roberta-base-wechsel-swahili / trainer_state.json
benjamin's picture
initial commit
f890be9
raw
history blame
65.7 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 208.6811352253756,
"global_step": 250000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.42,
"learning_rate": 2.0000000000000003e-06,
"loss": 7.8213,
"step": 500
},
{
"epoch": 0.83,
"learning_rate": 4.000000000000001e-06,
"loss": 6.0676,
"step": 1000
},
{
"epoch": 1.25,
"learning_rate": 6e-06,
"loss": 5.2769,
"step": 1500
},
{
"epoch": 1.67,
"learning_rate": 8.000000000000001e-06,
"loss": 4.7202,
"step": 2000
},
{
"epoch": 2.09,
"learning_rate": 1e-05,
"loss": 4.3225,
"step": 2500
},
{
"epoch": 2.5,
"learning_rate": 1.2e-05,
"loss": 3.9793,
"step": 3000
},
{
"epoch": 2.92,
"learning_rate": 1.4000000000000001e-05,
"loss": 3.6914,
"step": 3500
},
{
"epoch": 3.34,
"learning_rate": 1.6000000000000003e-05,
"loss": 3.4487,
"step": 4000
},
{
"epoch": 3.76,
"learning_rate": 1.8e-05,
"loss": 3.2435,
"step": 4500
},
{
"epoch": 4.17,
"learning_rate": 2e-05,
"loss": 3.0638,
"step": 5000
},
{
"epoch": 4.59,
"learning_rate": 2.2000000000000003e-05,
"loss": 2.9136,
"step": 5500
},
{
"epoch": 5.01,
"learning_rate": 2.4e-05,
"loss": 2.7839,
"step": 6000
},
{
"epoch": 5.43,
"learning_rate": 2.6000000000000002e-05,
"loss": 2.665,
"step": 6500
},
{
"epoch": 5.84,
"learning_rate": 2.8000000000000003e-05,
"loss": 2.5639,
"step": 7000
},
{
"epoch": 6.26,
"learning_rate": 3e-05,
"loss": 2.4721,
"step": 7500
},
{
"epoch": 6.68,
"learning_rate": 3.2000000000000005e-05,
"loss": 2.3918,
"step": 8000
},
{
"epoch": 7.1,
"learning_rate": 3.4000000000000007e-05,
"loss": 2.3195,
"step": 8500
},
{
"epoch": 7.51,
"learning_rate": 3.6e-05,
"loss": 2.2511,
"step": 9000
},
{
"epoch": 7.93,
"learning_rate": 3.8e-05,
"loss": 2.1977,
"step": 9500
},
{
"epoch": 8.35,
"learning_rate": 4e-05,
"loss": 2.1409,
"step": 10000
},
{
"epoch": 8.76,
"learning_rate": 4.2e-05,
"loss": 2.093,
"step": 10500
},
{
"epoch": 9.18,
"learning_rate": 4.4000000000000006e-05,
"loss": 2.0491,
"step": 11000
},
{
"epoch": 9.6,
"learning_rate": 4.600000000000001e-05,
"loss": 2.0059,
"step": 11500
},
{
"epoch": 10.02,
"learning_rate": 4.8e-05,
"loss": 1.9673,
"step": 12000
},
{
"epoch": 10.43,
"learning_rate": 5e-05,
"loss": 1.9332,
"step": 12500
},
{
"epoch": 10.43,
"eval_loss": 1.8155752420425415,
"eval_runtime": 140.158,
"eval_samples_per_second": 486.158,
"eval_steps_per_second": 3.803,
"step": 12500
},
{
"epoch": 10.85,
"learning_rate": 5.2000000000000004e-05,
"loss": 1.8996,
"step": 13000
},
{
"epoch": 11.27,
"learning_rate": 5.4000000000000005e-05,
"loss": 1.868,
"step": 13500
},
{
"epoch": 11.69,
"learning_rate": 5.6000000000000006e-05,
"loss": 1.8409,
"step": 14000
},
{
"epoch": 12.1,
"learning_rate": 5.8e-05,
"loss": 1.8127,
"step": 14500
},
{
"epoch": 12.52,
"learning_rate": 6e-05,
"loss": 1.787,
"step": 15000
},
{
"epoch": 12.94,
"learning_rate": 6.2e-05,
"loss": 1.7647,
"step": 15500
},
{
"epoch": 13.36,
"learning_rate": 6.400000000000001e-05,
"loss": 1.7386,
"step": 16000
},
{
"epoch": 13.77,
"learning_rate": 6.6e-05,
"loss": 1.7192,
"step": 16500
},
{
"epoch": 14.19,
"learning_rate": 6.800000000000001e-05,
"loss": 1.6991,
"step": 17000
},
{
"epoch": 14.61,
"learning_rate": 7e-05,
"loss": 1.6812,
"step": 17500
},
{
"epoch": 15.03,
"learning_rate": 7.2e-05,
"loss": 1.661,
"step": 18000
},
{
"epoch": 15.44,
"learning_rate": 7.4e-05,
"loss": 1.6418,
"step": 18500
},
{
"epoch": 15.86,
"learning_rate": 7.6e-05,
"loss": 1.6236,
"step": 19000
},
{
"epoch": 16.28,
"learning_rate": 7.800000000000001e-05,
"loss": 1.6107,
"step": 19500
},
{
"epoch": 16.69,
"learning_rate": 8e-05,
"loss": 1.5962,
"step": 20000
},
{
"epoch": 17.11,
"learning_rate": 8.2e-05,
"loss": 1.5812,
"step": 20500
},
{
"epoch": 17.53,
"learning_rate": 8.4e-05,
"loss": 1.5685,
"step": 21000
},
{
"epoch": 17.95,
"learning_rate": 8.6e-05,
"loss": 1.5554,
"step": 21500
},
{
"epoch": 18.36,
"learning_rate": 8.800000000000001e-05,
"loss": 1.5439,
"step": 22000
},
{
"epoch": 18.78,
"learning_rate": 9e-05,
"loss": 1.5343,
"step": 22500
},
{
"epoch": 19.2,
"learning_rate": 9.200000000000001e-05,
"loss": 1.521,
"step": 23000
},
{
"epoch": 19.62,
"learning_rate": 9.4e-05,
"loss": 1.5101,
"step": 23500
},
{
"epoch": 20.03,
"learning_rate": 9.6e-05,
"loss": 1.5014,
"step": 24000
},
{
"epoch": 20.45,
"learning_rate": 9.8e-05,
"loss": 1.4894,
"step": 24500
},
{
"epoch": 20.87,
"learning_rate": 0.0001,
"loss": 1.4842,
"step": 25000
},
{
"epoch": 20.87,
"eval_loss": 1.4012646675109863,
"eval_runtime": 125.2851,
"eval_samples_per_second": 543.871,
"eval_steps_per_second": 4.254,
"step": 25000
},
{
"epoch": 21.29,
"learning_rate": 9.977777777777779e-05,
"loss": 1.4737,
"step": 25500
},
{
"epoch": 21.7,
"learning_rate": 9.955555555555556e-05,
"loss": 1.4685,
"step": 26000
},
{
"epoch": 22.12,
"learning_rate": 9.933333333333334e-05,
"loss": 1.4549,
"step": 26500
},
{
"epoch": 22.54,
"learning_rate": 9.911111111111112e-05,
"loss": 1.4472,
"step": 27000
},
{
"epoch": 22.95,
"learning_rate": 9.888888888888889e-05,
"loss": 1.4382,
"step": 27500
},
{
"epoch": 23.37,
"learning_rate": 9.866666666666668e-05,
"loss": 1.4292,
"step": 28000
},
{
"epoch": 23.79,
"learning_rate": 9.844444444444444e-05,
"loss": 1.4244,
"step": 28500
},
{
"epoch": 24.21,
"learning_rate": 9.822222222222223e-05,
"loss": 1.4167,
"step": 29000
},
{
"epoch": 24.62,
"learning_rate": 9.8e-05,
"loss": 1.4092,
"step": 29500
},
{
"epoch": 25.04,
"learning_rate": 9.777777777777778e-05,
"loss": 1.4061,
"step": 30000
},
{
"epoch": 25.46,
"learning_rate": 9.755555555555555e-05,
"loss": 1.3953,
"step": 30500
},
{
"epoch": 25.88,
"learning_rate": 9.733333333333335e-05,
"loss": 1.3929,
"step": 31000
},
{
"epoch": 26.29,
"learning_rate": 9.711111111111111e-05,
"loss": 1.3849,
"step": 31500
},
{
"epoch": 26.71,
"learning_rate": 9.68888888888889e-05,
"loss": 1.3824,
"step": 32000
},
{
"epoch": 27.13,
"learning_rate": 9.666666666666667e-05,
"loss": 1.3748,
"step": 32500
},
{
"epoch": 27.55,
"learning_rate": 9.644444444444445e-05,
"loss": 1.3683,
"step": 33000
},
{
"epoch": 27.96,
