": 236000 |
|
}, |
|
{ |
|
"epoch": 12.75, |
|
"grad_norm": 0.8309066295623779, |
|
"learning_rate": 3.9714916151809357e-05, |
|
"loss": 2.9637, |
|
"step": 237000 |
|
}, |
|
{ |
|
"epoch": 12.8, |
|
"grad_norm": 0.818537175655365, |
|
"learning_rate": 3.942071197411004e-05, |
|
"loss": 2.9648, |
|
"step": 238000 |
|
}, |
|
{ |
|
"epoch": 12.85, |
|
"grad_norm": 0.829467236995697, |
|
"learning_rate": 3.912650779641071e-05, |
|
"loss": 2.9676, |
|
"step": 239000 |
|
}, |
|
{ |
|
"epoch": 12.91, |
|
"grad_norm": 0.8041892647743225, |
|
"learning_rate": 3.883230361871139e-05, |
|
"loss": 2.9631, |
|
"step": 240000 |
|
}, |
|
{ |
|
"epoch": 12.96, |
|
"grad_norm": 0.8085007667541504, |
|
"learning_rate": 3.8538393645189764e-05, |
|
"loss": 2.967, |
|
"step": 241000 |
|
}, |
|
{ |
|
"epoch": 13.0, |
|
"eval_accuracy": 0.40449210160668586, |
|
"eval_loss": 3.3969216346740723, |
|
"eval_runtime": 154.1242, |
|
"eval_samples_per_second": 375.788, |
|
"eval_steps_per_second": 5.872, |
|
"step": 241735 |
|
}, |
|
{ |
|
"epoch": 13.01, |
|
"grad_norm": 0.8514643311500549, |
|
"learning_rate": 3.824418946749044e-05, |
|
"loss": 2.9558, |
|
"step": 242000 |
|
}, |
|
{ |
|
"epoch": 13.07, |
|
"grad_norm": 0.8870530724525452, |
|
"learning_rate": 3.794998528979112e-05, |
|
"loss": 2.9199, |
|
"step": 243000 |
|
}, |
|
{ |
|
"epoch": 13.12, |
|
"grad_norm": 0.8721070885658264, |
|
"learning_rate": 3.765607531626949e-05, |
|
"loss": 2.9276, |
|
"step": 244000 |
|
}, |
|
{ |
|
"epoch": 13.18, |
|
"grad_norm": 0.8669698238372803, |
|
"learning_rate": 3.736187113857017e-05, |
|
"loss": 2.9317, |
|
"step": 245000 |
|
}, |
|
{ |
|
"epoch": 13.23, |
|
"grad_norm": 0.8661383390426636, |
|
"learning_rate": 3.7067666960870846e-05, |
|
"loss": 2.9308, |
|
"step": 246000 |
|
}, |
|
{ |
|
"epoch": 13.28, |
|
"grad_norm": 0.860162079334259, |
|
"learning_rate": 3.677346278317153e-05, |
|
"loss": 2.9322, |
|
"step": 247000 |
|
}, |
|
{ |
|
"epoch": 13.34, |
|
"grad_norm": 0.8323468565940857, |
|
"learning_rate": 3.64795528096499e-05, |
|
"loss": 2.9336, |
|
"step": 248000 |
|
}, |
|
{ |
|
"epoch": 13.39, |
|
"grad_norm": 0.8697391748428345, |
|
"learning_rate": 3.6185642836128275e-05, |
|
"loss": 2.9342, |
|
"step": 249000 |
|
}, |
|
{ |
|
"epoch": 13.44, |
|
"grad_norm": 0.8538952469825745, |
|
"learning_rate": 3.589143865842895e-05, |
|
"loss": 2.9373, |
|
"step": 250000 |
|
}, |
|
{ |
|
"epoch": 13.5, |
|
"grad_norm": 0.848420262336731, |
|
"learning_rate": 3.559723448072963e-05, |
|
"loss": 2.9357, |
|
"step": 251000 |
|
}, |
|
{ |
|
"epoch": 13.55, |
|
"grad_norm": 0.8527274131774902, |
|
"learning_rate": 3.5303324507208e-05, |
|
"loss": 2.9394, |
|
"step": 252000 |
|
}, |
|
{ |
|
"epoch": 13.61, |
|
"grad_norm": 0.8854946494102478, |
|
"learning_rate": 3.500912032950868e-05, |
|
"loss": 2.9423, |
|
"step": 253000 |
|
}, |
|
{ |
|
"epoch": 13.66, |
|
