|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 20.0, |
|
"eval_steps": 500, |
|
"global_step": 371840, |
|
"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": 6.2298, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 6.25e-05, |
|
"loss": 5.0094, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"learning_rate": 9.375e-05, |
|
"loss": 4.6802, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"learning_rate": 0.000125, |
|
"loss": 4.4542, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 0.00015625, |
|
"loss": 4.2925, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 0.0001875, |
|
"loss": 4.1748, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"learning_rate": 0.00021875, |
|
"loss": 4.0701, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"learning_rate": 0.00025, |
|
"loss": 3.9851, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"learning_rate": 0.00028121875, |
|
"loss": 3.913, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"learning_rate": 0.00031246875000000003, |
|
"loss": 3.8497, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 0.59, |
|
"learning_rate": 0.00034368749999999997, |
|
"loss": 3.8077, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 0.65, |
|
"learning_rate": 0.0003749375, |
|
"loss": 3.7627, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"learning_rate": 0.00040615625, |
|
"loss": 3.7397, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"learning_rate": 0.00043737500000000005, |
|
"loss": 3.701, |
|
"step": 14000 |
|
}, |
|
{ |
|
"epoch": 0.81, |
|
"learning_rate": 0.000468625, |
|
"loss": 3.6693, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"learning_rate": 0.00049984375, |
|
"loss": 3.6554, |
|
"step": 16000 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"learning_rate": 0.00053109375, |
|
"loss": 3.6301, |
|
"step": 17000 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"learning_rate": 0.00056234375, |
|
"loss": 3.6072, |
|
"step": 18000 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.3577459544837306, |
|
"eval_loss": 3.797168016433716, |
|
"eval_runtime": 152.3829, |
|
"eval_samples_per_second": 380.089, |
|
"eval_steps_per_second": 5.939, |
|
"step": 18592 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"learning_rate": 0.0005935625, |
|
"loss": 3.5799, |
|
"step": 19000 |
|
}, |
|
{ |
|
"epoch": 1.08, |
|
"learning_rate": 0.0006248125, |
|
"loss": 3.5485, |
|
"step": 20000 |
|
}, |
|
{ |
|
"epoch": 1.13, |
|
"learning_rate": 0.0006560625, |
|
"loss": 3.5406, |
|
"step": 21000 |
|
}, |
|
{ |
|
"epoch": 1.18, |
|
"learning_rate": 0.00068728125, |
|
"loss": 3.5264, |
|
"step": 22000 |
|
}, |
|
{ |
|
"epoch": 1.24, |
|
"learning_rate": 0.00071853125, |
|
"loss": 3.518, |
|
"step": 23000 |
|
}, |
|
{ |
|
"epoch": 1.29, |
|
"learning_rate": 0.00074978125, |
|
"loss": 3.5133, |
|
"step": 24000 |
|
}, |
|
{ |
|
"epoch": 1.34, |
|
"learning_rate": 0.0007810312499999999, |
|
"loss": 3.4955, |
|
"step": 25000 |
|
}, |
|
{ |
|
"epoch": 1.4, |
|
"learning_rate": 0.00081225, |
|
"loss": 3.4875, |
|
"step": 26000 |
|
}, |
|
{ |
|
"epoch": 1.45, |
|
"learning_rate": 0.0008435000000000001, |
|
"loss": 3.4788, |
|
"step": 27000 |
|
}, |
|
{ |
|
"epoch": 1.51, |
|
"learning_rate": 0.00087471875, |
|
"loss": 3.4699, |
|
"step": 28000 |
|
}, |
|
{ |
|
"epoch": 1.56, |
|
"learning_rate": 0.00090596875, |
|
"loss": 3.4636, |
|
"step": 29000 |
|
}, |
|
{ |
|
"epoch": 1.61, |
|
"learning_rate": 0.00093721875, |
|
"loss": 3.4553, |
|
"step": 30000 |
|
}, |
|
{ |
|
"epoch": 1.67, |
|
"learning_rate": 0.0009684375, |
|
"loss": 3.4444, |
|
"step": 31000 |
|
}, |
|
{ |
|
"epoch": 1.72, |
|
"learning_rate": 0.0009996875, |
|
"loss": 3.437, |
|
"step": 32000 |
|
}, |
|
{ |
|
"epoch": 1.77, |
|
"learning_rate": 0.0009970868644067796, |
|
"loss": 3.4315, |
|
"step": 33000 |
|
}, |
|
{ |
|
"epoch": 1.83, |
|
"learning_rate": 0.0009941443032015066, |
|
"loss": 3.4193, |
|
"step": 34000 |
|
}, |
|
{ |
|
"epoch": 1.88, |
|
"learning_rate": 0.0009912017419962335, |
|
"loss": 3.4093, |
|
"step": 35000 |
|
}, |
|
{ |
|
"epoch": 1.94, |
|
"learning_rate": 0.0009882621233521657, |
|
"loss": 3.3989, |
|
"step": 36000 |
|
}, |
|
{ |
|
"epoch": 1.99, |
|
"learning_rate": 0.0009853225047080979, |
|
"loss": 3.3863, |
|
"step": 37000 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.38044785878682175, |
|
"eval_loss": 3.5804407596588135, |
|
"eval_runtime": 153.7328, |
|
"eval_samples_per_second": 376.751, |
|
"eval_steps_per_second": 5.887, |
|
"step": 37184 |
|
}, |
|
{ |
|
"epoch": 2.04, |
|
"learning_rate": 0.000982379943502825, |
|
"loss": 3.3396, |
|
"step": 38000 |
|
}, |
|
{ |
|
"epoch": 2.1, |
|
"learning_rate": 0.0009794373822975518, |
|
"loss": 3.328, |
|
"step": 39000 |
|
}, |
|
{ |
|
"epoch": 2.15, |
|
"learning_rate": 0.000976497763653484, |
|
"loss": 3.3247, |
|
"step": 40000 |
|
}, |
|
{ |
|
"epoch": 2.21, |
|
"learning_rate": 0.0009735552024482109, |
|
"loss": 3.3252, |
|
"step": 41000 |
|
}, |
|
{ |
|
"epoch": 2.26, |
|
"learning_rate": 0.0009706155838041431, |
|
"loss": 3.3166, |
|
"step": 42000 |
|
}, |
|
{ |
|
"epoch": 2.31, |
|
"learning_rate": 0.00096767302259887, |
|
"loss": 3.3188, |
|
"step": 43000 |
|
}, |
|
{ |
|
"epoch": 2.37, |
|
"learning_rate": 0.0009647334039548023, |
|
"loss": 3.3129, |
|
"step": 44000 |
|
}, |
|
{ |
|
"epoch": 2.42, |
|
"learning_rate": 0.0009617908427495293, |
|
"loss": 3.304, |
|
"step": 45000 |
|
}, |
|
{ |
|
"epoch": 2.47, |
|
"learning_rate": 0.0009588482815442562, |
|
"loss": 3.2975, |
|
"step": 46000 |
|
}, |
|
{ |
|
"epoch": 2.53, |
|
"learning_rate": 0.0009559086629001883, |
|
"loss": 3.2928, |
|
"step": 47000 |
|
}, |
|
{ |
|
"epoch": 2.58, |
|
"learning_rate": 0.0009529661016949153, |
|
"loss": 3.2871, |
|
"step": 48000 |
|
}, |
|
{ |
|
"epoch": 2.64, |
|
"learning_rate": 0.0009500235404896422, |
|
"loss": 3.2821, |
|
"step": 49000 |
|
}, |
|
{ |
|
"epoch": 2.69, |
|
"learning_rate": 0.0009470809792843691, |
|
"loss": 3.2765, |
|