"learning_rate": 9.622222222222222e-05,
"loss": 1.3655,
"step": 33500
},
{
"epoch": 28.38,
"learning_rate": 9.6e-05,
"loss": 1.359,
"step": 34000
},
{
"epoch": 28.8,
"learning_rate": 9.577777777777777e-05,
"loss": 1.3541,
"step": 34500
},
{
"epoch": 29.22,
"learning_rate": 9.555555555555557e-05,
"loss": 1.3509,
"step": 35000
},
{
"epoch": 29.63,
"learning_rate": 9.533333333333334e-05,
"loss": 1.3463,
"step": 35500
},
{
"epoch": 30.05,
"learning_rate": 9.511111111111112e-05,
"loss": 1.3416,
"step": 36000
},
{
"epoch": 30.47,
"learning_rate": 9.488888888888889e-05,
"loss": 1.3358,
"step": 36500
},
{
"epoch": 30.88,
"learning_rate": 9.466666666666667e-05,
"loss": 1.3326,
"step": 37000
},
{
"epoch": 31.3,
"learning_rate": 9.444444444444444e-05,
"loss": 1.3267,
"step": 37500
},
{
"epoch": 31.3,
"eval_loss": 1.268040418624878,
"eval_runtime": 125.1415,
"eval_samples_per_second": 544.496,
"eval_steps_per_second": 4.259,
"step": 37500
},
{
"epoch": 31.72,
"learning_rate": 9.422222222222223e-05,
"loss": 1.3276,
"step": 38000
},
{
"epoch": 32.14,
"learning_rate": 9.4e-05,
"loss": 1.3216,
"step": 38500
},
{
"epoch": 32.55,
"learning_rate": 9.377777777777779e-05,
"loss": 1.3181,
"step": 39000
},
{
"epoch": 32.97,
"learning_rate": 9.355555555555556e-05,
"loss": 1.3146,
"step": 39500
},
{
"epoch": 33.39,
"learning_rate": 9.333333333333334e-05,
"loss": 1.3091,
"step": 40000
},
{
"epoch": 33.81,
"learning_rate": 9.311111111111111e-05,
"loss": 1.3072,
"step": 40500
},
{
"epoch": 34.22,
"learning_rate": 9.28888888888889e-05,
"loss": 1.3041,
"step": 41000
},
{
"epoch": 34.64,
"learning_rate": 9.266666666666666e-05,
"loss": 1.2998,
"step": 41500
},
{
"epoch": 35.06,
"learning_rate": 9.244444444444445e-05,
"loss": 1.2995,
"step": 42000
},
{
"epoch": 35.48,
"learning_rate": 9.222222222222223e-05,
"loss": 1.2942,
"step": 42500
},
{
"epoch": 35.89,
"learning_rate": 9.200000000000001e-05,
"loss": 1.2928,
"step": 43000
},
{
"epoch": 36.31,
"learning_rate": 9.177777777777778e-05,
"loss": 1.289,
"step": 43500
},
{
"epoch": 36.73,
"learning_rate": 9.155555555555557e-05,
"loss": 1.2874,
"step": 44000
},
{
"epoch": 37.15,
"learning_rate": 9.133333333333334e-05,
"loss": 1.2859,
"step": 44500
},
{
"epoch": 37.56,
"learning_rate": 9.111111111111112e-05,
"loss": 1.2771,
"step": 45000
},
{
"epoch": 37.98,
"learning_rate": 9.088888888888889e-05,
"loss": 1.2809,
"step": 45500
},
{
"epoch": 38.4,
"learning_rate": 9.066666666666667e-05,
"loss": 1.2749,
"step": 46000
},
{
"epoch": 38.81,
"learning_rate": 9.044444444444445e-05,
"loss": 1.272,
"step": 46500
},
{
"epoch": 39.23,
"learning_rate": 9.022222222222224e-05,
"loss": 1.2708,
"step": 47000
},
{
"epoch": 39.65,
"learning_rate": 9e-05,
"loss": 1.2685,
"step": 47500
},
{
"epoch": 40.07,
"learning_rate": 8.977777777777779e-05,
"loss": 1.2678,
"step": 48000
},
{
"epoch": 40.48,
"learning_rate": 8.955555555555556e-05,
"loss": 1.2627,
"step": 48500
},
{
"epoch": 40.9,
"learning_rate": 8.933333333333334e-05,
"loss": 1.262,
"step": 49000
},
{
"epoch": 41.32,
"learning_rate": 8.911111111111111e-05,
"loss": 1.2555,
"step": 49500
},
{
"epoch": 41.74,
"learning_rate": 8.888888888888889e-05,
"loss": 1.2581,
"step": 50000
},
{
"epoch": 41.74,
"eval_loss": 1.2052675485610962,
"eval_runtime": 125.1251,
"eval_samples_per_second": 544.567,
"eval_steps_per_second": 4.26,
"step": 50000
},
{
"epoch": 42.15,
"learning_rate": 8.866666666666668e-05,
"loss": 1.2531,
"step": 50500
},
{
"epoch": 42.57,
"learning_rate": 8.844444444444445e-05,
"loss": 1.2518,
"step": 51000
},
{
"epoch": 42.99,
"learning_rate": 8.822222222222223e-05,
"loss": 1.2515,
"step": 51500
},
{
"epoch": 43.41,
"learning_rate": 8.800000000000001e-05,
"loss": 1.2481,
"step": 52000
},
{
"epoch": 43.82,
"learning_rate": 8.777777777777778e-05,
"loss": 1.2489,
"step": 52500
},
{
"epoch": 44.24,
"learning_rate": 8.755555555555556e-05,
"loss": 1.2431,
"step": 53000
},
{
"epoch": 44.66,
"learning_rate": 8.733333333333333e-05,
"loss": 1.242,
"step": 53500
},
{
"epoch": 45.08,
"learning_rate": 8.711111111111112e-05,
"loss": 1.2403,
"step": 54000
},
{
"epoch": 45.49,
"learning_rate": 8.68888888888889e-05,
"loss": 1.2356,
"step": 54500
},
{
"epoch": 45.91,
"learning_rate": 8.666666666666667e-05,
"loss": 1.2381,
"step": 55000
},
{
"epoch": 46.33,
"learning_rate": 8.644444444444445e-05,
"loss": 1.2336,
"step": 55500
},
{
"epoch": 46.74,
"learning_rate": 8.622222222222222e-05,
"loss": 1.2325,
"step": 56000
},
{
"epoch": 47.16,
"learning_rate": 8.6e-05,
"loss": 1.2316,
"step": 56500
},
{
"epoch": 47.58,
"learning_rate": 8.577777777777777e-05,
"loss": 1.2277,
"step": 57000
},
{
"epoch": 48.0,
"learning_rate": 8.555555555555556e-05,
"loss": 1.2287,
"step": 57500
},
{
"epoch": 48.41,
"learning_rate": 8.533333333333334e-05,
"loss": 1.2243,
"step": 58000
},
{
"epoch": 48.83,
"learning_rate": 8.511111111111112e-05,
"loss": 1.2253,
"step": 58500
},
{
"epoch": 49.25,
"learning_rate": 8.488888888888889e-05,
"loss": 1.2209,
"step": 59000
},
{
"epoch": 49.67,
"learning_rate": 8.466666666666667e-05,
"loss": 1.2209,
"step": 59500
},
{
"epoch": 50.08,
"learning_rate": 8.444444444444444e-05,
"loss": 1.2201,
"step": 60000
},
{
"epoch": 50.5,
"learning_rate": 8.422222222222223e-05,
"loss": 1.2164,
"step": 60500
},
{
"epoch": 50.92,
"learning_rate": 8.4e-05,
"loss": 1.2168,
"step": 61000
},
{
"epoch": 51.34,
"learning_rate": 8.377777777777778e-05,
"loss": 1.2139,
"step": 61500
},
{
"epoch": 51.75,
"learning_rate": 8.355555555555556e-05,
"loss": 1.212,
"step": 62000
},
{
"epoch": 52.17,
"learning_rate": 8.333333333333334e-05,
"loss": 1.211,
"step": 62500
},
{
"epoch": 52.17,
"eval_loss": 1.1704248189926147,
"eval_runtime": 125.3013,
"eval_samples_per_second": 543.801,
"eval_steps_per_second": 4.254,
"step": 62500
},
{
"epoch": 52.59,
"learning_rate": 8.311111111111111e-05,
"loss": 1.2118,
"step": 63000
},
{
"epoch": 53.01,
"learning_rate": 8.28888888888889e-05,
"loss": 1.2088,
"step": 63500
},
{
"epoch": 53.42,
"learning_rate": 8.266666666666667e-05,
"loss": 1.2054,
"step": 64000
},
{
"epoch": 53.84,
"learning_rate": 8.244444444444445e-05,
"loss": 1.2065,