"grad_norm": 0.8356419801712036, |
|
"learning_rate": 3.471491615180936e-05, |
|
"loss": 2.9412, |
|
"step": 254000 |
|
}, |
|
{ |
|
"epoch": 13.71, |
|
"grad_norm": 0.8529571890830994, |
|
"learning_rate": 3.442071197411003e-05, |
|
"loss": 2.9442, |
|
"step": 255000 |
|
}, |
|
{ |
|
"epoch": 13.77, |
|
"grad_norm": 0.8669166564941406, |
|
"learning_rate": 3.412650779641071e-05, |
|
"loss": 2.945, |
|
"step": 256000 |
|
}, |
|
{ |
|
"epoch": 13.82, |
|
"grad_norm": 0.8501659035682678, |
|
"learning_rate": 3.3832892027066786e-05, |
|
"loss": 2.9491, |
|
"step": 257000 |
|
}, |
|
{ |
|
"epoch": 13.87, |
|
"grad_norm": 0.8519834876060486, |
|
"learning_rate": 3.353868784936746e-05, |
|
"loss": 2.946, |
|
"step": 258000 |
|
}, |
|
{ |
|
"epoch": 13.93, |
|
"grad_norm": 0.8449692130088806, |
|
"learning_rate": 3.324477787584584e-05, |
|
"loss": 2.9439, |
|
"step": 259000 |
|
}, |
|
{ |
|
"epoch": 13.98, |
|
"grad_norm": 0.8286080360412598, |
|
"learning_rate": 3.2950867902324215e-05, |
|
"loss": 2.9457, |
|
"step": 260000 |
|
}, |
|
{ |
|
"epoch": 14.0, |
|
"eval_accuracy": 0.4048622747603851, |
|
"eval_loss": 3.4079537391662598, |
|
"eval_runtime": 153.3879, |
|
"eval_samples_per_second": 377.592, |
|
"eval_steps_per_second": 5.9, |
|
"step": 260330 |
|
}, |
|
{ |
|
"epoch": 14.04, |
|
"grad_norm": 0.8589282631874084, |
|
"learning_rate": 3.265666372462489e-05, |
|
"loss": 2.918, |
|
"step": 261000 |
|
}, |
|
{ |
|
"epoch": 14.09, |
|
"grad_norm": 0.8760070204734802, |
|
"learning_rate": 3.2362459546925564e-05, |
|
"loss": 2.9044, |
|
"step": 262000 |
|
}, |
|
{ |
|
"epoch": 14.14, |
|
"grad_norm": 0.883084237575531, |
|
"learning_rate": 3.2068255369226245e-05, |
|
"loss": 2.9054, |
|
"step": 263000 |
|
}, |
|
{ |
|
"epoch": 14.2, |
|
"grad_norm": 0.8560861349105835, |
|
"learning_rate": 3.177405119152692e-05, |
|
"loss": 2.9076, |
|
"step": 264000 |
|
}, |
|
{ |
|
"epoch": 14.25, |
|
"grad_norm": 0.9030760526657104, |
|
"learning_rate": 3.1479847013827594e-05, |
|
"loss": 2.9111, |
|
"step": 265000 |
|
}, |
|
{ |
|
"epoch": 14.3, |
|
"grad_norm": 0.8720868229866028, |
|
"learning_rate": 3.118593704030597e-05, |
|
"loss": 2.9138, |
|
"step": 266000 |
|
}, |
|
{ |
|
"epoch": 14.36, |
|
"grad_norm": 0.8849419951438904, |
|
"learning_rate": 3.089202706678435e-05, |
|
"loss": 2.9211, |
|
"step": 267000 |
|
}, |
|
{ |
|
"epoch": 14.41, |
|
"grad_norm": 0.8608791828155518, |
|
"learning_rate": 3.059782288908502e-05, |
|
"loss": 2.9176, |
|
"step": 268000 |
|
}, |
|
{ |
|
"epoch": 14.47, |
|
"grad_norm": 0.868854284286499, |
|
"learning_rate": 3.03036187113857e-05, |
|
"loss": 2.9216, |
|
"step": 269000 |
|
}, |
|
{ |
|
"epoch": 14.52, |
|
"grad_norm": 0.8694607615470886, |
|
"learning_rate": 3.0009414533686382e-05, |
|
"loss": 2.9244, |
|
"step": 270000 |
|
}, |
|
{ |
|
"epoch": 14.57, |
|
"grad_norm": 0.8707175254821777, |
|
"learning_rate": 2.9715504560164753e-05, |