"step": 50000 |
|
}, |
|
{ |
|
"epoch": 2.74, |
|
"learning_rate": 0.0009441413606403013, |
|
"loss": 3.2783, |
|
"step": 51000 |
|
}, |
|
{ |
|
"epoch": 2.8, |
|
"learning_rate": 0.0009411987994350282, |
|
"loss": 3.2737, |
|
"step": 52000 |
|
}, |
|
{ |
|
"epoch": 2.85, |
|
"learning_rate": 0.0009382591807909605, |
|
"loss": 3.2713, |
|
"step": 53000 |
|
}, |
|
{ |
|
"epoch": 2.9, |
|
"learning_rate": 0.0009353166195856874, |
|
"loss": 3.2633, |
|
"step": 54000 |
|
}, |
|
{ |
|
"epoch": 2.96, |
|
"learning_rate": 0.0009323770009416196, |
|
"loss": 3.2556, |
|
"step": 55000 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.3910322096812649, |
|
"eval_loss": 3.472982168197632, |
|
"eval_runtime": 154.4664, |
|
"eval_samples_per_second": 374.962, |
|
"eval_steps_per_second": 5.859, |
|
"step": 55776 |
|
}, |
|
{ |
|
"epoch": 3.01, |
|
"learning_rate": 0.0009294344397363465, |
|
"loss": 3.2424, |
|
"step": 56000 |
|
}, |
|
{ |
|
"epoch": 3.07, |
|
"learning_rate": 0.0009264918785310734, |
|
"loss": 3.1944, |
|
"step": 57000 |
|
}, |
|
{ |
|
"epoch": 3.12, |
|
"learning_rate": 0.0009235493173258004, |
|
"loss": 3.1947, |
|
"step": 58000 |
|
}, |
|
{ |
|
"epoch": 3.17, |
|
"learning_rate": 0.0009206096986817327, |
|
"loss": 3.1979, |
|
"step": 59000 |
|
}, |
|
{ |
|
"epoch": 3.23, |
|
"learning_rate": 0.0009176671374764596, |
|
"loss": 3.1974, |
|
"step": 60000 |
|
}, |
|
{ |
|
"epoch": 3.28, |
|
"learning_rate": 0.0009147245762711865, |
|
"loss": 3.1978, |
|
"step": 61000 |
|
}, |
|
{ |
|
"epoch": 3.33, |
|
"learning_rate": 0.0009117849576271187, |
|
"loss": 3.196, |
|
"step": 62000 |
|
}, |
|
{ |
|
"epoch": 3.39, |
|
"learning_rate": 0.0009088423964218456, |
|
"loss": 3.1975, |
|
"step": 63000 |
|
}, |
|
{ |
|
"epoch": 3.44, |
|
"learning_rate": 0.0009059057203389832, |
|
"loss": 3.1989, |
|
"step": 64000 |
|
}, |
|
{ |
|
"epoch": 3.5, |
|
"learning_rate": 0.00090296315913371, |
|
"loss": 3.2003, |
|
"step": 65000 |
|
}, |
|
{ |
|
"epoch": 3.55, |
|
"learning_rate": 0.000900020597928437, |
|
"loss": 3.1914, |
|
"step": 66000 |
|
}, |
|
{ |
|
"epoch": 3.6, |
|
"learning_rate": 0.0008970809792843692, |
|
"loss": 3.1914, |
|
"step": 67000 |
|
}, |
|
{ |
|
"epoch": 3.66, |
|
"learning_rate": 0.0008941413606403014, |
|
"loss": 3.1919, |
|
"step": 68000 |
|
}, |
|
{ |
|
"epoch": 3.71, |
|
"learning_rate": 0.0008911987994350282, |
|
"loss": 3.1876, |
|
"step": 69000 |
|
}, |
|
{ |
|
"epoch": 3.77, |
|
"learning_rate": 0.0008882562382297552, |
|
"loss": 3.1891, |
|
"step": 70000 |
|
}, |
|
{ |
|
"epoch": 3.82, |
|
"learning_rate": 0.0008853136770244821, |
|
"loss": 3.1853, |
|
"step": 71000 |
|
}, |
|
{ |
|
"epoch": 3.87, |
|
"learning_rate": 0.000882371115819209, |
|
"loss": 3.184, |
|
"step": 72000 |
|
}, |
|
{ |
|
"epoch": 3.93, |
|
"learning_rate": 0.000879428554613936, |
|
"loss": 3.1845, |
|
"step": 73000 |
|
}, |
|
{ |
|
"epoch": 3.98, |
|
"learning_rate": 0.0008764889359698683, |
|
"loss": 3.1829, |
|
"step": 74000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.3992117530567142, |
|
"eval_loss": 3.4019179344177246, |
|
"eval_runtime": 153.2776, |
|
"eval_samples_per_second": 377.87, |
|
"eval_steps_per_second": 5.904, |
|
"step": 74368 |
|
}, |
|
{ |
|
"epoch": 4.03, |
|
"learning_rate": 0.0008735463747645951, |
|
"loss": 3.1352, |
|
"step": 75000 |
|
}, |
|
{ |
|
"epoch": 4.09, |
|
"learning_rate": 0.0008706067561205273, |
|
"loss": 3.1132, |
|
"step": 76000 |
|
}, |
|
{ |
|
"epoch": 4.14, |
|
"learning_rate": 0.0008676641949152543, |
|
"loss": 3.1218, |
|
"step": 77000 |
|
}, |
|
{ |
|
"epoch": 4.2, |
|
"learning_rate": 0.0008647216337099811, |
|
"loss": 3.1223, |
|
"step": 78000 |
|
}, |
|
{ |
|
"epoch": 4.25, |
|
"learning_rate": 0.0008617790725047081, |
|
"loss": 3.1286, |
|
"step": 79000 |
|
}, |
|
{ |
|
"epoch": 4.3, |
|
"learning_rate": 0.0008588394538606403, |
|
"loss": 3.1235, |
|
"step": 80000 |
|
}, |
|
{ |
|
"epoch": 4.36, |
|
"learning_rate": 0.0008558998352165726, |
|
"loss": 3.1243, |
|
"step": 81000 |
|
}, |
|
{ |
|
"epoch": 4.41, |
|
"learning_rate": 0.0008529572740112994, |
|
"loss": 3.13, |
|
"step": 82000 |
|
}, |
|
{ |
|
"epoch": 4.46, |
|
"learning_rate": 0.0008500147128060264, |
|
"loss": 3.1321, |
|
"step": 83000 |
|
}, |
|
{ |
|
"epoch": 4.52, |
|
"learning_rate": 0.0008470750941619586, |
|
"loss": 3.1309, |
|
"step": 84000 |
|
}, |
|
{ |
|
"epoch": 4.57, |
|
"learning_rate": 0.0008441325329566854, |
|
"loss": 3.132, |
|
"step": 85000 |
|
}, |
|
{ |
|
"epoch": 4.63, |
|
"learning_rate": 0.0008411899717514124, |
|
"loss": 3.1335, |
|
"step": 86000 |
|
}, |
|
{ |
|
"epoch": 4.68, |
|
"learning_rate": 0.0008382474105461394, |
|
"loss": 3.1249, |
|
"step": 87000 |
|
}, |
|
{ |
|
"epoch": 4.73, |
|
"learning_rate": 0.0008353077919020717, |
|
"loss": 3.125, |
|
"step": 88000 |
|
}, |
|
{ |
|
"epoch": 4.79, |
|
"learning_rate": 0.0008323652306967985, |
|
"loss": 3.13, |
|
"step": 89000 |
|
}, |
|
{ |
|
"epoch": 4.84, |
|
"learning_rate": 0.0008294226694915255, |
|
"loss": 3.1299, |
|
"step": 90000 |
|
}, |
|
{ |
|
"epoch": 4.89, |
|
"learning_rate": 0.0008264801082862525, |
|
"loss": 3.1265, |
|
"step": 91000 |
|
}, |
|
{ |
|
"epoch": 4.95, |
|
"learning_rate": 0.0008235404896421845, |
|
"loss": 3.1264, |
|
"step": 92000 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.4020474588723565, |
|
"eval_loss": 3.382812976837158, |
|
"eval_runtime": 153.1227, |
|
"eval_samples_per_second": 378.252, |
|
"eval_steps_per_second": 5.91, |
|
"step": 92960 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"learning_rate": 0.0008205979284369115, |
|
"loss": 3.1204, |
|
"step": 93000 |
|
}, |
|
{ |
|
"epoch": 5.06, |
|
"learning_rate": 0.0008176553672316385, |
|
"loss": 3.0543, |
|
"step": 94000 |
|
}, |
|
{ |
|
"epoch": 5.11, |
|
"learning_rate": 0.0008147157485875707, |
|
"loss": 3.0628, |
|
"step": 95000 |
|
}, |
|
{ |
|
"epoch": 5.16, |
|
"learning_rate": 0.0008117761299435028, |
|
"loss": 3.067, |
|
"step": 96000 |
|
}, |
|
{ |
|
"epoch": 5.22, |
|