"step": 64500
},
{
"epoch": 54.26,
"learning_rate": 8.222222222222222e-05,
"loss": 1.2048,
"step": 65000
},
{
"epoch": 54.67,
"learning_rate": 8.2e-05,
"loss": 1.202,
"step": 65500
},
{
"epoch": 55.09,
"learning_rate": 8.177777777777778e-05,
"loss": 1.2035,
"step": 66000
},
{
"epoch": 55.51,
"learning_rate": 8.155555555555557e-05,
"loss": 1.1989,
"step": 66500
},
{
"epoch": 55.93,
"learning_rate": 8.133333333333334e-05,
"loss": 1.201,
"step": 67000
},
{
"epoch": 56.34,
"learning_rate": 8.111111111111112e-05,
"loss": 1.1975,
"step": 67500
},
{
"epoch": 56.76,
"learning_rate": 8.088888888888889e-05,
"loss": 1.1971,
"step": 68000
},
{
"epoch": 57.18,
"learning_rate": 8.066666666666667e-05,
"loss": 1.1967,
"step": 68500
},
{
"epoch": 57.6,
"learning_rate": 8.044444444444444e-05,
"loss": 1.1931,
"step": 69000
},
{
"epoch": 58.01,
"learning_rate": 8.022222222222222e-05,
"loss": 1.1942,
"step": 69500
},
{
"epoch": 58.43,
"learning_rate": 8e-05,
"loss": 1.1901,
"step": 70000
},
{
"epoch": 58.85,
"learning_rate": 7.977777777777779e-05,
"loss": 1.1903,
"step": 70500
},
{
"epoch": 59.27,
"learning_rate": 7.955555555555556e-05,
"loss": 1.189,
"step": 71000
},
{
"epoch": 59.68,
"learning_rate": 7.933333333333334e-05,
"loss": 1.1855,
"step": 71500
},
{
"epoch": 60.1,
"learning_rate": 7.911111111111111e-05,
"loss": 1.1876,
"step": 72000
},
{
"epoch": 60.52,
"learning_rate": 7.88888888888889e-05,
"loss": 1.183,
"step": 72500
},
{
"epoch": 60.93,
"learning_rate": 7.866666666666666e-05,
"loss": 1.1853,
"step": 73000
},
{
"epoch": 61.35,
"learning_rate": 7.844444444444446e-05,
"loss": 1.1813,
"step": 73500
},
{
"epoch": 61.77,
"learning_rate": 7.822222222222223e-05,
"loss": 1.1809,
"step": 74000
},
{
"epoch": 62.19,
"learning_rate": 7.800000000000001e-05,
"loss": 1.1796,
"step": 74500
},
{
"epoch": 62.6,
"learning_rate": 7.777777777777778e-05,
"loss": 1.18,
"step": 75000
},
{
"epoch": 62.6,
"eval_loss": 1.1421879529953003,
"eval_runtime": 125.314,
"eval_samples_per_second": 543.746,
"eval_steps_per_second": 4.253,
"step": 75000
},
{
"epoch": 63.02,
"learning_rate": 7.755555555555556e-05,
"loss": 1.1821,
"step": 75500
},
{
"epoch": 63.44,
"learning_rate": 7.733333333333333e-05,
"loss": 1.1732,
"step": 76000
},
{
"epoch": 63.86,
"learning_rate": 7.711111111111112e-05,
"loss": 1.1786,
"step": 76500
},
{
"epoch": 64.27,
"learning_rate": 7.688888888888889e-05,
"loss": 1.1729,
"step": 77000
},
{
"epoch": 64.69,
"learning_rate": 7.666666666666667e-05,
"loss": 1.1737,
"step": 77500
},
{
"epoch": 65.11,
"learning_rate": 7.644444444444445e-05,
"loss": 1.1727,
"step": 78000
},
{
"epoch": 65.53,
"learning_rate": 7.622222222222223e-05,
"loss": 1.1708,
"step": 78500
},
{
"epoch": 65.94,
"learning_rate": 7.6e-05,
"loss": 1.1712,
"step": 79000
},
{
"epoch": 66.36,
"learning_rate": 7.577777777777779e-05,
"loss": 1.169,
"step": 79500
},
{
"epoch": 66.78,
"learning_rate": 7.555555555555556e-05,
"loss": 1.1697,
"step": 80000
},
{
"epoch": 67.2,
"learning_rate": 7.533333333333334e-05,
"loss": 1.1681,
"step": 80500
},
{
"epoch": 67.61,
"learning_rate": 7.511111111111111e-05,
"loss": 1.1651,
"step": 81000
},
{
"epoch": 68.03,
"learning_rate": 7.488888888888889e-05,
"loss": 1.1684,
"step": 81500
},
{
"epoch": 68.45,
"learning_rate": 7.466666666666667e-05,
"loss": 1.1644,
"step": 82000
},
{
"epoch": 68.86,
"learning_rate": 7.444444444444444e-05,
"loss": 1.1643,
"step": 82500
},
{
"epoch": 69.28,
"learning_rate": 7.422222222222223e-05,
"loss": 1.1608,
"step": 83000
},
{
"epoch": 69.7,
"learning_rate": 7.4e-05,
"loss": 1.164,
"step": 83500
},
{
"epoch": 70.12,
"learning_rate": 7.377777777777778e-05,
"loss": 1.1612,
"step": 84000
},
{
"epoch": 70.53,
"learning_rate": 7.355555555555556e-05,
"loss": 1.1593,
"step": 84500
},
{
"epoch": 70.95,
"learning_rate": 7.333333333333333e-05,
"loss": 1.1585,
"step": 85000
},
{
"epoch": 71.37,
"learning_rate": 7.311111111111111e-05,
"loss": 1.1584,
"step": 85500
},
{
"epoch": 71.79,
"learning_rate": 7.28888888888889e-05,
"loss": 1.1571,
"step": 86000
},
{
"epoch": 72.2,
"learning_rate": 7.266666666666667e-05,
"loss": 1.1565,
"step": 86500
},
{
"epoch": 72.62,
"learning_rate": 7.244444444444445e-05,
"loss": 1.1545,
"step": 87000
},
{
"epoch": 73.04,
"learning_rate": 7.222222222222222e-05,
"loss": 1.1545,
"step": 87500
},
{
"epoch": 73.04,
"eval_loss": 1.119368314743042,
"eval_runtime": 125.2645,
"eval_samples_per_second": 543.961,
"eval_steps_per_second": 4.255,
"step": 87500
},
{
"epoch": 73.46,
"learning_rate": 7.2e-05,
"loss": 1.1528,
"step": 88000
},
{
"epoch": 73.87,
"learning_rate": 7.177777777777777e-05,
"loss": 1.1546,
"step": 88500
},
{
"epoch": 74.29,
"learning_rate": 7.155555555555555e-05,
"loss": 1.152,
"step": 89000
},
{
"epoch": 74.71,
"learning_rate": 7.133333333333334e-05,
"loss": 1.15,
"step": 89500
},
{
"epoch": 75.13,
"learning_rate": 7.111111111111112e-05,
"loss": 1.1511,
"step": 90000
},
{
"epoch": 75.54,
"learning_rate": 7.088888888888889e-05,
"loss": 1.1492,
"step": 90500
},
{
"epoch": 75.96,
"learning_rate": 7.066666666666667e-05,
"loss": 1.1495,
"step": 91000
},
{
"epoch": 76.38,
"learning_rate": 7.044444444444444e-05,
"loss": 1.1466,
"step": 91500
},
{
"epoch": 76.79,
"learning_rate": 7.022222222222222e-05,
"loss": 1.1464,
"step": 92000
},
{
"epoch": 77.21,
"learning_rate": 7e-05,
"loss": 1.1458,
"step": 92500
},
{
"epoch": 77.63,
"learning_rate": 6.977777777777779e-05,
"loss": 1.1451,
"step": 93000
},
{
"epoch": 78.05,
"learning_rate": 6.955555555555556e-05,
"loss": 1.1452,
"step": 93500
},
{
"epoch": 78.46,
"learning_rate": 6.933333333333334e-05,
"loss": 1.1428,
"step": 94000
},
{
"epoch": 78.88,
"learning_rate": 6.911111111111111e-05,
"loss": 1.1439,
"step": 94500
},
{
"epoch": 79.3,
"learning_rate": 6.88888888888889e-05,
"loss": 1.1433,
"step": 95000
},
{
"epoch": 79.72,
"learning_rate": 6.866666666666666e-05,
"loss": 1.143,
"step": 95500
},
{
"epoch": 80.13,
"learning_rate": 6.844444444444445e-05,
"loss": 1.1394,
"step": 96000
},
{
"epoch": 80.55,
"learning_rate": 6.822222222222222e-05,
"loss": 1.1416,
"step": 96500
},
{
"epoch": 80.97,
"learning_rate": 6.800000000000001e-05,