|
"loss": 2.9216, |
|
"step": 271000 |
|
}, |
|
{ |
|
"epoch": 14.63, |
|
"grad_norm": 0.907792329788208, |
|
"learning_rate": 2.9421300382465434e-05, |
|
"loss": 2.927, |
|
"step": 272000 |
|
}, |
|
{ |
|
"epoch": 14.68, |
|
"grad_norm": 0.8800376057624817, |
|
"learning_rate": 2.912739040894381e-05, |
|
"loss": 2.9259, |
|
"step": 273000 |
|
}, |
|
{ |
|
"epoch": 14.74, |
|
"grad_norm": 0.8789490461349487, |
|
"learning_rate": 2.8833186231244486e-05, |
|
"loss": 2.9261, |
|
"step": 274000 |
|
}, |
|
{ |
|
"epoch": 14.79, |
|
"grad_norm": 0.8529285192489624, |
|
"learning_rate": 2.8539276257722863e-05, |
|
"loss": 2.9234, |
|
"step": 275000 |
|
}, |
|
{ |
|
"epoch": 14.84, |
|
"grad_norm": 0.8620697855949402, |
|
"learning_rate": 2.8245072080023538e-05, |
|
"loss": 2.9255, |
|
"step": 276000 |
|
}, |
|
{ |
|
"epoch": 14.9, |
|
"grad_norm": 0.8713541626930237, |
|
"learning_rate": 2.7950867902324212e-05, |
|
"loss": 2.9287, |
|
"step": 277000 |
|
}, |
|
{ |
|
"epoch": 14.95, |
|
"grad_norm": 0.8321398496627808, |
|
"learning_rate": 2.765695792880259e-05, |
|
"loss": 2.9268, |
|
"step": 278000 |
|
}, |
|
{ |
|
"epoch": 15.0, |
|
"eval_accuracy": 0.40448188993348033, |
|
"eval_loss": 3.4128763675689697, |
|
"eval_runtime": 153.2847, |
|
"eval_samples_per_second": 377.846, |
|
"eval_steps_per_second": 5.904, |
|
"step": 278925 |
|
}, |
|
{ |
|
"epoch": 15.0, |
|
"grad_norm": 0.8870017528533936, |
|
"learning_rate": 2.7363047955280967e-05, |
|
"loss": 2.9236, |
|
"step": 279000 |
|
}, |
|
{ |
|
"epoch": 15.06, |
|
"grad_norm": 0.902341902256012, |
|
"learning_rate": 2.706884377758164e-05, |
|
"loss": 2.8917, |
|
"step": 280000 |
|
}, |
|
{ |
|
"epoch": 15.11, |
|
"grad_norm": 0.8734650015830994, |
|
"learning_rate": 2.677463959988232e-05, |
|
"loss": 2.8882, |
|
"step": 281000 |
|
}, |
|
{ |
|
"epoch": 15.17, |
|
"grad_norm": 0.8736613392829895, |
|
"learning_rate": 2.6480435422182997e-05, |
|
"loss": 2.895, |
|
"step": 282000 |
|
}, |
|
{ |
|
"epoch": 15.22, |
|
"grad_norm": 0.8876488208770752, |
|
"learning_rate": 2.618652544866137e-05, |
|
"loss": 2.8917, |
|
"step": 283000 |
|
}, |
|
{ |
|
"epoch": 15.27, |
|
"grad_norm": 0.9243654012680054, |
|
"learning_rate": 2.589232127096205e-05, |
|
"loss": 2.8952, |
|
"step": 284000 |
|
}, |
|
{ |
|
"epoch": 15.33, |
|
"grad_norm": 0.8820899724960327, |
|
"learning_rate": 2.5598117093262723e-05, |
|
"loss": 2.8973, |
|
"step": 285000 |
|
}, |
|
{ |
|
"epoch": 15.38, |
|
"grad_norm": 0.9063917398452759, |
|
"learning_rate": 2.5304207119741104e-05, |
|
"loss": 2.8972, |
|
"step": 286000 |
|
}, |
|
{ |
|
"epoch": 15.43, |
|
"grad_norm": 0.9035106301307678, |
|
"learning_rate": 2.501000294204178e-05, |
|
"loss": 2.9064, |
|
"step": 287000 |
|
}, |
|
{ |
|
"epoch": 15.49, |
|
"grad_norm": 0.8639585971832275, |
|
"learning_rate": 2.4716092968520156e-05, |
|
"loss": 2.9013, |
|
"step": 288000 |