"learning_rate": 0.0008088335687382298, |
|
"loss": 3.0676, |
|
"step": 97000 |
|
}, |
|
{ |
|
"epoch": 5.27, |
|
"learning_rate": 0.0008058910075329568, |
|
"loss": 3.0752, |
|
"step": 98000 |
|
}, |
|
{ |
|
"epoch": 5.32, |
|
"learning_rate": 0.0008029484463276836, |
|
"loss": 3.0826, |
|
"step": 99000 |
|
}, |
|
{ |
|
"epoch": 5.38, |
|
"learning_rate": 0.0008000088276836158, |
|
"loss": 3.0768, |
|
"step": 100000 |
|
}, |
|
{ |
|
"epoch": 5.43, |
|
"learning_rate": 0.0007970662664783428, |
|
"loss": 3.0752, |
|
"step": 101000 |
|
}, |
|
{ |
|
"epoch": 5.49, |
|
"learning_rate": 0.000794126647834275, |
|
"loss": 3.0835, |
|
"step": 102000 |
|
}, |
|
{ |
|
"epoch": 5.54, |
|
"learning_rate": 0.0007911840866290018, |
|
"loss": 3.0767, |
|
"step": 103000 |
|
}, |
|
{ |
|
"epoch": 5.59, |
|
"learning_rate": 0.0007882415254237288, |
|
"loss": 3.0816, |
|
"step": 104000 |
|
}, |
|
{ |
|
"epoch": 5.65, |
|
"learning_rate": 0.0007852989642184559, |
|
"loss": 3.0783, |
|
"step": 105000 |
|
}, |
|
{ |
|
"epoch": 5.7, |
|
"learning_rate": 0.0007823564030131827, |
|
"loss": 3.0838, |
|
"step": 106000 |
|
}, |
|
{ |
|
"epoch": 5.76, |
|
"learning_rate": 0.0007794138418079097, |
|
"loss": 3.0834, |
|
"step": 107000 |
|
}, |
|
{ |
|
"epoch": 5.81, |
|
"learning_rate": 0.0007764712806026365, |
|
"loss": 3.0844, |
|
"step": 108000 |
|
}, |
|
{ |
|
"epoch": 5.86, |
|
"learning_rate": 0.0007735316619585687, |
|
"loss": 3.0811, |
|
"step": 109000 |
|
}, |
|
{ |
|
"epoch": 5.92, |
|
"learning_rate": 0.0007705891007532957, |
|
"loss": 3.0826, |
|
"step": 110000 |
|
}, |
|
{ |
|
"epoch": 5.97, |
|
"learning_rate": 0.0007676465395480227, |
|
"loss": 3.0827, |
|
"step": 111000 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.40310327390667616, |
|
"eval_loss": 3.3848841190338135, |
|
"eval_runtime": 153.1292, |
|
"eval_samples_per_second": 378.236, |
|
"eval_steps_per_second": 5.91, |
|
"step": 111552 |
|
}, |
|
{ |
|
"epoch": 6.02, |
|
"learning_rate": 0.0007647098634651602, |
|
"loss": 3.0519, |
|
"step": 112000 |
|
}, |
|
{ |
|
"epoch": 6.08, |
|
"learning_rate": 0.000761767302259887, |
|
"loss": 3.0168, |
|
"step": 113000 |
|
}, |
|
{ |
|
"epoch": 6.13, |
|
"learning_rate": 0.000758824741054614, |
|
"loss": 3.0242, |
|
"step": 114000 |
|
}, |
|
{ |
|
"epoch": 6.19, |
|
"learning_rate": 0.000755882179849341, |
|
"loss": 3.0258, |
|
"step": 115000 |
|
}, |
|
{ |
|
"epoch": 6.24, |
|
"learning_rate": 0.0007529396186440678, |
|
"loss": 3.0341, |
|
"step": 116000 |
|
}, |
|
{ |
|
"epoch": 6.29, |
|
"learning_rate": 0.00075, |
|
"loss": 3.0319, |
|
"step": 117000 |
|
}, |
|
{ |
|
"epoch": 6.35, |
|
"learning_rate": 0.000747057438794727, |
|
"loss": 3.0359, |
|
"step": 118000 |
|
}, |
|
{ |
|
"epoch": 6.4, |
|
"learning_rate": 0.0007441178201506591, |
|
"loss": 3.0377, |
|
"step": 119000 |
|
}, |
|
{ |
|
"epoch": 6.45, |
|
"learning_rate": 0.000741175258945386, |
|
"loss": 3.0378, |
|
"step": 120000 |
|
}, |
|
{ |
|
"epoch": 6.51, |
|
"learning_rate": 0.000738232697740113, |
|
"loss": 3.0402, |
|
"step": 121000 |
|
}, |
|
{ |
|
"epoch": 6.56, |
|
"learning_rate": 0.0007352901365348398, |
|
"loss": 3.046, |
|
"step": 122000 |
|
}, |
|
{ |
|
"epoch": 6.62, |
|
"learning_rate": 0.0007323505178907721, |
|
"loss": 3.0444, |
|
"step": 123000 |
|
}, |
|
{ |
|
"epoch": 6.67, |
|
"learning_rate": 0.0007294079566854991, |
|
"loss": 3.0424, |
|
"step": 124000 |
|
}, |
|
{ |
|
"epoch": 6.72, |
|
"learning_rate": 0.0007264712806026365, |
|
"loss": 3.0428, |
|
"step": 125000 |
|
}, |
|
{ |
|
"epoch": 6.78, |
|
"learning_rate": 0.0007235316619585688, |
|
"loss": 3.0425, |
|
"step": 126000 |
|
}, |
|
{ |
|
"epoch": 6.83, |
|
"learning_rate": 0.0007205891007532957, |
|
"loss": 3.0448, |
|
"step": 127000 |
|
}, |
|
{ |
|
"epoch": 6.88, |
|
"learning_rate": 0.0007176465395480226, |
|
"loss": 3.049, |
|
"step": 128000 |
|
}, |
|
{ |
|
"epoch": 6.94, |
|
"learning_rate": 0.0007147039783427496, |
|
"loss": 3.0425, |
|
"step": 129000 |
|
}, |
|
{ |
|
"epoch": 6.99, |
|
"learning_rate": 0.0007117614171374765, |
|
"loss": 3.0461, |
|
"step": 130000 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.4050249835692312, |
|
"eval_loss": 3.3727521896362305, |
|
"eval_runtime": 153.1303, |
|
"eval_samples_per_second": 378.233, |
|
"eval_steps_per_second": 5.91, |
|
"step": 130144 |
|
}, |
|
{ |
|
"epoch": 7.05, |
|
"learning_rate": 0.0007088188559322034, |
|
"loss": 2.9834, |
|
"step": 131000 |
|
}, |
|
{ |
|
"epoch": 7.1, |
|
"learning_rate": 0.0007058792372881356, |
|
"loss": 2.9854, |
|
"step": 132000 |
|
}, |
|
{ |
|
"epoch": 7.15, |
|
"learning_rate": 0.0007029366760828625, |
|
"loss": 2.9855, |
|
"step": 133000 |
|
}, |
|
{ |
|
"epoch": 7.21, |
|
"learning_rate": 0.0006999970574387948, |
|
"loss": 2.9948, |
|
"step": 134000 |
|
}, |
|
{ |
|
"epoch": 7.26, |
|
"learning_rate": 0.0006970544962335217, |
|
"loss": 2.9986, |
|
"step": 135000 |
|
}, |
|
{ |
|
"epoch": 7.31, |
|
"learning_rate": 0.0006941148775894539, |
|
"loss": 3.0007, |
|
"step": 136000 |
|
}, |
|
{ |
|
"epoch": 7.37, |
|
"learning_rate": 0.0006911723163841808, |
|
"loss": 3.0032, |
|
"step": 137000 |
|
}, |
|
{ |
|
"epoch": 7.42, |
|
"learning_rate": 0.0006882297551789077, |
|
"loss": 3.0065, |
|
"step": 138000 |
|
}, |
|
{ |
|
"epoch": 7.48, |
|
"learning_rate": 0.0006852871939736347, |
|
"loss": 3.0083, |
|
"step": 139000 |
|
}, |
|
{ |
|
"epoch": 7.53, |
|
"learning_rate": 0.0006823505178907722, |
|
"loss": 3.0086, |
|
"step": 140000 |
|
}, |
|
{ |
|
"epoch": 7.58, |
|
"learning_rate": 0.0006794079566854991, |
|
"loss": 3.0111, |
|
"step": 141000 |
|
}, |
|
{ |
|
"epoch": 7.64, |
|
"learning_rate": 0.000676465395480226, |
|
"loss": 3.0091, |
|
"step": 142000 |
|
}, |
|
{ |
|
"epoch": 7.69, |
|
"learning_rate": 0.000673522834274953, |
|
"loss": 3.0138, |
|
"step": 143000 |
|
}, |
|
{ |
|
"epoch": 7.75, |
|
"learning_rate": 0.0006705802730696799, |
|
"loss": 3.0099, |
|
"step": 144000 |
|
}, |
|
{ |
|
"epoch": 7.8, |
|
"learning_rate": 0.0006676406544256121, |