"loss": 1.1392,
"step": 97000
},
{
"epoch": 81.39,
"learning_rate": 6.777777777777778e-05,
"loss": 1.1389,
"step": 97500
},
{
"epoch": 81.8,
"learning_rate": 6.755555555555557e-05,
"loss": 1.138,
"step": 98000
},
{
"epoch": 82.22,
"learning_rate": 6.733333333333333e-05,
"loss": 1.1381,
"step": 98500
},
{
"epoch": 82.64,
"learning_rate": 6.711111111111112e-05,
"loss": 1.1357,
"step": 99000
},
{
"epoch": 83.06,
"learning_rate": 6.688888888888889e-05,
"loss": 1.137,
"step": 99500
},
{
"epoch": 83.47,
"learning_rate": 6.666666666666667e-05,
"loss": 1.1324,
"step": 100000
},
{
"epoch": 83.47,
"eval_loss": 1.1047865152359009,
"eval_runtime": 125.3085,
"eval_samples_per_second": 543.77,
"eval_steps_per_second": 4.254,
"step": 100000
},
{
"epoch": 83.89,
"learning_rate": 6.644444444444444e-05,
"loss": 1.1346,
"step": 100500
},
{
"epoch": 84.31,
"learning_rate": 6.622222222222224e-05,
"loss": 1.134,
"step": 101000
},
{
"epoch": 84.72,
"learning_rate": 6.6e-05,
"loss": 1.1329,
"step": 101500
},
{
"epoch": 85.14,
"learning_rate": 6.577777777777779e-05,
"loss": 1.1338,
"step": 102000
},
{
"epoch": 85.56,
"learning_rate": 6.555555555555556e-05,
"loss": 1.1304,
"step": 102500
},
{
"epoch": 85.98,
"learning_rate": 6.533333333333334e-05,
"loss": 1.1303,
"step": 103000
},
{
"epoch": 86.39,
"learning_rate": 6.511111111111111e-05,
"loss": 1.1299,
"step": 103500
},
{
"epoch": 86.81,
"learning_rate": 6.488888888888889e-05,
"loss": 1.1298,
"step": 104000
},
{
"epoch": 87.23,
"learning_rate": 6.466666666666666e-05,
"loss": 1.1276,
"step": 104500
},
{
"epoch": 87.65,
"learning_rate": 6.444444444444446e-05,
"loss": 1.1285,
"step": 105000
},
{
"epoch": 88.06,
"learning_rate": 6.422222222222223e-05,
"loss": 1.1263,
"step": 105500
},
{
"epoch": 88.48,
"learning_rate": 6.400000000000001e-05,
"loss": 1.1243,
"step": 106000
},
{
"epoch": 88.9,
"learning_rate": 6.377777777777778e-05,
"loss": 1.1267,
"step": 106500
},
{
"epoch": 89.32,
"learning_rate": 6.355555555555556e-05,
"loss": 1.1236,
"step": 107000
},
{
"epoch": 89.73,
"learning_rate": 6.333333333333333e-05,
"loss": 1.1242,
"step": 107500
},
{
"epoch": 90.15,
"learning_rate": 6.311111111111112e-05,
"loss": 1.1226,
"step": 108000
},
{
"epoch": 90.57,
"learning_rate": 6.28888888888889e-05,
"loss": 1.1216,
"step": 108500
},
{
"epoch": 90.98,
"learning_rate": 6.266666666666667e-05,
"loss": 1.1233,
"step": 109000
},
{
"epoch": 91.4,
"learning_rate": 6.244444444444445e-05,
"loss": 1.1216,
"step": 109500
},
{
"epoch": 91.82,
"learning_rate": 6.222222222222222e-05,
"loss": 1.1229,
"step": 110000
},
{
"epoch": 92.24,
"learning_rate": 6.2e-05,
"loss": 1.1203,
"step": 110500
},
{
"epoch": 92.65,
"learning_rate": 6.177777777777779e-05,
"loss": 1.1184,
"step": 111000
},
{
"epoch": 93.07,
"learning_rate": 6.155555555555555e-05,
"loss": 1.1217,
"step": 111500
},
{
"epoch": 93.49,
"learning_rate": 6.133333333333334e-05,
"loss": 1.1177,
"step": 112000
},
{
"epoch": 93.91,
"learning_rate": 6.111111111111112e-05,
"loss": 1.1171,
"step": 112500
},
{
"epoch": 93.91,
"eval_loss": 1.0929380655288696,
"eval_runtime": 125.1906,
"eval_samples_per_second": 544.282,
"eval_steps_per_second": 4.258,
"step": 112500
},
{
"epoch": 94.32,
"learning_rate": 6.08888888888889e-05,
"loss": 1.1157,
"step": 113000
},
{
"epoch": 94.74,
"learning_rate": 6.066666666666667e-05,
"loss": 1.1171,
"step": 113500
},
{
"epoch": 95.16,
"learning_rate": 6.044444444444445e-05,
"loss": 1.1163,
"step": 114000
},
{
"epoch": 95.58,
"learning_rate": 6.0222222222222225e-05,
"loss": 1.1163,
"step": 114500
},
{
"epoch": 95.99,
"learning_rate": 6e-05,
"loss": 1.1163,
"step": 115000
},
{
"epoch": 96.41,
"learning_rate": 5.977777777777778e-05,
"loss": 1.1124,
"step": 115500
},
{
"epoch": 96.83,
"learning_rate": 5.9555555555555554e-05,
"loss": 1.1139,
"step": 116000
},
{
"epoch": 97.25,
"learning_rate": 5.9333333333333343e-05,
"loss": 1.1134,
"step": 116500
},
{
"epoch": 97.66,
"learning_rate": 5.911111111111112e-05,
"loss": 1.1119,
"step": 117000
},
{
"epoch": 98.08,
"learning_rate": 5.8888888888888896e-05,
"loss": 1.1138,
"step": 117500
},
{
"epoch": 98.5,
"learning_rate": 5.866666666666667e-05,
"loss": 1.1088,
"step": 118000
},
{
"epoch": 98.91,
"learning_rate": 5.844444444444445e-05,
"loss": 1.112,
"step": 118500
},
{
"epoch": 99.33,
"learning_rate": 5.8222222222222224e-05,
"loss": 1.1076,
"step": 119000
},
{
"epoch": 99.75,
"learning_rate": 5.8e-05,
"loss": 1.1097,
"step": 119500
},
{
"epoch": 100.17,
"learning_rate": 5.7777777777777776e-05,
"loss": 1.1081,
"step": 120000
},
{
"epoch": 100.58,
"learning_rate": 5.755555555555556e-05,
"loss": 1.1051,
"step": 120500
},
{
"epoch": 101.0,
"learning_rate": 5.7333333333333336e-05,
"loss": 1.1101,
"step": 121000
},
{
"epoch": 101.42,
"learning_rate": 5.711111111111112e-05,
"loss": 1.104,
"step": 121500
},
{
"epoch": 101.84,
"learning_rate": 5.6888888888888895e-05,
"loss": 1.108,
"step": 122000
},
{
"epoch": 102.25,
"learning_rate": 5.666666666666667e-05,
"loss": 1.1058,
"step": 122500
},
{
"epoch": 102.67,
"learning_rate": 5.644444444444445e-05,
"loss": 1.1069,
"step": 123000
},
{
"epoch": 103.09,
"learning_rate": 5.622222222222222e-05,
"loss": 1.1042,
"step": 123500
},
{
"epoch": 103.51,
"learning_rate": 5.6000000000000006e-05,
"loss": 1.1031,
"step": 124000
},
{
"epoch": 103.92,
"learning_rate": 5.577777777777778e-05,
"loss": 1.1032,
"step": 124500
},
{
"epoch": 104.34,
"learning_rate": 5.555555555555556e-05,
"loss": 1.1044,
"step": 125000
},
{
"epoch": 104.34,
"eval_loss": 1.0788438320159912,
"eval_runtime": 125.2368,
"eval_samples_per_second": 544.081,
"eval_steps_per_second": 4.256,
"step": 125000
},
{
"epoch": 104.76,
"learning_rate": 5.5333333333333334e-05,
"loss": 1.1023,
"step": 125500
},
{
"epoch": 105.18,
"learning_rate": 5.511111111111111e-05,
"loss": 1.1028,
"step": 126000
},
{
"epoch": 105.59,
"learning_rate": 5.488888888888889e-05,
"loss": 1.1027,
"step": 126500
},
{
"epoch": 106.01,
"learning_rate": 5.466666666666666e-05,
"loss": 1.1019,
"step": 127000
},
{
"epoch": 106.43,
"learning_rate": 5.4444444444444446e-05,
"loss": 1.0994,
"step": 127500
},
{