|
}, |
|
{ |
|
"epoch": 15.54, |
|
"grad_norm": 0.8969417810440063, |
|
"learning_rate": 2.4421888790820833e-05, |
|
"loss": 2.9056, |
|
"step": 289000 |
|
}, |
|
{ |
|
"epoch": 15.6, |
|
"grad_norm": 0.8919821977615356, |
|
"learning_rate": 2.4127684613121508e-05, |
|
"loss": 2.9073, |
|
"step": 290000 |
|
}, |
|
{ |
|
"epoch": 15.65, |
|
"grad_norm": 0.8989285230636597, |
|
"learning_rate": 2.3833774639599885e-05, |
|
"loss": 2.9062, |
|
"step": 291000 |
|
}, |
|
{ |
|
"epoch": 15.7, |
|
"grad_norm": 0.8888868689537048, |
|
"learning_rate": 2.353986466607826e-05, |
|
"loss": 2.9088, |
|
"step": 292000 |
|
}, |
|
{ |
|
"epoch": 15.76, |
|
"grad_norm": 0.9275554418563843, |
|
"learning_rate": 2.3245660488378937e-05, |
|
"loss": 2.9069, |
|
"step": 293000 |
|
}, |
|
{ |
|
"epoch": 15.81, |
|
"grad_norm": 0.8558235168457031, |
|
"learning_rate": 2.295145631067961e-05, |
|
"loss": 2.9037, |
|
"step": 294000 |
|
}, |
|
{ |
|
"epoch": 15.86, |
|
"grad_norm": 0.8763130307197571, |
|
"learning_rate": 2.265725213298029e-05, |
|
"loss": 2.9064, |
|
"step": 295000 |
|
}, |
|
{ |
|
"epoch": 15.92, |
|
"grad_norm": 0.8805882334709167, |
|
"learning_rate": 2.2363342159458667e-05, |
|
"loss": 2.9079, |
|
"step": 296000 |
|
}, |
|
{ |
|
"epoch": 15.97, |
|
"grad_norm": 0.9118674993515015, |
|
"learning_rate": 2.206943218593704e-05, |
|
"loss": 2.911, |
|
"step": 297000 |
|
}, |
|
{ |
|
"epoch": 16.0, |
|
"eval_accuracy": 0.4047090324802421, |
|
"eval_loss": 3.41589617729187, |
|
"eval_runtime": 153.0438, |
|
"eval_samples_per_second": 378.441, |
|
"eval_steps_per_second": 5.913, |
|
"step": 297520 |
|
}, |
|
{ |
|
"epoch": 16.03, |
|
"grad_norm": 0.9044952392578125, |
|
"learning_rate": 2.177522800823772e-05, |
|
"loss": 2.8973, |
|
"step": 298000 |
|
}, |
|
{ |
|
"epoch": 16.08, |
|
"grad_norm": 0.8891879320144653, |
|
"learning_rate": 2.1481023830538393e-05, |
|
"loss": 2.8726, |
|
"step": 299000 |
|
}, |
|
{ |
|
"epoch": 16.13, |
|
"grad_norm": 0.9434779286384583, |
|
"learning_rate": 2.118681965283907e-05, |
|
"loss": 2.8738, |
|
"step": 300000 |
|
}, |
|
{ |
|
"epoch": 16.19, |
|
"grad_norm": 0.9484353065490723, |
|
"learning_rate": 2.0892909679317445e-05, |
|
"loss": 2.8755, |
|
"step": 301000 |
|
}, |
|
{ |
|
"epoch": 16.24, |
|
"grad_norm": 0.9076915979385376, |
|
"learning_rate": 2.0598705501618122e-05, |
|
"loss": 2.8804, |
|
"step": 302000 |
|
}, |
|
{ |
|
"epoch": 16.29, |
|
"grad_norm": 0.9162052869796753, |
|
"learning_rate": 2.03047955280965e-05, |
|
"loss": 2.8824, |
|
"step": 303000 |
|
}, |
|
{ |
|
"epoch": 16.35, |
|
"grad_norm": 0.9373551607131958, |
|
"learning_rate": 2.0010591350397174e-05, |
|
"loss": 2.8834, |
|
"step": 304000 |
|
}, |
|
{ |
|
"epoch": 16.4, |
|
"grad_norm": 0.9191654324531555, |
|
"learning_rate": 1.9716387172697852e-05, |
|
"loss": 2.8875, |
|
"step": 305000 |
|
}, |
|
{ |
|
"epoch": 16.46, |
|