|
"loss": 3.0149, |
|
"step": 145000 |
|
}, |
|
{ |
|
"epoch": 7.85, |
|
"learning_rate": 0.0006647010357815442, |
|
"loss": 3.0104, |
|
"step": 146000 |
|
}, |
|
{ |
|
"epoch": 7.91, |
|
"learning_rate": 0.0006617584745762712, |
|
"loss": 3.0183, |
|
"step": 147000 |
|
}, |
|
{ |
|
"epoch": 7.96, |
|
"learning_rate": 0.0006588159133709981, |
|
"loss": 3.0111, |
|
"step": 148000 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.40690403335988545, |
|
"eval_loss": 3.3608620166778564, |
|
"eval_runtime": 153.1275, |
|
"eval_samples_per_second": 378.24, |
|
"eval_steps_per_second": 5.91, |
|
"step": 148736 |
|
}, |
|
{ |
|
"epoch": 8.01, |
|
"learning_rate": 0.0006558762947269304, |
|
"loss": 2.9952, |
|
"step": 149000 |
|
}, |
|
{ |
|
"epoch": 8.07, |
|
"learning_rate": 0.0006529337335216573, |
|
"loss": 2.9533, |
|
"step": 150000 |
|
}, |
|
{ |
|
"epoch": 8.12, |
|
"learning_rate": 0.0006499911723163842, |
|
"loss": 2.9567, |
|
"step": 151000 |
|
}, |
|
{ |
|
"epoch": 8.18, |
|
"learning_rate": 0.0006470515536723164, |
|
"loss": 2.9616, |
|
"step": 152000 |
|
}, |
|
{ |
|
"epoch": 8.23, |
|
"learning_rate": 0.0006441089924670433, |
|
"loss": 2.964, |
|
"step": 153000 |
|
}, |
|
{ |
|
"epoch": 8.28, |
|
"learning_rate": 0.0006411664312617702, |
|
"loss": 2.9719, |
|
"step": 154000 |
|
}, |
|
{ |
|
"epoch": 8.34, |
|
"learning_rate": 0.0006382238700564972, |
|
"loss": 2.97, |
|
"step": 155000 |
|
}, |
|
{ |
|
"epoch": 8.39, |
|
"learning_rate": 0.0006352813088512241, |
|
"loss": 2.9762, |
|
"step": 156000 |
|
}, |
|
{ |
|
"epoch": 8.44, |
|
"learning_rate": 0.0006323446327683616, |
|
"loss": 2.9736, |
|
"step": 157000 |
|
}, |
|
{ |
|
"epoch": 8.5, |
|
"learning_rate": 0.0006294020715630885, |
|
"loss": 2.983, |
|
"step": 158000 |
|
}, |
|
{ |
|
"epoch": 8.55, |
|
"learning_rate": 0.0006264624529190207, |
|
"loss": 2.9757, |
|
"step": 159000 |
|
}, |
|
{ |
|
"epoch": 8.61, |
|
"learning_rate": 0.0006235198917137477, |
|
"loss": 2.9828, |
|
"step": 160000 |
|
}, |
|
{ |
|
"epoch": 8.66, |
|
"learning_rate": 0.0006205773305084745, |
|
"loss": 2.9829, |
|
"step": 161000 |
|
}, |
|
{ |
|
"epoch": 8.71, |
|
"learning_rate": 0.0006176347693032015, |
|
"loss": 2.9784, |
|
"step": 162000 |
|
}, |
|
{ |
|
"epoch": 8.77, |
|
"learning_rate": 0.0006146922080979284, |
|
"loss": 2.9807, |
|
"step": 163000 |
|
}, |
|
{ |
|
"epoch": 8.82, |
|
"learning_rate": 0.0006117525894538607, |
|
"loss": 2.9858, |
|
"step": 164000 |
|
}, |
|
{ |
|
"epoch": 8.87, |
|
"learning_rate": 0.0006088100282485876, |
|
"loss": 2.9856, |
|
"step": 165000 |
|
}, |
|
{ |
|
"epoch": 8.93, |
|
"learning_rate": 0.0006058704096045198, |
|
"loss": 2.9875, |
|
"step": 166000 |
|
}, |
|
{ |
|
"epoch": 8.98, |
|
"learning_rate": 0.0006029278483992468, |
|
"loss": 2.9857, |
|
"step": 167000 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_accuracy": 0.4082261534843072, |
|
"eval_loss": 3.3496146202087402, |
|
"eval_runtime": 153.15, |
|
"eval_samples_per_second": 378.185, |
|
"eval_steps_per_second": 5.909, |
|
"step": 167328 |
|
}, |
|
{ |
|
"epoch": 9.04, |
|
"learning_rate": 0.0005999852871939736, |
|
"loss": 2.9408, |
|
"step": 168000 |
|
}, |
|
{ |
|
"epoch": 9.09, |
|
"learning_rate": 0.0005970427259887006, |
|
"loss": 2.9243, |
|
"step": 169000 |
|
}, |
|
{ |
|
"epoch": 9.14, |
|
"learning_rate": 0.0005941031073446328, |
|
"loss": 2.9328, |
|
"step": 170000 |
|
}, |
|
{ |
|
"epoch": 9.2, |
|
"learning_rate": 0.0005911634887005651, |
|
"loss": 2.9327, |
|
"step": 171000 |
|
}, |
|
{ |
|
"epoch": 9.25, |
|
"learning_rate": 0.0005882209274952919, |
|
"loss": 2.9438, |
|
"step": 172000 |
|
}, |
|
{ |
|
"epoch": 9.31, |
|
"learning_rate": 0.0005852783662900189, |
|
"loss": 2.9379, |
|
"step": 173000 |
|
}, |
|
{ |
|
"epoch": 9.36, |
|
"learning_rate": 0.0005823358050847458, |
|
"loss": 2.9497, |
|
"step": 174000 |
|
}, |
|
{ |
|
"epoch": 9.41, |
|
"learning_rate": 0.0005793961864406779, |
|
"loss": 2.9508, |
|
"step": 175000 |
|
}, |
|
{ |
|
"epoch": 9.47, |
|
"learning_rate": 0.0005764536252354049, |
|
"loss": 2.9503, |
|
"step": 176000 |
|
}, |
|
{ |
|
"epoch": 9.52, |
|
"learning_rate": 0.0005735110640301319, |
|
"loss": 2.9494, |
|
"step": 177000 |
|
}, |
|
{ |
|
"epoch": 9.57, |
|
"learning_rate": 0.0005705685028248587, |
|
"loss": 2.954, |
|
"step": 178000 |
|
}, |
|
{ |
|
"epoch": 9.63, |
|
"learning_rate": 0.0005676259416195857, |
|
"loss": 2.9555, |
|
"step": 179000 |
|
}, |
|
{ |
|
"epoch": 9.68, |
|
"learning_rate": 0.0005646863229755179, |
|
"loss": 2.9602, |
|
"step": 180000 |
|
}, |
|
{ |
|
"epoch": 9.74, |
|
"learning_rate": 0.0005617437617702449, |
|
"loss": 2.9581, |
|
"step": 181000 |
|
}, |
|
{ |
|
"epoch": 9.79, |
|
"learning_rate": 0.000558804143126177, |
|
"loss": 2.9585, |
|
"step": 182000 |
|
}, |
|
{ |
|
"epoch": 9.84, |
|
"learning_rate": 0.000555861581920904, |
|
"loss": 2.96, |
|
"step": 183000 |
|
}, |
|
{ |
|
"epoch": 9.9, |
|
"learning_rate": 0.000552919020715631, |
|
"loss": 2.9614, |
|
"step": 184000 |
|
}, |
|
{ |
|
"epoch": 9.95, |
|
"learning_rate": 0.000549979402071563, |
|
"loss": 2.9608, |
|
"step": 185000 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_accuracy": 0.4075345932932098, |
|
"eval_loss": 3.3682754039764404, |
|
"eval_runtime": 153.8415, |
|
"eval_samples_per_second": 376.485, |
|
"eval_steps_per_second": 5.883, |
|
"step": 185920 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"learning_rate": 0.0005470397834274953, |
|
"loss": 2.9544, |
|
"step": 186000 |
|
}, |
|
{ |
|
"epoch": 10.06, |
|
"learning_rate": 0.0005440972222222223, |
|
"loss": 2.8967, |
|
"step": 187000 |
|
}, |
|
{ |
|
"epoch": 10.11, |
|
"learning_rate": 0.0005411576035781545, |
|
"loss": 2.9034, |
|
"step": 188000 |
|
}, |
|
{ |
|
"epoch": 10.17, |
|
"learning_rate": 0.0005382150423728813, |
|
"loss": 2.9113, |
|
"step": 189000 |
|
}, |
|
{ |
|
"epoch": 10.22, |
|
"learning_rate": 0.0005352724811676083, |
|
"loss": 2.9107, |
|
"step": 190000 |
|
}, |
|
{ |
|