"epoch": 106.84,
"learning_rate": 5.422222222222223e-05,
"loss": 1.1012,
"step": 128000
},
{
"epoch": 107.26,
"learning_rate": 5.4000000000000005e-05,
"loss": 1.0995,
"step": 128500
},
{
"epoch": 107.68,
"learning_rate": 5.377777777777778e-05,
"loss": 1.0994,
"step": 129000
},
{
"epoch": 108.1,
"learning_rate": 5.355555555555556e-05,
"loss": 1.0999,
"step": 129500
},
{
"epoch": 108.51,
"learning_rate": 5.333333333333333e-05,
"loss": 1.0971,
"step": 130000
},
{
"epoch": 108.93,
"learning_rate": 5.311111111111111e-05,
"loss": 1.0982,
"step": 130500
},
{
"epoch": 109.35,
"learning_rate": 5.2888888888888885e-05,
"loss": 1.0954,
"step": 131000
},
{
"epoch": 109.77,
"learning_rate": 5.266666666666666e-05,
"loss": 1.0979,
"step": 131500
},
{
"epoch": 110.18,
"learning_rate": 5.244444444444445e-05,
"loss": 1.0966,
"step": 132000
},
{
"epoch": 110.6,
"learning_rate": 5.222222222222223e-05,
"loss": 1.0954,
"step": 132500
},
{
"epoch": 111.02,
"learning_rate": 5.2000000000000004e-05,
"loss": 1.0959,
"step": 133000
},
{
"epoch": 111.44,
"learning_rate": 5.177777777777778e-05,
"loss": 1.0943,
"step": 133500
},
{
"epoch": 111.85,
"learning_rate": 5.1555555555555556e-05,
"loss": 1.0941,
"step": 134000
},
{
"epoch": 112.27,
"learning_rate": 5.133333333333333e-05,
"loss": 1.0937,
"step": 134500
},
{
"epoch": 112.69,
"learning_rate": 5.111111111111111e-05,
"loss": 1.093,
"step": 135000
},
{
"epoch": 113.11,
"learning_rate": 5.0888888888888884e-05,
"loss": 1.0931,
"step": 135500
},
{
"epoch": 113.52,
"learning_rate": 5.0666666666666674e-05,
"loss": 1.0901,
"step": 136000
},
{
"epoch": 113.94,
"learning_rate": 5.044444444444445e-05,
"loss": 1.0948,
"step": 136500
},
{
"epoch": 114.36,
"learning_rate": 5.0222222222222226e-05,
"loss": 1.0903,
"step": 137000
},
{
"epoch": 114.77,
"learning_rate": 5e-05,
"loss": 1.0922,
"step": 137500
},
{
"epoch": 114.77,
"eval_loss": 1.0676383972167969,
"eval_runtime": 125.398,
"eval_samples_per_second": 543.382,
"eval_steps_per_second": 4.25,
"step": 137500
},
{
"epoch": 115.19,
"learning_rate": 4.977777777777778e-05,
"loss": 1.0883,
"step": 138000
},
{
"epoch": 115.61,
"learning_rate": 4.955555555555556e-05,
"loss": 1.0905,
"step": 138500
},
{
"epoch": 116.03,
"learning_rate": 4.933333333333334e-05,
"loss": 1.0912,
"step": 139000
},
{
"epoch": 116.44,
"learning_rate": 4.9111111111111114e-05,
"loss": 1.0873,
"step": 139500
},
{
"epoch": 116.86,
"learning_rate": 4.888888888888889e-05,
"loss": 1.0893,
"step": 140000
},
{
"epoch": 117.28,
"learning_rate": 4.866666666666667e-05,
"loss": 1.0857,
"step": 140500
},
{
"epoch": 117.7,
"learning_rate": 4.844444444444445e-05,
"loss": 1.0877,
"step": 141000
},
{
"epoch": 118.11,
"learning_rate": 4.8222222222222225e-05,
"loss": 1.0879,
"step": 141500
},
{
"epoch": 118.53,
"learning_rate": 4.8e-05,
"loss": 1.0858,
"step": 142000
},
{
"epoch": 118.95,
"learning_rate": 4.7777777777777784e-05,
"loss": 1.0876,
"step": 142500
},
{
"epoch": 119.37,
"learning_rate": 4.755555555555556e-05,
"loss": 1.0849,
"step": 143000
},
{
"epoch": 119.78,
"learning_rate": 4.7333333333333336e-05,
"loss": 1.0851,
"step": 143500
},
{
"epoch": 120.2,
"learning_rate": 4.711111111111111e-05,
"loss": 1.0833,
"step": 144000
},
{
"epoch": 120.62,
"learning_rate": 4.6888888888888895e-05,
"loss": 1.0841,
"step": 144500
},
{
"epoch": 121.04,
"learning_rate": 4.666666666666667e-05,
"loss": 1.0848,
"step": 145000
},
{
"epoch": 121.45,
"learning_rate": 4.644444444444445e-05,
"loss": 1.0832,
"step": 145500
},
{
"epoch": 121.87,
"learning_rate": 4.6222222222222224e-05,
"loss": 1.0829,
"step": 146000
},
{
"epoch": 122.29,
"learning_rate": 4.600000000000001e-05,
"loss": 1.0832,
"step": 146500
},
{
"epoch": 122.7,
"learning_rate": 4.577777777777778e-05,
"loss": 1.0804,
"step": 147000
},
{
"epoch": 123.12,
"learning_rate": 4.555555555555556e-05,
"loss": 1.084,
"step": 147500
},
{
"epoch": 123.54,
"learning_rate": 4.5333333333333335e-05,
"loss": 1.0809,
"step": 148000
},
{
"epoch": 123.96,
"learning_rate": 4.511111111111112e-05,
"loss": 1.0816,
"step": 148500
},
{
"epoch": 124.37,
"learning_rate": 4.4888888888888894e-05,
"loss": 1.0781,
"step": 149000
},
{
"epoch": 124.79,
"learning_rate": 4.466666666666667e-05,
"loss": 1.0811,
"step": 149500
},
{
"epoch": 125.21,
"learning_rate": 4.4444444444444447e-05,
"loss": 1.0772,
"step": 150000
},
{
"epoch": 125.21,
"eval_loss": 1.0623060464859009,
"eval_runtime": 125.1984,
"eval_samples_per_second": 544.248,
"eval_steps_per_second": 4.257,
"step": 150000
},
{
"epoch": 125.63,
"learning_rate": 4.422222222222222e-05,
"loss": 1.0786,
"step": 150500
},
{
"epoch": 126.04,
"learning_rate": 4.4000000000000006e-05,
"loss": 1.0799,
"step": 151000
},
{
"epoch": 126.46,
"learning_rate": 4.377777777777778e-05,
"loss": 1.0782,
"step": 151500
},
{
"epoch": 126.88,
"learning_rate": 4.355555555555556e-05,
"loss": 1.0788,
"step": 152000
},
{
"epoch": 127.3,
"learning_rate": 4.3333333333333334e-05,
"loss": 1.0769,
"step": 152500
},
{
"epoch": 127.71,
"learning_rate": 4.311111111111111e-05,
"loss": 1.0752,
"step": 153000
},
{
"epoch": 128.13,
"learning_rate": 4.2888888888888886e-05,
"loss": 1.0765,
"step": 153500
},
{
"epoch": 128.55,
"learning_rate": 4.266666666666667e-05,
"loss": 1.075,
"step": 154000
},
{
"epoch": 128.96,
"learning_rate": 4.2444444444444445e-05,
"loss": 1.0749,
"step": 154500
},
{
"epoch": 129.38,
"learning_rate": 4.222222222222222e-05,
"loss": 1.0743,
"step": 155000
},
{
"epoch": 129.8,
"learning_rate": 4.2e-05,
"loss": 1.0732,
"step": 155500
},
{
"epoch": 130.22,
"learning_rate": 4.177777777777778e-05,
"loss": 1.0753,
"step": 156000
},
{
"epoch": 130.63,
"learning_rate": 4.155555555555556e-05,
"loss": 1.0745,
"step": 156500
},
{
"epoch": 131.05,
"learning_rate": 4.133333333333333e-05,
"loss": 1.0738,
"step": 157000
},
{
"epoch": 131.47,
"learning_rate": 4.111111111111111e-05,
"loss": 1.073,
"step": 157500
},
{
"epoch": 131.89,
"learning_rate": 4.088888888888889e-05,
"loss": 1.0712,
"step": 158000
},
{
"epoch": 132.3,
"learning_rate": 4.066666666666667e-05,
"loss": 1.0708,
"step": 158500
},
{
"epoch": 132.72,
"learning_rate": 4.0444444444444444e-05,