"grad_norm": 0.9333674907684326, |
|
"learning_rate": 1.942218299499853e-05, |
|
"loss": 2.8867, |
|
"step": 306000 |
|
}, |
|
{ |
|
"epoch": 16.51, |
|
"grad_norm": 0.9224890470504761, |
|
"learning_rate": 1.9128273021476907e-05, |
|
"loss": 2.886, |
|
"step": 307000 |
|
}, |
|
{ |
|
"epoch": 16.56, |
|
"grad_norm": 0.9154072403907776, |
|
"learning_rate": 1.8834068843777582e-05, |
|
"loss": 2.8859, |
|
"step": 308000 |
|
}, |
|
{ |
|
"epoch": 16.62, |
|
"grad_norm": 0.8822410702705383, |
|
"learning_rate": 1.854015887025596e-05, |
|
"loss": 2.8887, |
|
"step": 309000 |
|
}, |
|
{ |
|
"epoch": 16.67, |
|
"grad_norm": 0.9022771716117859, |
|
"learning_rate": 1.8245954692556637e-05, |
|
"loss": 2.8884, |
|
"step": 310000 |
|
}, |
|
{ |
|
"epoch": 16.72, |
|
"grad_norm": 0.8923472762107849, |
|
"learning_rate": 1.795175051485731e-05, |
|
"loss": 2.893, |
|
"step": 311000 |
|
}, |
|
{ |
|
"epoch": 16.78, |
|
"grad_norm": 0.9239156246185303, |
|
"learning_rate": 1.765784054133569e-05, |
|
"loss": 2.8919, |
|
"step": 312000 |
|
}, |
|
{ |
|
"epoch": 16.83, |
|
"grad_norm": 0.9343294501304626, |
|
"learning_rate": 1.7363636363636366e-05, |
|
"loss": 2.8876, |
|
"step": 313000 |
|
}, |
|
{ |
|
"epoch": 16.89, |
|
"grad_norm": 0.9064814448356628, |
|
"learning_rate": 1.706972639011474e-05, |
|
"loss": 2.8913, |
|
"step": 314000 |
|
}, |
|
{ |
|
"epoch": 16.94, |
|
"grad_norm": 0.9277246594429016, |
|
"learning_rate": 1.6775522212415418e-05, |
|
"loss": 2.8909, |
|
"step": 315000 |
|
}, |
|
{ |
|
"epoch": 16.99, |
|
"grad_norm": 0.9258543848991394, |
|
"learning_rate": 1.6481612238893792e-05, |
|
"loss": 2.8964, |
|
"step": 316000 |
|
}, |
|
{ |
|
"epoch": 17.0, |
|
"eval_accuracy": 0.40512663616869843, |
|
"eval_loss": 3.422145366668701, |
|
"eval_runtime": 153.2527, |
|
"eval_samples_per_second": 377.925, |
|
"eval_steps_per_second": 5.905, |
|
"step": 316115 |
|
}, |
|
{ |
|
"epoch": 17.05, |
|
"grad_norm": 0.9131640195846558, |
|
"learning_rate": 1.6187702265372166e-05, |
|
"loss": 2.864, |
|
"step": 317000 |
|
}, |
|
{ |
|
"epoch": 17.1, |
|
"grad_norm": 1.1945741176605225, |
|
"learning_rate": 1.5893498087672844e-05, |
|
"loss": 2.8617, |
|
"step": 318000 |
|
}, |
|
{ |
|
"epoch": 17.16, |
|
"grad_norm": 0.9226692914962769, |
|
"learning_rate": 1.5599588114151225e-05, |
|
"loss": 2.8644, |
|
"step": 319000 |
|
}, |
|
{ |
|
"epoch": 17.21, |
|
"grad_norm": 0.9040902256965637, |
|
"learning_rate": 1.53053839364519e-05, |
|
"loss": 2.8643, |
|
"step": 320000 |
|
}, |
|
{ |
|
"epoch": 17.26, |
|
"grad_norm": 0.9265015721321106, |
|
"learning_rate": 1.5011179758752575e-05, |
|
"loss": 2.8684, |
|
"step": 321000 |
|
}, |
|
{ |
|
"epoch": 17.32, |
|
"grad_norm": 0.940202534198761, |
|
"learning_rate": 1.4716975581053253e-05, |
|
"loss": 2.8653, |
|
"step": 322000 |
|
}, |
|
{ |
|
"epoch": 17.37, |
|
"grad_norm": 0.9231876730918884, |
|