"epoch": 10.27, |
|
"learning_rate": 0.0005323328625235405, |
|
"loss": 2.9144, |
|
"step": 191000 |
|
}, |
|
{ |
|
"epoch": 10.33, |
|
"learning_rate": 0.0005293903013182673, |
|
"loss": 2.9206, |
|
"step": 192000 |
|
}, |
|
{ |
|
"epoch": 10.38, |
|
"learning_rate": 0.0005264477401129943, |
|
"loss": 2.9229, |
|
"step": 193000 |
|
}, |
|
{ |
|
"epoch": 10.43, |
|
"learning_rate": 0.0005235051789077213, |
|
"loss": 2.9233, |
|
"step": 194000 |
|
}, |
|
{ |
|
"epoch": 10.49, |
|
"learning_rate": 0.0005205626177024481, |
|
"loss": 2.9323, |
|
"step": 195000 |
|
}, |
|
{ |
|
"epoch": 10.54, |
|
"learning_rate": 0.0005176229990583804, |
|
"loss": 2.9275, |
|
"step": 196000 |
|
}, |
|
{ |
|
"epoch": 10.6, |
|
"learning_rate": 0.0005146804378531074, |
|
"loss": 2.9309, |
|
"step": 197000 |
|
}, |
|
{ |
|
"epoch": 10.65, |
|
"learning_rate": 0.0005117408192090396, |
|
"loss": 2.9346, |
|
"step": 198000 |
|
}, |
|
{ |
|
"epoch": 10.7, |
|
"learning_rate": 0.0005088012005649718, |
|
"loss": 2.9339, |
|
"step": 199000 |
|
}, |
|
{ |
|
"epoch": 10.76, |
|
"learning_rate": 0.0005058586393596986, |
|
"loss": 2.9358, |
|
"step": 200000 |
|
}, |
|
{ |
|
"epoch": 10.81, |
|
"learning_rate": 0.0005029160781544256, |
|
"loss": 2.9336, |
|
"step": 201000 |
|
}, |
|
{ |
|
"epoch": 10.86, |
|
"learning_rate": 0.0004999735169491526, |
|
"loss": 2.938, |
|
"step": 202000 |
|
}, |
|
{ |
|
"epoch": 10.92, |
|
"learning_rate": 0.0004970309557438794, |
|
"loss": 2.9378, |
|
"step": 203000 |
|
}, |
|
{ |
|
"epoch": 10.97, |
|
"learning_rate": 0.0004940913370998117, |
|
"loss": 2.9402, |
|
"step": 204000 |
|
}, |
|
{ |
|
"epoch": 11.0, |
|
"eval_accuracy": 0.40864939315828364, |
|
"eval_loss": 3.372835159301758, |
|
"eval_runtime": 153.3961, |
|
"eval_samples_per_second": 377.578, |
|
"eval_steps_per_second": 5.9, |
|
"step": 204512 |
|
}, |
|
{ |
|
"epoch": 11.03, |
|
"learning_rate": 0.0004911487758945386, |
|
"loss": 2.9052, |
|
"step": 205000 |
|
}, |
|
{ |
|
"epoch": 11.08, |
|
"learning_rate": 0.0004882091572504708, |
|
"loss": 2.8772, |
|
"step": 206000 |
|
}, |
|
{ |
|
"epoch": 11.13, |
|
"learning_rate": 0.0004852665960451977, |
|
"loss": 2.8834, |
|
"step": 207000 |
|
}, |
|
{ |
|
"epoch": 11.19, |
|
"learning_rate": 0.0004823240348399247, |
|
"loss": 2.8883, |
|
"step": 208000 |
|
}, |
|
{ |
|
"epoch": 11.24, |
|
"learning_rate": 0.0004793844161958569, |
|
"loss": 2.8927, |
|
"step": 209000 |
|
}, |
|
{ |
|
"epoch": 11.3, |
|
"learning_rate": 0.0004764418549905838, |
|
"loss": 2.8952, |
|
"step": 210000 |
|
}, |
|
{ |
|
"epoch": 11.35, |
|
"learning_rate": 0.00047349929378531075, |
|
"loss": 2.9015, |
|
"step": 211000 |
|
}, |
|
{ |
|
"epoch": 11.4, |
|
"learning_rate": 0.00047055967514124294, |
|
"loss": 2.9018, |
|
"step": 212000 |
|
}, |
|
{ |
|
"epoch": 11.46, |
|
"learning_rate": 0.00046762005649717513, |
|
"loss": 2.9079, |
|
"step": 213000 |
|
}, |
|
{ |
|
"epoch": 11.51, |
|
"learning_rate": 0.00046467749529190204, |
|
"loss": 2.9015, |
|
"step": 214000 |
|
}, |
|
{ |
|
"epoch": 11.56, |
|
"learning_rate": 0.00046173493408662906, |
|
"loss": 2.9078, |
|
"step": 215000 |
|
}, |
|
{ |
|
"epoch": 11.62, |
|
"learning_rate": 0.00045879237288135597, |
|
"loss": 2.9094, |
|
"step": 216000 |
|
}, |
|
{ |
|
"epoch": 11.67, |
|
"learning_rate": 0.00045585275423728816, |
|
"loss": 2.9103, |
|
"step": 217000 |
|
}, |
|
{ |
|
"epoch": 11.73, |
|
"learning_rate": 0.00045291019303201507, |
|
"loss": 2.912, |
|
"step": 218000 |
|
}, |
|
{ |
|
"epoch": 11.78, |
|
"learning_rate": 0.00044997057438794726, |
|
"loss": 2.9127, |
|
"step": 219000 |
|
}, |
|
{ |
|
"epoch": 11.83, |
|
"learning_rate": 0.0004470309557438795, |
|
"loss": 2.9159, |
|
"step": 220000 |
|
}, |
|
{ |
|
"epoch": 11.89, |
|
"learning_rate": 0.0004440883945386064, |
|
"loss": 2.9151, |
|
"step": 221000 |
|
}, |
|
{ |
|
"epoch": 11.94, |
|
"learning_rate": 0.0004411458333333334, |
|
"loss": 2.9164, |
|
"step": 222000 |
|
}, |
|
{ |
|
"epoch": 11.99, |
|
"learning_rate": 0.0004382062146892655, |
|
"loss": 2.9154, |
|
"step": 223000 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"eval_accuracy": 0.4082875568274841, |
|
"eval_loss": 3.384491205215454, |
|
"eval_runtime": 153.6213, |
|
"eval_samples_per_second": 377.025, |
|
"eval_steps_per_second": 5.891, |
|
"step": 223104 |
|
}, |
|
{ |
|
"epoch": 12.05, |
|
"learning_rate": 0.0004352636534839925, |
|
"loss": 2.8564, |
|
"step": 224000 |
|
}, |
|
{ |
|
"epoch": 12.1, |
|
"learning_rate": 0.00043232403483992466, |
|
"loss": 2.8614, |
|
"step": 225000 |
|
}, |
|
{ |
|
"epoch": 12.16, |
|
"learning_rate": 0.00042938147363465163, |
|
"loss": 2.8663, |
|
"step": 226000 |
|
}, |
|
{ |
|
"epoch": 12.21, |
|
"learning_rate": 0.00042643891242937854, |
|
"loss": 2.8715, |
|
"step": 227000 |
|
}, |
|
{ |
|
"epoch": 12.26, |
|
"learning_rate": 0.00042349635122410545, |
|
"loss": 2.8779, |
|
"step": 228000 |
|
}, |
|
{ |
|
"epoch": 12.32, |
|
"learning_rate": 0.0004205567325800377, |
|
"loss": 2.8745, |
|
"step": 229000 |
|
}, |
|
{ |
|
"epoch": 12.37, |
|
"learning_rate": 0.0004176141713747646, |
|
"loss": 2.8812, |
|
"step": 230000 |
|
}, |
|
{ |
|
"epoch": 12.42, |
|
"learning_rate": 0.0004146716101694915, |
|
"loss": 2.8815, |
|
"step": 231000 |
|
}, |
|
{ |
|
"epoch": 12.48, |
|
"learning_rate": 0.0004117290489642185, |
|
"loss": 2.8818, |
|
"step": 232000 |
|
}, |
|
{ |
|
"epoch": 12.53, |
|
"learning_rate": 0.00040878943032015067, |
|
"loss": 2.8834, |
|
"step": 233000 |
|
}, |
|
{ |
|
"epoch": 12.59, |
|
"learning_rate": 0.00040584981167608286, |
|
"loss": 2.8884, |
|
"step": 234000 |
|
}, |
|
{ |
|
"epoch": 12.64, |
|
"learning_rate": 0.00040290725047080977, |
|
"loss": 2.8882, |
|
"step": 235000 |
|
}, |
|
{ |
|
"epoch": 12.69, |
|
"learning_rate": 0.00039996468926553673, |
|
"loss": 2.8902, |
|
"step": 236000 |
|
}, |
|
{ |
|
"epoch": 12.75, |
|
"learning_rate": 0.00039702212806026364, |
|
"loss": 2.8924, |
|
"step": 237000 |
|
}, |
|
{ |
|
"epoch": 12.8, |
|