"loss": 1.072,
"step": 159000
},
{
"epoch": 133.14,
"learning_rate": 4.022222222222222e-05,
"loss": 1.0724,
"step": 159500
},
{
"epoch": 133.56,
"learning_rate": 4e-05,
"loss": 1.0711,
"step": 160000
},
{
"epoch": 133.97,
"learning_rate": 3.977777777777778e-05,
"loss": 1.0713,
"step": 160500
},
{
"epoch": 134.39,
"learning_rate": 3.9555555555555556e-05,
"loss": 1.0692,
"step": 161000
},
{
"epoch": 134.81,
"learning_rate": 3.933333333333333e-05,
"loss": 1.0701,
"step": 161500
},
{
"epoch": 135.23,
"learning_rate": 3.9111111111111115e-05,
"loss": 1.0701,
"step": 162000
},
{
"epoch": 135.64,
"learning_rate": 3.888888888888889e-05,
"loss": 1.0709,
"step": 162500
},
{
"epoch": 135.64,
"eval_loss": 1.0541090965270996,
"eval_runtime": 125.3801,
"eval_samples_per_second": 543.46,
"eval_steps_per_second": 4.251,
"step": 162500
},
{
"epoch": 136.06,
"learning_rate": 3.866666666666667e-05,
"loss": 1.0692,
"step": 163000
},
{
"epoch": 136.48,
"learning_rate": 3.844444444444444e-05,
"loss": 1.0683,
"step": 163500
},
{
"epoch": 136.89,
"learning_rate": 3.8222222222222226e-05,
"loss": 1.0682,
"step": 164000
},
{
"epoch": 137.31,
"learning_rate": 3.8e-05,
"loss": 1.069,
"step": 164500
},
{
"epoch": 137.73,
"learning_rate": 3.777777777777778e-05,
"loss": 1.0683,
"step": 165000
},
{
"epoch": 138.15,
"learning_rate": 3.7555555555555554e-05,
"loss": 1.0677,
"step": 165500
},
{
"epoch": 138.56,
"learning_rate": 3.733333333333334e-05,
"loss": 1.0666,
"step": 166000
},
{
"epoch": 138.98,
"learning_rate": 3.7111111111111113e-05,
"loss": 1.0656,
"step": 166500
},
{
"epoch": 139.4,
"learning_rate": 3.688888888888889e-05,
"loss": 1.0646,
"step": 167000
},
{
"epoch": 139.82,
"learning_rate": 3.6666666666666666e-05,
"loss": 1.0667,
"step": 167500
},
{
"epoch": 140.23,
"learning_rate": 3.644444444444445e-05,
"loss": 1.0646,
"step": 168000
},
{
"epoch": 140.65,
"learning_rate": 3.6222222222222225e-05,
"loss": 1.0635,
"step": 168500
},
{
"epoch": 141.07,
"learning_rate": 3.6e-05,
"loss": 1.066,
"step": 169000
},
{
"epoch": 141.49,
"learning_rate": 3.577777777777778e-05,
"loss": 1.0618,
"step": 169500
},
{
"epoch": 141.9,
"learning_rate": 3.555555555555556e-05,
"loss": 1.0641,
"step": 170000
},
{
"epoch": 142.32,
"learning_rate": 3.5333333333333336e-05,
"loss": 1.0626,
"step": 170500
},
{
"epoch": 142.74,
"learning_rate": 3.511111111111111e-05,
"loss": 1.0622,
"step": 171000
},
{
"epoch": 143.16,
"learning_rate": 3.4888888888888895e-05,
"loss": 1.0623,
"step": 171500
},
{
"epoch": 143.57,
"learning_rate": 3.466666666666667e-05,
"loss": 1.0636,
"step": 172000
},
{
"epoch": 143.99,
"learning_rate": 3.444444444444445e-05,
"loss": 1.0624,
"step": 172500
},
{
"epoch": 144.41,
"learning_rate": 3.4222222222222224e-05,
"loss": 1.0625,
"step": 173000
},
{
"epoch": 144.82,
"learning_rate": 3.4000000000000007e-05,
"loss": 1.061,
"step": 173500
},
{
"epoch": 145.24,
"learning_rate": 3.377777777777778e-05,
"loss": 1.061,
"step": 174000
},
{
"epoch": 145.66,
"learning_rate": 3.355555555555556e-05,
"loss": 1.0604,
"step": 174500
},
{
"epoch": 146.08,
"learning_rate": 3.3333333333333335e-05,
"loss": 1.0612,
"step": 175000
},
{
"epoch": 146.08,
"eval_loss": 1.0464940071105957,
"eval_runtime": 125.4785,
"eval_samples_per_second": 543.033,
"eval_steps_per_second": 4.248,
"step": 175000
},
{
"epoch": 146.49,
"learning_rate": 3.311111111111112e-05,
"loss": 1.0602,
"step": 175500
},
{
"epoch": 146.91,
"learning_rate": 3.2888888888888894e-05,
"loss": 1.0588,
"step": 176000
},
{
"epoch": 147.33,
"learning_rate": 3.266666666666667e-05,
"loss": 1.0601,
"step": 176500
},
{
"epoch": 147.75,
"learning_rate": 3.2444444444444446e-05,
"loss": 1.0575,
"step": 177000
},
{
"epoch": 148.16,
"learning_rate": 3.222222222222223e-05,
"loss": 1.0583,
"step": 177500
},
{
"epoch": 148.58,
"learning_rate": 3.2000000000000005e-05,
"loss": 1.0574,
"step": 178000
},
{
"epoch": 149.0,
"learning_rate": 3.177777777777778e-05,
"loss": 1.0572,
"step": 178500
},
{
"epoch": 149.42,
"learning_rate": 3.155555555555556e-05,
"loss": 1.0574,
"step": 179000
},
{
"epoch": 149.83,
"learning_rate": 3.1333333333333334e-05,
"loss": 1.0568,
"step": 179500
},
{
"epoch": 150.25,
"learning_rate": 3.111111111111111e-05,
"loss": 1.057,
"step": 180000
},
{
"epoch": 150.67,
"learning_rate": 3.088888888888889e-05,
"loss": 1.0571,
"step": 180500
},
{
"epoch": 151.09,
"learning_rate": 3.066666666666667e-05,
"loss": 1.0571,
"step": 181000
},
{
"epoch": 151.5,
"learning_rate": 3.044444444444445e-05,
"loss": 1.0555,
"step": 181500
},
{
"epoch": 151.92,
"learning_rate": 3.0222222222222225e-05,
"loss": 1.0558,
"step": 182000
},
{
"epoch": 152.34,
"learning_rate": 3e-05,
"loss": 1.0538,
"step": 182500
},
{
"epoch": 152.75,
"learning_rate": 2.9777777777777777e-05,
"loss": 1.0547,
"step": 183000
},
{
"epoch": 153.17,
"learning_rate": 2.955555555555556e-05,
"loss": 1.0545,
"step": 183500
},
{
"epoch": 153.59,
"learning_rate": 2.9333333333333336e-05,
"loss": 1.0544,
"step": 184000
},
{
"epoch": 154.01,
"learning_rate": 2.9111111111111112e-05,
"loss": 1.0543,
"step": 184500
},
{
"epoch": 154.42,
"learning_rate": 2.8888888888888888e-05,
"loss": 1.0527,
"step": 185000
},
{
"epoch": 154.84,
"learning_rate": 2.8666666666666668e-05,
"loss": 1.0528,
"step": 185500
},
{
"epoch": 155.26,
"learning_rate": 2.8444444444444447e-05,
"loss": 1.0527,
"step": 186000
},
{
"epoch": 155.68,
"learning_rate": 2.8222222222222223e-05,
"loss": 1.0527,
"step": 186500
},
{
"epoch": 156.09,
"learning_rate": 2.8000000000000003e-05,
"loss": 1.0523,
"step": 187000
},
{
"epoch": 156.51,
"learning_rate": 2.777777777777778e-05,
"loss": 1.0519,
"step": 187500
},
{
"epoch": 156.51,
"eval_loss": 1.0381312370300293,
"eval_runtime": 125.1331,
"eval_samples_per_second": 544.532,
"eval_steps_per_second": 4.259,
"step": 187500
},
{
"epoch": 156.93,
"learning_rate": 2.7555555555555555e-05,
"loss": 1.0498,
"step": 188000
},
{
"epoch": 157.35,
"learning_rate": 2.733333333333333e-05,
"loss": 1.0497,
"step": 188500
},
{
"epoch": 157.76,
"learning_rate": 2.7111111111111114e-05,
"loss": 1.0521,
"step": 189000
},
{
"epoch": 158.18,