"learning_rate": 1.4423065607531627e-05, |
|
"loss": 2.8729, |
|
"step": 323000 |
|
}, |
|
{ |
|
"epoch": 17.42, |
|
"grad_norm": 0.9177024960517883, |
|
"learning_rate": 1.4128861429832305e-05, |
|
"loss": 2.8688, |
|
"step": 324000 |
|
}, |
|
{ |
|
"epoch": 17.48, |
|
"grad_norm": 0.9014251828193665, |
|
"learning_rate": 1.3834657252132983e-05, |
|
"loss": 2.8711, |
|
"step": 325000 |
|
}, |
|
{ |
|
"epoch": 17.53, |
|
"grad_norm": 0.9649085402488708, |
|
"learning_rate": 1.3540453074433657e-05, |
|
"loss": 2.8721, |
|
"step": 326000 |
|
}, |
|
{ |
|
"epoch": 17.59, |
|
"grad_norm": 0.9398193955421448, |
|
"learning_rate": 1.3246543100912034e-05, |
|
"loss": 2.8725, |
|
"step": 327000 |
|
}, |
|
{ |
|
"epoch": 17.64, |
|
"grad_norm": 0.9121034145355225, |
|
"learning_rate": 1.2952338923212709e-05, |
|
"loss": 2.8771, |
|
"step": 328000 |
|
}, |
|
{ |
|
"epoch": 17.69, |
|
"grad_norm": 0.8918709754943848, |
|
"learning_rate": 1.2658134745513387e-05, |
|
"loss": 2.8721, |
|
"step": 329000 |
|
}, |
|
{ |
|
"epoch": 17.75, |
|
"grad_norm": 0.8925199508666992, |
|
"learning_rate": 1.2363930567814063e-05, |
|
"loss": 2.8722, |
|
"step": 330000 |
|
}, |
|
{ |
|
"epoch": 17.8, |
|
"grad_norm": 0.9330604076385498, |
|
"learning_rate": 1.207031479847014e-05, |
|
"loss": 2.8763, |
|
"step": 331000 |
|
}, |
|
{ |
|
"epoch": 17.85, |
|
"grad_norm": 0.9173546433448792, |
|
"learning_rate": 1.1776110620770816e-05, |
|
"loss": 2.8743, |
|
"step": 332000 |
|
}, |
|
{ |
|
"epoch": 17.91, |
|
"grad_norm": 0.9605908393859863, |
|
"learning_rate": 1.1481906443071492e-05, |
|
"loss": 2.8716, |
|
"step": 333000 |
|
}, |
|
{ |
|
"epoch": 17.96, |
|
"grad_norm": 0.9416488409042358, |
|
"learning_rate": 1.1187702265372168e-05, |
|
"loss": 2.8758, |
|
"step": 334000 |
|
}, |
|
{ |
|
"epoch": 18.0, |
|
"eval_accuracy": 0.4054102116463327, |
|
"eval_loss": 3.428605079650879, |
|
"eval_runtime": 153.3009, |
|
"eval_samples_per_second": 377.806, |
|
"eval_steps_per_second": 5.903, |
|
"step": 334710 |
|
}, |
|
{ |
|
"epoch": 18.02, |
|
"grad_norm": 0.9360882639884949, |
|
"learning_rate": 1.0893498087672846e-05, |
|
"loss": 2.8649, |
|
"step": 335000 |
|
}, |
|
{ |
|
"epoch": 18.07, |
|
"grad_norm": 0.9502159357070923, |
|
"learning_rate": 1.0599588114151222e-05, |
|
"loss": 2.8497, |
|
"step": 336000 |
|
}, |
|
{ |
|
"epoch": 18.12, |
|
"grad_norm": 0.9194748997688293, |
|
"learning_rate": 1.0305383936451898e-05, |
|
"loss": 2.8492, |
|
"step": 337000 |
|
}, |
|
{ |
|
"epoch": 18.18, |
|
"grad_norm": 0.9241055250167847, |
|
"learning_rate": 1.0011473962930273e-05, |
|
"loss": 2.8539, |
|
"step": 338000 |
|
}, |
|
{ |
|
"epoch": 18.23, |
|
"grad_norm": 0.9047831296920776, |
|
"learning_rate": 9.717563989408649e-06, |
|
"loss": 2.8555, |
|
"step": 339000 |
|
}, |
|
{ |
|
"epoch": 18.28, |
|
"grad_norm": 0.9505220651626587, |
|
"learning_rate": 9.423359811709327e-06, |
|
"loss": 2.8542, |