"learning_rate": 0.0003940795668549906, |
|
"loss": 2.8938, |
|
"step": 238000 |
|
}, |
|
{ |
|
"epoch": 12.85, |
|
"learning_rate": 0.0003911399482109228, |
|
"loss": 2.8921, |
|
"step": 239000 |
|
}, |
|
{ |
|
"epoch": 12.91, |
|
"learning_rate": 0.00038820032956685504, |
|
"loss": 2.8936, |
|
"step": 240000 |
|
}, |
|
{ |
|
"epoch": 12.96, |
|
"learning_rate": 0.00038525776836158195, |
|
"loss": 2.891, |
|
"step": 241000 |
|
}, |
|
{ |
|
"epoch": 13.0, |
|
"eval_accuracy": 0.4097539815264616, |
|
"eval_loss": 3.374075174331665, |
|
"eval_runtime": 153.5328, |
|
"eval_samples_per_second": 377.242, |
|
"eval_steps_per_second": 5.895, |
|
"step": 241696 |
|
}, |
|
{ |
|
"epoch": 13.02, |
|
"learning_rate": 0.00038231520715630886, |
|
"loss": 2.8823, |
|
"step": 242000 |
|
}, |
|
{ |
|
"epoch": 13.07, |
|
"learning_rate": 0.00037937264595103577, |
|
"loss": 2.8372, |
|
"step": 243000 |
|
}, |
|
{ |
|
"epoch": 13.12, |
|
"learning_rate": 0.00037643596986817324, |
|
"loss": 2.8443, |
|
"step": 244000 |
|
}, |
|
{ |
|
"epoch": 13.18, |
|
"learning_rate": 0.0003734934086629002, |
|
"loss": 2.8465, |
|
"step": 245000 |
|
}, |
|
{ |
|
"epoch": 13.23, |
|
"learning_rate": 0.0003705508474576271, |
|
"loss": 2.8528, |
|
"step": 246000 |
|
}, |
|
{ |
|
"epoch": 13.29, |
|
"learning_rate": 0.000367608286252354, |
|
"loss": 2.8543, |
|
"step": 247000 |
|
}, |
|
{ |
|
"epoch": 13.34, |
|
"learning_rate": 0.000364665725047081, |
|
"loss": 2.8555, |
|
"step": 248000 |
|
}, |
|
{ |
|
"epoch": 13.39, |
|
"learning_rate": 0.0003617231638418079, |
|
"loss": 2.863, |
|
"step": 249000 |
|
}, |
|
{ |
|
"epoch": 13.45, |
|
"learning_rate": 0.00035878060263653486, |
|
"loss": 2.8651, |
|
"step": 250000 |
|
}, |
|
{ |
|
"epoch": 13.5, |
|
"learning_rate": 0.00035584686911487755, |
|
"loss": 2.866, |
|
"step": 251000 |
|
}, |
|
{ |
|
"epoch": 13.55, |
|
"learning_rate": 0.0003529043079096045, |
|
"loss": 2.8672, |
|
"step": 252000 |
|
}, |
|
{ |
|
"epoch": 13.61, |
|
"learning_rate": 0.00034996174670433143, |
|
"loss": 2.8659, |
|
"step": 253000 |
|
}, |
|
{ |
|
"epoch": 13.66, |
|
"learning_rate": 0.00034701918549905834, |
|
"loss": 2.8692, |
|
"step": 254000 |
|
}, |
|
{ |
|
"epoch": 13.72, |
|
"learning_rate": 0.00034407662429378536, |
|
"loss": 2.8718, |
|
"step": 255000 |
|
}, |
|
{ |
|
"epoch": 13.77, |
|
"learning_rate": 0.0003411370056497175, |
|
"loss": 2.8712, |
|
"step": 256000 |
|
}, |
|
{ |
|
"epoch": 13.82, |
|
"learning_rate": 0.00033819738700564974, |
|
"loss": 2.8723, |
|
"step": 257000 |
|
}, |
|
{ |
|
"epoch": 13.88, |
|
"learning_rate": 0.00033525482580037665, |
|
"loss": 2.8748, |
|
"step": 258000 |
|
}, |
|
{ |
|
"epoch": 13.93, |
|
"learning_rate": 0.0003323122645951036, |
|
"loss": 2.8718, |
|
"step": 259000 |
|
}, |
|
{ |
|
"epoch": 13.98, |
|
"learning_rate": 0.0003293726459510358, |
|
"loss": 2.8754, |
|
"step": 260000 |
|
}, |
|
{ |
|
"epoch": 14.0, |
|
"eval_accuracy": 0.41060670869817323, |
|
"eval_loss": 3.367408514022827, |
|
"eval_runtime": 152.6297, |
|
"eval_samples_per_second": 379.474, |
|
"eval_steps_per_second": 5.929, |
|
"step": 260288 |
|
}, |
|
{ |
|
"epoch": 14.04, |
|
"learning_rate": 0.00032643008474576277, |
|
"loss": 2.8386, |
|
"step": 261000 |
|
}, |
|
{ |
|
"epoch": 14.09, |
|
"learning_rate": 0.0003234875235404897, |
|
"loss": 2.8267, |
|
"step": 262000 |
|
}, |
|
{ |
|
"epoch": 14.15, |
|
"learning_rate": 0.0003205449623352166, |
|
"loss": 2.8268, |
|
"step": 263000 |
|
}, |
|
{ |
|
"epoch": 14.2, |
|
"learning_rate": 0.0003176024011299435, |
|
"loss": 2.8328, |
|
"step": 264000 |
|
}, |
|
{ |
|
"epoch": 14.25, |
|
"learning_rate": 0.0003146627824858757, |
|
"loss": 2.8368, |
|
"step": 265000 |
|
}, |
|
{ |
|
"epoch": 14.31, |
|
"learning_rate": 0.0003117202212806026, |
|
"loss": 2.8369, |
|
"step": 266000 |
|
}, |
|
{ |
|
"epoch": 14.36, |
|
"learning_rate": 0.00030878060263653484, |
|
"loss": 2.8409, |
|
"step": 267000 |
|
}, |
|
{ |
|
"epoch": 14.41, |
|
"learning_rate": 0.00030583804143126175, |
|
"loss": 2.8439, |
|
"step": 268000 |
|
}, |
|
{ |
|
"epoch": 14.47, |
|
"learning_rate": 0.000302898422787194, |
|
"loss": 2.8428, |
|
"step": 269000 |
|
}, |
|
{ |
|
"epoch": 14.52, |
|
"learning_rate": 0.0002999588041431262, |
|
"loss": 2.8454, |
|
"step": 270000 |
|
}, |
|
{ |
|
"epoch": 14.58, |
|
"learning_rate": 0.00029701624293785315, |
|
"loss": 2.8435, |
|
"step": 271000 |
|
}, |
|
{ |
|
"epoch": 14.63, |
|
"learning_rate": 0.00029407368173258006, |
|
"loss": 2.8488, |
|
"step": 272000 |
|
}, |
|
{ |
|
"epoch": 14.68, |
|
"learning_rate": 0.00029113112052730697, |
|
"loss": 2.8501, |
|
"step": 273000 |
|
}, |
|
{ |
|
"epoch": 14.74, |
|
"learning_rate": 0.00028818855932203393, |
|
"loss": 2.8496, |
|
"step": 274000 |
|
}, |
|
{ |
|
"epoch": 14.79, |
|
"learning_rate": 0.00028524894067796607, |
|
"loss": 2.8539, |
|
"step": 275000 |
|
}, |
|
{ |
|
"epoch": 14.85, |
|
"learning_rate": 0.00028230637947269303, |
|
"loss": 2.8539, |
|
"step": 276000 |
|
}, |
|
{ |
|
"epoch": 14.9, |
|
"learning_rate": 0.0002793667608286252, |
|
"loss": 2.8558, |
|
"step": 277000 |
|
}, |
|
{ |
|
"epoch": 14.95, |
|
"learning_rate": 0.00027642714218455746, |
|
"loss": 2.8555, |
|
"step": 278000 |
|
}, |
|
{ |
|
"epoch": 15.0, |
|
"eval_accuracy": 0.4095201248113644, |
|
"eval_loss": 3.3868112564086914, |
|
"eval_runtime": 153.6538, |
|
"eval_samples_per_second": 376.945, |
|
"eval_steps_per_second": 5.89, |
|
"step": 278880 |
|
}, |
|
{ |
|
"epoch": 15.01, |
|
"learning_rate": 0.0002734845809792844, |
|
"loss": 2.8488, |
|
"step": 279000 |
|
}, |
|
{ |
|
"epoch": 15.06, |
|
"learning_rate": 0.00027054201977401134, |
|
"loss": 2.8093, |
|
"step": 280000 |
|
}, |
|
{ |
|
"epoch": 15.11, |
|
"learning_rate": 0.00026759945856873825, |
|
"loss": 2.8137, |
|
"step": 281000 |
|
}, |
|
{ |
|
"epoch": 15.17, |
|
"learning_rate": 0.00026465983992467044, |
|
"loss": 2.8098, |
|
"step": 282000 |
|
}, |
|
{ |
|
"epoch": 15.22, |
|
"learning_rate": 0.0002617172787193974, |
|
"loss": 2.815, |