"learning_rate": 2.688888888888889e-05,
"loss": 1.0511,
"step": 189500
},
{
"epoch": 158.6,
"learning_rate": 2.6666666666666667e-05,
"loss": 1.0495,
"step": 190000
},
{
"epoch": 159.02,
"learning_rate": 2.6444444444444443e-05,
"loss": 1.0499,
"step": 190500
},
{
"epoch": 159.43,
"learning_rate": 2.6222222222222226e-05,
"loss": 1.0472,
"step": 191000
},
{
"epoch": 159.85,
"learning_rate": 2.6000000000000002e-05,
"loss": 1.0499,
"step": 191500
},
{
"epoch": 160.27,
"learning_rate": 2.5777777777777778e-05,
"loss": 1.0485,
"step": 192000
},
{
"epoch": 160.68,
"learning_rate": 2.5555555555555554e-05,
"loss": 1.0483,
"step": 192500
},
{
"epoch": 161.1,
"learning_rate": 2.5333333333333337e-05,
"loss": 1.0494,
"step": 193000
},
{
"epoch": 161.52,
"learning_rate": 2.5111111111111113e-05,
"loss": 1.0469,
"step": 193500
},
{
"epoch": 161.94,
"learning_rate": 2.488888888888889e-05,
"loss": 1.0496,
"step": 194000
},
{
"epoch": 162.35,
"learning_rate": 2.466666666666667e-05,
"loss": 1.0466,
"step": 194500
},
{
"epoch": 162.77,
"learning_rate": 2.4444444444444445e-05,
"loss": 1.0475,
"step": 195000
},
{
"epoch": 163.19,
"learning_rate": 2.4222222222222224e-05,
"loss": 1.0472,
"step": 195500
},
{
"epoch": 163.61,
"learning_rate": 2.4e-05,
"loss": 1.0462,
"step": 196000
},
{
"epoch": 164.02,
"learning_rate": 2.377777777777778e-05,
"loss": 1.0459,
"step": 196500
},
{
"epoch": 164.44,
"learning_rate": 2.3555555555555556e-05,
"loss": 1.0456,
"step": 197000
},
{
"epoch": 164.86,
"learning_rate": 2.3333333333333336e-05,
"loss": 1.0451,
"step": 197500
},
{
"epoch": 165.28,
"learning_rate": 2.3111111111111112e-05,
"loss": 1.0464,
"step": 198000
},
{
"epoch": 165.69,
"learning_rate": 2.288888888888889e-05,
"loss": 1.0441,
"step": 198500
},
{
"epoch": 166.11,
"learning_rate": 2.2666666666666668e-05,
"loss": 1.0442,
"step": 199000
},
{
"epoch": 166.53,
"learning_rate": 2.2444444444444447e-05,
"loss": 1.0459,
"step": 199500
},
{
"epoch": 166.94,
"learning_rate": 2.2222222222222223e-05,
"loss": 1.0434,
"step": 200000
},
{
"epoch": 166.94,
"eval_loss": 1.0337976217269897,
"eval_runtime": 125.3173,
"eval_samples_per_second": 543.732,
"eval_steps_per_second": 4.253,
"step": 200000
},
{
"epoch": 167.36,
"learning_rate": 2.2000000000000003e-05,
"loss": 1.0442,
"step": 200500
},
{
"epoch": 167.78,
"learning_rate": 2.177777777777778e-05,
"loss": 1.0451,
"step": 201000
},
{
"epoch": 168.2,
"learning_rate": 2.1555555555555555e-05,
"loss": 1.0426,
"step": 201500
},
{
"epoch": 168.61,
"learning_rate": 2.1333333333333335e-05,
"loss": 1.0439,
"step": 202000
},
{
"epoch": 169.03,
"learning_rate": 2.111111111111111e-05,
"loss": 1.0437,
"step": 202500
},
{
"epoch": 169.45,
"learning_rate": 2.088888888888889e-05,
"loss": 1.0437,
"step": 203000
},
{
"epoch": 169.87,
"learning_rate": 2.0666666666666666e-05,
"loss": 1.043,
"step": 203500
},
{
"epoch": 170.28,
"learning_rate": 2.0444444444444446e-05,
"loss": 1.0435,
"step": 204000
},
{
"epoch": 170.7,
"learning_rate": 2.0222222222222222e-05,
"loss": 1.0418,
"step": 204500
},
{
"epoch": 171.12,
"learning_rate": 2e-05,
"loss": 1.0424,
"step": 205000
},
{
"epoch": 171.54,
"learning_rate": 1.9777777777777778e-05,
"loss": 1.0412,
"step": 205500
},
{
"epoch": 171.95,
"learning_rate": 1.9555555555555557e-05,
"loss": 1.0415,
"step": 206000
},
{
"epoch": 172.37,
"learning_rate": 1.9333333333333333e-05,
"loss": 1.0423,
"step": 206500
},
{
"epoch": 172.79,
"learning_rate": 1.9111111111111113e-05,
"loss": 1.0414,
"step": 207000
},
{
"epoch": 173.21,
"learning_rate": 1.888888888888889e-05,
"loss": 1.0395,
"step": 207500
},
{
"epoch": 173.62,
"learning_rate": 1.866666666666667e-05,
"loss": 1.0386,
"step": 208000
},
{
"epoch": 174.04,
"learning_rate": 1.8444444444444445e-05,
"loss": 1.0417,
"step": 208500
},
{
"epoch": 174.46,
"learning_rate": 1.8222222222222224e-05,
"loss": 1.0396,
"step": 209000
},
{
"epoch": 174.87,
"learning_rate": 1.8e-05,
"loss": 1.0387,
"step": 209500
},
{
"epoch": 175.29,
"learning_rate": 1.777777777777778e-05,
"loss": 1.0375,
"step": 210000
},
{
"epoch": 175.71,
"learning_rate": 1.7555555555555556e-05,
"loss": 1.0386,
"step": 210500
},
{
"epoch": 176.13,
"learning_rate": 1.7333333333333336e-05,
"loss": 1.0392,
"step": 211000
},
{
"epoch": 176.54,
"learning_rate": 1.7111111111111112e-05,
"loss": 1.0373,
"step": 211500
},
{
"epoch": 176.96,
"learning_rate": 1.688888888888889e-05,
"loss": 1.0375,
"step": 212000
},
{
"epoch": 177.38,
"learning_rate": 1.6666666666666667e-05,
"loss": 1.0368,
"step": 212500
},
{
"epoch": 177.38,
"eval_loss": 1.0263112783432007,
"eval_runtime": 125.2736,
"eval_samples_per_second": 543.921,
"eval_steps_per_second": 4.255,
"step": 212500
},
{
"epoch": 177.8,
"learning_rate": 1.6444444444444447e-05,
"loss": 1.0362,
"step": 213000
},
{
"epoch": 178.21,
"learning_rate": 1.6222222222222223e-05,
"loss": 1.0365,
"step": 213500
},
{
"epoch": 178.63,
"learning_rate": 1.6000000000000003e-05,
"loss": 1.0369,
"step": 214000
},
{
"epoch": 179.05,
"learning_rate": 1.577777777777778e-05,
"loss": 1.0358,
"step": 214500
},
{
"epoch": 179.47,
"learning_rate": 1.5555555555555555e-05,
"loss": 1.0361,
"step": 215000
},
{
"epoch": 179.88,
"learning_rate": 1.5333333333333334e-05,
"loss": 1.0374,
"step": 215500
},
{
"epoch": 180.3,
"learning_rate": 1.5111111111111112e-05,
"loss": 1.0342,
"step": 216000
},
{
"epoch": 180.72,
"learning_rate": 1.4888888888888888e-05,
"loss": 1.0363,
"step": 216500
},
{
"epoch": 181.14,
"learning_rate": 1.4666666666666668e-05,
"loss": 1.0366,
"step": 217000
},
{
"epoch": 181.55,
"learning_rate": 1.4444444444444444e-05,
"loss": 1.0354,
"step": 217500
},
{
"epoch": 181.97,
"learning_rate": 1.4222222222222224e-05,
"loss": 1.0331,
"step": 218000
},
{
"epoch": 182.39,
"learning_rate": 1.4000000000000001e-05,
"loss": 1.0354,
"step": 218500
},
{
"epoch": 182.8,
"learning_rate": 1.3777777777777778e-05,
"loss": 1.0331,
"step": 219000
},
{
"epoch": 183.22,
"learning_rate": 1.3555555555555557e-05,
"loss": 1.0326,
"step": 219500
},
{
"epoch": 183.64,
"learning_rate": 1.3333333333333333e-05,
"loss": 1.0337,
"step": 220000
},
{
"epoch": 184.06,