|
"step": 340000 |
|
}, |
|
{ |
|
"epoch": 18.34, |
|
"grad_norm": 0.9681676030158997, |
|
"learning_rate": 9.129155634010003e-06, |
|
"loss": 2.8526, |
|
"step": 341000 |
|
}, |
|
{ |
|
"epoch": 18.39, |
|
"grad_norm": 0.9679902791976929, |
|
"learning_rate": 8.834951456310681e-06, |
|
"loss": 2.8543, |
|
"step": 342000 |
|
}, |
|
{ |
|
"epoch": 18.45, |
|
"grad_norm": 0.9757332801818848, |
|
"learning_rate": 8.540747278611357e-06, |
|
"loss": 2.8569, |
|
"step": 343000 |
|
}, |
|
{ |
|
"epoch": 18.5, |
|
"grad_norm": 0.9353688359260559, |
|
"learning_rate": 8.246837305089733e-06, |
|
"loss": 2.8563, |
|
"step": 344000 |
|
}, |
|
{ |
|
"epoch": 18.55, |
|
"grad_norm": 0.9887840747833252, |
|
"learning_rate": 7.952633127390409e-06, |
|
"loss": 2.8577, |
|
"step": 345000 |
|
}, |
|
{ |
|
"epoch": 18.61, |
|
"grad_norm": 0.9348433613777161, |
|
"learning_rate": 7.658723153868786e-06, |
|
"loss": 2.8606, |
|
"step": 346000 |
|
}, |
|
{ |
|
"epoch": 18.66, |
|
"grad_norm": 0.9570315480232239, |
|
"learning_rate": 7.364518976169461e-06, |
|
"loss": 2.8603, |
|
"step": 347000 |
|
}, |
|
{ |
|
"epoch": 18.71, |
|
"grad_norm": 0.9226345419883728, |
|
"learning_rate": 7.0703147984701384e-06, |
|
"loss": 2.8586, |
|
"step": 348000 |
|
}, |
|
{ |
|
"epoch": 18.77, |
|
"grad_norm": 0.9210349321365356, |
|
"learning_rate": 6.776110620770815e-06, |
|
"loss": 2.8566, |
|
"step": 349000 |
|
}, |
|
{ |
|
"epoch": 18.82, |
|
"grad_norm": 0.9562284350395203, |
|
"learning_rate": 6.482200647249191e-06, |
|
"loss": 2.8609, |
|
"step": 350000 |
|
}, |
|
{ |
|
"epoch": 18.88, |
|
"grad_norm": 0.9210054874420166, |
|
"learning_rate": 6.187996469549868e-06, |
|
"loss": 2.8607, |
|
"step": 351000 |
|
}, |
|
{ |
|
"epoch": 18.93, |
|
"grad_norm": 0.9457345008850098, |
|
"learning_rate": 5.893792291850544e-06, |
|
"loss": 2.8598, |
|
"step": 352000 |
|
}, |
|
{ |
|
"epoch": 18.98, |
|
"grad_norm": 0.9460088610649109, |
|
"learning_rate": 5.59988231832892e-06, |
|
"loss": 2.858, |
|
"step": 353000 |
|
}, |
|
{ |
|
"epoch": 19.0, |
|
"eval_accuracy": 0.4053840106427133, |
|
"eval_loss": 3.426475763320923, |
|
"eval_runtime": 153.612, |
|
"eval_samples_per_second": 377.041, |
|
"eval_steps_per_second": 5.891, |
|
"step": 353305 |
|
}, |
|
{ |
|
"epoch": 19.04, |
|
"grad_norm": 0.9623994827270508, |
|
"learning_rate": 5.305678140629597e-06, |
|
"loss": 2.8452, |
|
"step": 354000 |
|
}, |
|
{ |
|
"epoch": 19.09, |
|
"grad_norm": 0.9756667613983154, |
|
"learning_rate": 5.011473962930274e-06, |
|
"loss": 2.842, |
|
"step": 355000 |
|
}, |
|
{ |
|
"epoch": 19.14, |
|
"grad_norm": 0.9565538763999939, |
|
"learning_rate": 4.71726978523095e-06, |
|
"loss": 2.8394, |
|
"step": 356000 |
|
}, |
|
{ |
|
"epoch": 19.2, |
|
"grad_norm": 0.9308847784996033, |
|
"learning_rate": 4.4233598117093264e-06, |
|
"loss": 2.8404, |
|
"step": 357000 |
|
}, |
|
{ |
|
"epoch": 19.25, |