|
"step": 283000 |
|
}, |
|
{ |
|
"epoch": 15.28, |
|
"learning_rate": 0.00025877766007532954, |
|
"loss": 2.8226, |
|
"step": 284000 |
|
}, |
|
{ |
|
"epoch": 15.33, |
|
"learning_rate": 0.0002558350988700565, |
|
"loss": 2.8227, |
|
"step": 285000 |
|
}, |
|
{ |
|
"epoch": 15.38, |
|
"learning_rate": 0.0002528925376647834, |
|
"loss": 2.8204, |
|
"step": 286000 |
|
}, |
|
{ |
|
"epoch": 15.44, |
|
"learning_rate": 0.0002499529190207156, |
|
"loss": 2.8239, |
|
"step": 287000 |
|
}, |
|
{ |
|
"epoch": 15.49, |
|
"learning_rate": 0.00024701035781544257, |
|
"loss": 2.8234, |
|
"step": 288000 |
|
}, |
|
{ |
|
"epoch": 15.54, |
|
"learning_rate": 0.0002440677966101695, |
|
"loss": 2.8268, |
|
"step": 289000 |
|
}, |
|
{ |
|
"epoch": 15.6, |
|
"learning_rate": 0.00024113112052730697, |
|
"loss": 2.8285, |
|
"step": 290000 |
|
}, |
|
{ |
|
"epoch": 15.65, |
|
"learning_rate": 0.0002381885593220339, |
|
"loss": 2.8301, |
|
"step": 291000 |
|
}, |
|
{ |
|
"epoch": 15.71, |
|
"learning_rate": 0.00023524599811676085, |
|
"loss": 2.831, |
|
"step": 292000 |
|
}, |
|
{ |
|
"epoch": 15.76, |
|
"learning_rate": 0.00023230343691148776, |
|
"loss": 2.8319, |
|
"step": 293000 |
|
}, |
|
{ |
|
"epoch": 15.81, |
|
"learning_rate": 0.00022936381826741997, |
|
"loss": 2.8331, |
|
"step": 294000 |
|
}, |
|
{ |
|
"epoch": 15.87, |
|
"learning_rate": 0.00022642125706214688, |
|
"loss": 2.837, |
|
"step": 295000 |
|
}, |
|
{ |
|
"epoch": 15.92, |
|
"learning_rate": 0.0002234816384180791, |
|
"loss": 2.835, |
|
"step": 296000 |
|
}, |
|
{ |
|
"epoch": 15.97, |
|
"learning_rate": 0.00022053907721280604, |
|
"loss": 2.8368, |
|
"step": 297000 |
|
}, |
|
{ |
|
"epoch": 16.0, |
|
"eval_accuracy": 0.40982626817554074, |
|
"eval_loss": 3.3892040252685547, |
|
"eval_runtime": 153.9496, |
|
"eval_samples_per_second": 376.22, |
|
"eval_steps_per_second": 5.879, |
|
"step": 297472 |
|
}, |
|
{ |
|
"epoch": 16.03, |
|
"learning_rate": 0.00021759651600753298, |
|
"loss": 2.8135, |
|
"step": 298000 |
|
}, |
|
{ |
|
"epoch": 16.08, |
|
"learning_rate": 0.00021465689736346516, |
|
"loss": 2.7892, |
|
"step": 299000 |
|
}, |
|
{ |
|
"epoch": 16.14, |
|
"learning_rate": 0.0002117143361581921, |
|
"loss": 2.7966, |
|
"step": 300000 |
|
}, |
|
{ |
|
"epoch": 16.19, |
|
"learning_rate": 0.0002087747175141243, |
|
"loss": 2.7981, |
|
"step": 301000 |
|
}, |
|
{ |
|
"epoch": 16.24, |
|
"learning_rate": 0.0002058350988700565, |
|
"loss": 2.7999, |
|
"step": 302000 |
|
}, |
|
{ |
|
"epoch": 16.3, |
|
"learning_rate": 0.00020289253766478344, |
|
"loss": 2.8005, |
|
"step": 303000 |
|
}, |
|
{ |
|
"epoch": 16.35, |
|
"learning_rate": 0.00019995291902071563, |
|
"loss": 2.8064, |
|
"step": 304000 |
|
}, |
|
{ |
|
"epoch": 16.4, |
|
"learning_rate": 0.00019701035781544257, |
|
"loss": 2.8087, |
|
"step": 305000 |
|
}, |
|
{ |
|
"epoch": 16.46, |
|
"learning_rate": 0.00019406779661016948, |
|
"loss": 2.8069, |
|
"step": 306000 |
|
}, |
|
{ |
|
"epoch": 16.51, |
|
"learning_rate": 0.00019112523540489642, |
|
"loss": 2.8089, |
|
"step": 307000 |
|
}, |
|
{ |
|
"epoch": 16.57, |
|
"learning_rate": 0.00018818561676082864, |
|
"loss": 2.8103, |
|
"step": 308000 |
|
}, |
|
{ |
|
"epoch": 16.62, |
|
"learning_rate": 0.00018524305555555557, |
|
"loss": 2.8099, |
|
"step": 309000 |
|
}, |
|
{ |
|
"epoch": 16.67, |
|
"learning_rate": 0.00018230343691148776, |
|
"loss": 2.8134, |
|
"step": 310000 |
|
}, |
|
{ |
|
"epoch": 16.73, |
|
"learning_rate": 0.0001793608757062147, |
|
"loss": 2.8105, |
|
"step": 311000 |
|
}, |
|
{ |
|
"epoch": 16.78, |
|
"learning_rate": 0.0001764212570621469, |
|
"loss": 2.8145, |
|
"step": 312000 |
|
}, |
|
{ |
|
"epoch": 16.84, |
|
"learning_rate": 0.00017347869585687383, |
|
"loss": 2.817, |
|
"step": 313000 |
|
}, |
|
{ |
|
"epoch": 16.89, |
|
"learning_rate": 0.00017053613465160076, |
|
"loss": 2.8195, |
|
"step": 314000 |
|
}, |
|
{ |
|
"epoch": 16.94, |
|
"learning_rate": 0.00016759651600753295, |
|
"loss": 2.8218, |
|
"step": 315000 |
|
}, |
|
{ |
|
"epoch": 17.0, |
|
"learning_rate": 0.0001646539548022599, |
|
"loss": 2.8185, |
|
"step": 316000 |
|
}, |
|
{ |
|
"epoch": 17.0, |
|
"eval_accuracy": 0.4105727151624538, |
|
"eval_loss": 3.3865416049957275, |
|
"eval_runtime": 155.0516, |
|
"eval_samples_per_second": 373.547, |
|
"eval_steps_per_second": 5.837, |
|
"step": 316064 |
|
}, |
|
{ |
|
"epoch": 17.05, |
|
"learning_rate": 0.00016171139359698683, |
|
"loss": 2.7805, |
|
"step": 317000 |
|
}, |
|
{ |
|
"epoch": 17.1, |
|
"learning_rate": 0.00015877177495291902, |
|
"loss": 2.7868, |
|
"step": 318000 |
|
}, |
|
{ |
|
"epoch": 17.16, |
|
"learning_rate": 0.00015582921374764595, |
|
"loss": 2.7826, |
|
"step": 319000 |
|
}, |
|
{ |
|
"epoch": 17.21, |
|
"learning_rate": 0.0001528866525423729, |
|
"loss": 2.783, |
|
"step": 320000 |
|
}, |
|
{ |
|
"epoch": 17.27, |
|
"learning_rate": 0.00014994703389830508, |
|
"loss": 2.7881, |
|
"step": 321000 |
|
}, |
|
{ |
|
"epoch": 17.32, |
|
"learning_rate": 0.00014700447269303202, |
|
"loss": 2.7889, |
|
"step": 322000 |
|
}, |
|
{ |
|
"epoch": 17.37, |
|
"learning_rate": 0.0001440648540489642, |
|
"loss": 2.7868, |
|
"step": 323000 |
|
}, |
|
{ |
|
"epoch": 17.43, |
|
"learning_rate": 0.00014112229284369115, |
|
"loss": 2.794, |
|
"step": 324000 |
|
}, |
|
{ |
|
"epoch": 17.48, |
|
"learning_rate": 0.00013818267419962336, |
|
"loss": 2.7907, |
|
"step": 325000 |
|
}, |
|
{ |
|
"epoch": 17.53, |
|
"learning_rate": 0.0001352401129943503, |
|
"loss": 2.7941, |
|
"step": 326000 |
|
}, |
|
{ |
|
"epoch": 17.59, |
|
"learning_rate": 0.0001322975517890772, |
|
"loss": 2.7927, |
|
"step": 327000 |
|
}, |
|
{ |
|
"epoch": 17.64, |
|
"learning_rate": 0.00012935499058380415, |
|
"loss": 2.7957, |
|
"step": 328000 |
|
}, |
|
{ |
|
"epoch": 17.7, |
|
"learning_rate": 0.00012641537193973634, |
|
"loss": 2.7919, |
|
"step": 329000 |
|
}, |
|
{ |
|
"epoch": 17.75, |
|
"learning_rate": 0.00012347281073446327, |
|
"loss": 2.7963, |
|
"step": 330000 |
|
}, |
|
{ |
|