"learning_rate": 1.3111111111111113e-05,
"loss": 1.0347,
"step": 220500
},
{
"epoch": 184.47,
"learning_rate": 1.2888888888888889e-05,
"loss": 1.0331,
"step": 221000
},
{
"epoch": 184.89,
"learning_rate": 1.2666666666666668e-05,
"loss": 1.034,
"step": 221500
},
{
"epoch": 185.31,
"learning_rate": 1.2444444444444445e-05,
"loss": 1.0324,
"step": 222000
},
{
"epoch": 185.73,
"learning_rate": 1.2222222222222222e-05,
"loss": 1.0333,
"step": 222500
},
{
"epoch": 186.14,
"learning_rate": 1.2e-05,
"loss": 1.0326,
"step": 223000
},
{
"epoch": 186.56,
"learning_rate": 1.1777777777777778e-05,
"loss": 1.0329,
"step": 223500
},
{
"epoch": 186.98,
"learning_rate": 1.1555555555555556e-05,
"loss": 1.0334,
"step": 224000
},
{
"epoch": 187.4,
"learning_rate": 1.1333333333333334e-05,
"loss": 1.0301,
"step": 224500
},
{
"epoch": 187.81,
"learning_rate": 1.1111111111111112e-05,
"loss": 1.032,
"step": 225000
},
{
"epoch": 187.81,
"eval_loss": 1.0231534242630005,
"eval_runtime": 125.0456,
"eval_samples_per_second": 544.913,
"eval_steps_per_second": 4.262,
"step": 225000
},
{
"epoch": 188.23,
"learning_rate": 1.088888888888889e-05,
"loss": 1.0321,
"step": 225500
},
{
"epoch": 188.65,
"learning_rate": 1.0666666666666667e-05,
"loss": 1.0316,
"step": 226000
},
{
"epoch": 189.07,
"learning_rate": 1.0444444444444445e-05,
"loss": 1.03,
"step": 226500
},
{
"epoch": 189.48,
"learning_rate": 1.0222222222222223e-05,
"loss": 1.0279,
"step": 227000
},
{
"epoch": 189.9,
"learning_rate": 1e-05,
"loss": 1.0311,
"step": 227500
},
{
"epoch": 190.32,
"learning_rate": 9.777777777777779e-06,
"loss": 1.0294,
"step": 228000
},
{
"epoch": 190.73,
"learning_rate": 9.555555555555556e-06,
"loss": 1.0284,
"step": 228500
},
{
"epoch": 191.15,
"learning_rate": 9.333333333333334e-06,
"loss": 1.0309,
"step": 229000
},
{
"epoch": 191.57,
"learning_rate": 9.111111111111112e-06,
"loss": 1.0288,
"step": 229500
},
{
"epoch": 191.99,
"learning_rate": 8.88888888888889e-06,
"loss": 1.0301,
"step": 230000
},
{
"epoch": 192.4,
"learning_rate": 8.666666666666668e-06,
"loss": 1.0304,
"step": 230500
},
{
"epoch": 192.82,
"learning_rate": 8.444444444444446e-06,
"loss": 1.0292,
"step": 231000
},
{
"epoch": 193.24,
"learning_rate": 8.222222222222223e-06,
"loss": 1.0288,
"step": 231500
},
{
"epoch": 193.66,
"learning_rate": 8.000000000000001e-06,
"loss": 1.0285,
"step": 232000
},
{
"epoch": 194.07,
"learning_rate": 7.777777777777777e-06,
"loss": 1.0281,
"step": 232500
},
{
"epoch": 194.49,
"learning_rate": 7.555555555555556e-06,
"loss": 1.0294,
"step": 233000
},
{
"epoch": 194.91,
"learning_rate": 7.333333333333334e-06,
"loss": 1.0297,
"step": 233500
},
{
"epoch": 195.33,
"learning_rate": 7.111111111111112e-06,
"loss": 1.0298,
"step": 234000
},
{
"epoch": 195.74,
"learning_rate": 6.888888888888889e-06,
"loss": 1.0302,
"step": 234500
},
{
"epoch": 196.16,
"learning_rate": 6.666666666666667e-06,
"loss": 1.0281,
"step": 235000
},
{
"epoch": 196.58,
"learning_rate": 6.4444444444444445e-06,
"loss": 1.0273,
"step": 235500
},
{
"epoch": 196.99,
"learning_rate": 6.222222222222222e-06,
"loss": 1.0261,
"step": 236000
},
{
"epoch": 197.41,
"learning_rate": 6e-06,
"loss": 1.0259,
"step": 236500
},
{
"epoch": 197.83,
"learning_rate": 5.777777777777778e-06,
"loss": 1.0277,
"step": 237000
},
{
"epoch": 198.25,
"learning_rate": 5.555555555555556e-06,
"loss": 1.0271,
"step": 237500
},
{
"epoch": 198.25,
"eval_loss": 1.0194706916809082,
"eval_runtime": 125.2786,
"eval_samples_per_second": 543.9,
"eval_steps_per_second": 4.255,
"step": 237500
},
{
"epoch": 198.66,
"learning_rate": 5.333333333333334e-06,
"loss": 1.0263,
"step": 238000
},
{
"epoch": 199.08,
"learning_rate": 5.1111111111111115e-06,
"loss": 1.0266,
"step": 238500
},
{
"epoch": 199.5,
"learning_rate": 4.888888888888889e-06,
"loss": 1.0254,
"step": 239000
},
{
"epoch": 199.92,
"learning_rate": 4.666666666666667e-06,
"loss": 1.0272,
"step": 239500
},
{
"epoch": 200.33,
"learning_rate": 4.444444444444445e-06,
"loss": 1.0253,
"step": 240000
},
{
"epoch": 200.75,
"learning_rate": 4.222222222222223e-06,
"loss": 1.026,
"step": 240500
},
{
"epoch": 201.17,
"learning_rate": 4.000000000000001e-06,
"loss": 1.0251,
"step": 241000
},
{
"epoch": 201.59,
"learning_rate": 3.777777777777778e-06,
"loss": 1.0251,
"step": 241500
},
{
"epoch": 202.0,
"learning_rate": 3.555555555555556e-06,
"loss": 1.0258,
"step": 242000
},
{
"epoch": 202.42,
"learning_rate": 3.3333333333333333e-06,
"loss": 1.0253,
"step": 242500
},
{
"epoch": 202.84,
"learning_rate": 3.111111111111111e-06,
"loss": 1.0243,
"step": 243000
},
{
"epoch": 203.26,
"learning_rate": 2.888888888888889e-06,
"loss": 1.0256,
"step": 243500
},
{
"epoch": 203.67,
"learning_rate": 2.666666666666667e-06,
"loss": 1.025,
"step": 244000
},
{
"epoch": 204.09,
"learning_rate": 2.4444444444444447e-06,
"loss": 1.0229,
"step": 244500
},
{
"epoch": 204.51,
"learning_rate": 2.2222222222222225e-06,
"loss": 1.0246,
"step": 245000
},
{
"epoch": 204.92,
"learning_rate": 2.0000000000000003e-06,
"loss": 1.0247,
"step": 245500
},
{
"epoch": 205.34,
"learning_rate": 1.777777777777778e-06,
"loss": 1.0237,
"step": 246000
},
{
"epoch": 205.76,
"learning_rate": 1.5555555555555556e-06,
"loss": 1.0237,
"step": 246500
},
{
"epoch": 206.18,
"learning_rate": 1.3333333333333334e-06,
"loss": 1.0217,
"step": 247000
},
{
"epoch": 206.59,
"learning_rate": 1.1111111111111112e-06,
"loss": 1.0232,
"step": 247500
},
{
"epoch": 207.01,
"learning_rate": 8.88888888888889e-07,
"loss": 1.0246,
"step": 248000
},
{
"epoch": 207.43,
"learning_rate": 6.666666666666667e-07,
"loss": 1.0225,
"step": 248500
},
{
"epoch": 207.85,
"learning_rate": 4.444444444444445e-07,
"loss": 1.0238,
"step": 249000
},
{
"epoch": 208.26,
"learning_rate": 2.2222222222222224e-07,
"loss": 1.023,
"step": 249500
},
{
"epoch": 208.68,
"learning_rate": 0.0,
"loss": 1.0216,
"step": 250000
},
{
"epoch": 208.68,
"eval_loss": 1.0167958736419678,
"eval_runtime": 125.3036,
"eval_samples_per_second": 543.791,
"eval_steps_per_second": 4.254,
"step": 250000
}
],
"max_steps": 250000,
"num_train_epochs": 209,
"total_flos": 4.212247412736e+18,
"trial_name": null,
"trial_params": null
}