|
"grad_norm": 0.9186875224113464, |
|
"learning_rate": 4.1291556340100035e-06, |
|
"loss": 2.8446, |
|
"step": 358000 |
|
}, |
|
{ |
|
"epoch": 19.31, |
|
"grad_norm": 0.9542786478996277, |
|
"learning_rate": 3.83495145631068e-06, |
|
"loss": 2.8453, |
|
"step": 359000 |
|
}, |
|
{ |
|
"epoch": 19.36, |
|
"grad_norm": 0.9696632027626038, |
|
"learning_rate": 3.5410414827890556e-06, |
|
"loss": 2.8454, |
|
"step": 360000 |
|
}, |
|
{ |
|
"epoch": 19.41, |
|
"grad_norm": 0.9518758058547974, |
|
"learning_rate": 3.2471315092674317e-06, |
|
"loss": 2.8447, |
|
"step": 361000 |
|
}, |
|
{ |
|
"epoch": 19.47, |
|
"grad_norm": 0.9522430896759033, |
|
"learning_rate": 2.9529273315681087e-06, |
|
"loss": 2.8461, |
|
"step": 362000 |
|
}, |
|
{ |
|
"epoch": 19.52, |
|
"grad_norm": 0.9725024700164795, |
|
"learning_rate": 2.6587231538687853e-06, |
|
"loss": 2.8434, |
|
"step": 363000 |
|
}, |
|
{ |
|
"epoch": 19.58, |
|
"grad_norm": 0.9441395401954651, |
|
"learning_rate": 2.364518976169462e-06, |
|
"loss": 2.8441, |
|
"step": 364000 |
|
}, |
|
{ |
|
"epoch": 19.63, |
|
"grad_norm": 0.9425933957099915, |
|
"learning_rate": 2.0706090026478374e-06, |
|
"loss": 2.8438, |
|
"step": 365000 |
|
}, |
|
{ |
|
"epoch": 19.68, |
|
"grad_norm": 0.9448084831237793, |
|
"learning_rate": 1.7764048249485142e-06, |
|
"loss": 2.8458, |
|
"step": 366000 |
|
}, |
|
{ |
|
"epoch": 19.74, |
|
"grad_norm": 0.9476189613342285, |
|
"learning_rate": 1.4824948514268903e-06, |
|
"loss": 2.8462, |
|
"step": 367000 |
|
}, |
|
{ |
|
"epoch": 19.79, |
|
"grad_norm": 0.9310097694396973, |
|
"learning_rate": 1.188290673727567e-06, |
|
"loss": 2.8437, |
|
"step": 368000 |
|
}, |
|
{ |
|
"epoch": 19.84, |
|
"grad_norm": 0.9563276171684265, |
|
"learning_rate": 8.940864960282438e-07, |
|
"loss": 2.8462, |
|
"step": 369000 |
|
}, |
|
{ |
|
"epoch": 19.9, |
|
"grad_norm": 0.9808979630470276, |
|
"learning_rate": 5.998823183289204e-07, |
|
"loss": 2.8412, |
|
"step": 370000 |
|
}, |
|
{ |
|
"epoch": 19.95, |
|
"grad_norm": 0.9566290378570557, |
|
"learning_rate": 3.0597234480729626e-07, |
|
"loss": 2.8493, |
|
"step": 371000 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"eval_accuracy": 0.40547080986496004, |
|
"eval_loss": 3.432330846786499, |
|
"eval_runtime": 154.1, |
|
"eval_samples_per_second": 375.847, |
|
"eval_steps_per_second": 5.873, |
|
"step": 371900 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"step": 371900, |
|
"total_flos": 1.56693365388288e+18, |
|
"train_loss": 3.1561156885198893, |
|
"train_runtime": 81199.8655, |
|
"train_samples_per_second": 146.561, |
|
"train_steps_per_second": 4.58 |
|
} |
|
], |
|
"logging_steps": 1000, |
|
"max_steps": 371900, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 20, |
|
"save_steps": 5000, |
|
"total_flos": 1.56693365388288e+18, |
|
"train_batch_size": 32, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|