"epoch": 17.8, |
|
"learning_rate": 0.00012053024952919021, |
|
"loss": 2.798, |
|
"step": 331000 |
|
}, |
|
{ |
|
"epoch": 17.86, |
|
"learning_rate": 0.00011759063088512241, |
|
"loss": 2.7969, |
|
"step": 332000 |
|
}, |
|
{ |
|
"epoch": 17.91, |
|
"learning_rate": 0.00011464806967984934, |
|
"loss": 2.7977, |
|
"step": 333000 |
|
}, |
|
{ |
|
"epoch": 17.96, |
|
"learning_rate": 0.00011170845103578154, |
|
"loss": 2.7969, |
|
"step": 334000 |
|
}, |
|
{ |
|
"epoch": 18.0, |
|
"eval_accuracy": 0.4098990250909243, |
|
"eval_loss": 3.400566577911377, |
|
"eval_runtime": 153.6726, |
|
"eval_samples_per_second": 376.899, |
|
"eval_steps_per_second": 5.889, |
|
"step": 334656 |
|
}, |
|
{ |
|
"epoch": 18.02, |
|
"learning_rate": 0.00010876588983050848, |
|
"loss": 2.7861, |
|
"step": 335000 |
|
}, |
|
{ |
|
"epoch": 18.07, |
|
"learning_rate": 0.0001058233286252354, |
|
"loss": 2.7663, |
|
"step": 336000 |
|
}, |
|
{ |
|
"epoch": 18.13, |
|
"learning_rate": 0.00010288076741996234, |
|
"loss": 2.769, |
|
"step": 337000 |
|
}, |
|
{ |
|
"epoch": 18.18, |
|
"learning_rate": 9.994114877589454e-05, |
|
"loss": 2.7697, |
|
"step": 338000 |
|
}, |
|
{ |
|
"epoch": 18.23, |
|
"learning_rate": 9.699858757062147e-05, |
|
"loss": 2.7771, |
|
"step": 339000 |
|
}, |
|
{ |
|
"epoch": 18.29, |
|
"learning_rate": 9.40560263653484e-05, |
|
"loss": 2.7719, |
|
"step": 340000 |
|
}, |
|
{ |
|
"epoch": 18.34, |
|
"learning_rate": 9.111346516007534e-05, |
|
"loss": 2.7793, |
|
"step": 341000 |
|
}, |
|
{ |
|
"epoch": 18.4, |
|
"learning_rate": 8.817384651600753e-05, |
|
"loss": 2.7736, |
|
"step": 342000 |
|
}, |
|
{ |
|
"epoch": 18.45, |
|
"learning_rate": 8.523422787193975e-05, |
|
"loss": 2.777, |
|
"step": 343000 |
|
}, |
|
{ |
|
"epoch": 18.5, |
|
"learning_rate": 8.229166666666667e-05, |
|
"loss": 2.7778, |
|
"step": 344000 |
|
}, |
|
{ |
|
"epoch": 18.56, |
|
"learning_rate": 7.935204802259887e-05, |
|
"loss": 2.7766, |
|
"step": 345000 |
|
}, |
|
{ |
|
"epoch": 18.61, |
|
"learning_rate": 7.640948681732581e-05, |
|
"loss": 2.7771, |
|
"step": 346000 |
|
}, |
|
{ |
|
"epoch": 18.66, |
|
"learning_rate": 7.3469868173258e-05, |
|
"loss": 2.7785, |
|
"step": 347000 |
|
}, |
|
{ |
|
"epoch": 18.72, |
|
"learning_rate": 7.052730696798494e-05, |
|
"loss": 2.7756, |
|
"step": 348000 |
|
}, |
|
{ |
|
"epoch": 18.77, |
|
"learning_rate": 6.758474576271187e-05, |
|
"loss": 2.7762, |
|
"step": 349000 |
|
}, |
|
{ |
|
"epoch": 18.83, |
|
"learning_rate": 6.464512711864406e-05, |
|
"loss": 2.7811, |
|
"step": 350000 |
|
}, |
|
{ |
|
"epoch": 18.88, |
|
"learning_rate": 6.1702565913371e-05, |
|
"loss": 2.7773, |
|
"step": 351000 |
|
}, |
|
{ |
|
"epoch": 18.93, |
|
"learning_rate": 5.8760004708097925e-05, |
|
"loss": 2.7786, |
|
"step": 352000 |
|
}, |
|
{ |
|
"epoch": 18.99, |
|
"learning_rate": 5.5820386064030134e-05, |
|
"loss": 2.7805, |
|
"step": 353000 |
|
}, |
|
{ |
|
"epoch": 19.0, |
|
"eval_accuracy": 0.41040066487593735, |
|
"eval_loss": 3.399707794189453, |
|
"eval_runtime": 153.9752, |
|
"eval_samples_per_second": 376.158, |
|
"eval_steps_per_second": 5.878, |
|
"step": 353248 |
|
}, |
|
{ |
|
"epoch": 19.04, |
|
"learning_rate": 5.2877824858757065e-05, |
|
"loss": 2.7638, |
|
"step": 354000 |
|
}, |
|
{ |
|
"epoch": 19.09, |
|
"learning_rate": 4.993526365348399e-05, |
|
"loss": 2.7559, |
|
"step": 355000 |
|
}, |
|
{ |
|
"epoch": 19.15, |
|
"learning_rate": 4.699270244821093e-05, |
|
"loss": 2.7604, |
|
"step": 356000 |
|
}, |
|
{ |
|
"epoch": 19.2, |
|
"learning_rate": 4.405308380414313e-05, |
|
"loss": 2.762, |
|
"step": 357000 |
|
}, |
|
{ |
|
"epoch": 19.26, |
|
"learning_rate": 4.111346516007533e-05, |
|
"loss": 2.7598, |
|
"step": 358000 |
|
}, |
|
{ |
|
"epoch": 19.31, |
|
"learning_rate": 3.8170903954802256e-05, |
|
"loss": 2.76, |
|
"step": 359000 |
|
}, |
|
{ |
|
"epoch": 19.36, |
|
"learning_rate": 3.523128531073446e-05, |
|
"loss": 2.7655, |
|
"step": 360000 |
|
}, |
|
{ |
|
"epoch": 19.42, |
|
"learning_rate": 3.2288724105461396e-05, |
|
"loss": 2.7589, |
|
"step": 361000 |
|
}, |
|
{ |
|
"epoch": 19.47, |
|
"learning_rate": 2.9346162900188327e-05, |
|
"loss": 2.7604, |
|
"step": 362000 |
|
}, |
|
{ |
|
"epoch": 19.52, |
|
"learning_rate": 2.6403601694915254e-05, |
|
"loss": 2.7634, |
|
"step": 363000 |
|
}, |
|
{ |
|
"epoch": 19.58, |
|
"learning_rate": 2.3461040489642185e-05, |
|
"loss": 2.7618, |
|
"step": 364000 |
|
}, |
|
{ |
|
"epoch": 19.63, |
|
"learning_rate": 2.0518479284369116e-05, |
|
"loss": 2.764, |
|
"step": 365000 |
|
}, |
|
{ |
|
"epoch": 19.69, |
|
"learning_rate": 1.7575918079096047e-05, |
|
"loss": 2.7651, |
|
"step": 366000 |
|
}, |
|
{ |
|
"epoch": 19.74, |
|
"learning_rate": 1.4636299435028249e-05, |
|
"loss": 2.7575, |
|
"step": 367000 |
|
}, |
|
{ |
|
"epoch": 19.79, |
|
"learning_rate": 1.169373822975518e-05, |
|
"loss": 2.7625, |
|
"step": 368000 |
|
}, |
|
{ |
|
"epoch": 19.85, |
|
"learning_rate": 8.754119585687382e-06, |
|
"loss": 2.7612, |
|
"step": 369000 |
|
}, |
|
{ |
|
"epoch": 19.9, |
|
"learning_rate": 5.814500941619586e-06, |
|
"loss": 2.7639, |
|
"step": 370000 |
|
}, |
|
{ |
|
"epoch": 19.95, |
|
"learning_rate": 2.871939736346516e-06, |
|
"loss": 2.7623, |
|
"step": 371000 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"eval_accuracy": 0.40988659662430754, |
|
"eval_loss": 3.4119651317596436, |
|
"eval_runtime": 154.9649, |
|
"eval_samples_per_second": 373.755, |
|
"eval_steps_per_second": 5.84, |
|
"step": 371840 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"step": 371840, |
|
"total_flos": 1.56667295384064e+18, |
|
"train_loss": 3.030422134530729, |
|
"train_runtime": 80926.4318, |
|
"train_samples_per_second": 147.032, |
|
"train_steps_per_second": 4.595 |
|
} |
|
], |
|
"logging_steps": 1000, |
|
"max_steps": 371840, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 20, |
|
"save_steps": 5000, |
|
"total_flos": 1.56667295384064e+18, |
|
"train_batch_size": 32, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|