roberta-base-wechsel-chinese / trainer_state.json
benjamin's picture
initial commit
c596805
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 94.3039502215518,
"global_step": 250000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.19,
"learning_rate": 2.0000000000000003e-06,
"loss": 8.2074,
"step": 500
},
{
"epoch": 0.38,
"learning_rate": 4.000000000000001e-06,
"loss": 7.333,
"step": 1000
},
{
"epoch": 0.57,
"learning_rate": 6e-06,
"loss": 6.7829,
"step": 1500
},
{
"epoch": 0.75,
"learning_rate": 8.000000000000001e-06,
"loss": 6.3812,
"step": 2000
},
{
"epoch": 0.94,
"learning_rate": 1e-05,
"loss": 6.1122,
"step": 2500
},
{
"epoch": 1.13,
"learning_rate": 1.2e-05,
"loss": 5.8855,
"step": 3000
},
{
"epoch": 1.32,
"learning_rate": 1.4000000000000001e-05,
"loss": 5.6587,
"step": 3500
},
{
"epoch": 1.51,
"learning_rate": 1.6000000000000003e-05,
"loss": 5.4578,
"step": 4000
},
{
"epoch": 1.7,
"learning_rate": 1.8e-05,
"loss": 5.2672,
"step": 4500
},
{
"epoch": 1.89,
"learning_rate": 2e-05,
"loss": 5.0883,
"step": 5000
},
{
"epoch": 2.07,
"learning_rate": 2.2000000000000003e-05,
"loss": 4.9455,
"step": 5500
},
{
"epoch": 2.26,
"learning_rate": 2.4e-05,
"loss": 4.7942,
"step": 6000
},
{
"epoch": 2.45,
"learning_rate": 2.6000000000000002e-05,
"loss": 4.6692,
"step": 6500
},
{
"epoch": 2.64,
"learning_rate": 2.8000000000000003e-05,
"loss": 4.5511,
"step": 7000
},
{
"epoch": 2.83,
"learning_rate": 3e-05,
"loss": 4.4469,
"step": 7500
},
{
"epoch": 3.02,
"learning_rate": 3.2000000000000005e-05,
"loss": 4.3566,
"step": 8000
},
{
"epoch": 3.21,
"learning_rate": 3.4000000000000007e-05,
"loss": 4.2571,
"step": 8500
},
{
"epoch": 3.39,
"learning_rate": 3.6e-05,
"loss": 4.1748,
"step": 9000
},
{
"epoch": 3.58,
"learning_rate": 3.8e-05,
"loss": 4.1022,
"step": 9500
},
{
"epoch": 3.77,
"learning_rate": 4e-05,
"loss": 4.027,
"step": 10000
},
{
"epoch": 3.96,
"learning_rate": 4.2e-05,
"loss": 3.9626,
"step": 10500
},
{
"epoch": 4.15,
"learning_rate": 4.4000000000000006e-05,
"loss": 3.9079,
"step": 11000
},
{
"epoch": 4.34,
"learning_rate": 4.600000000000001e-05,
"loss": 3.844,
"step": 11500
},
{
"epoch": 4.53,
"learning_rate": 4.8e-05,
"loss": 3.7907,
"step": 12000
},
{
"epoch": 4.71,
"learning_rate": 5e-05,
"loss": 3.7452,
"step": 12500
},
{
"epoch": 4.71,
"eval_loss": 3.5597941875457764,
"eval_runtime": 298.4056,
"eval_samples_per_second": 461.962,
"eval_steps_per_second": 3.609,
"step": 12500
},
{
"epoch": 4.9,
"learning_rate": 5.2000000000000004e-05,
"loss": 3.7008,
"step": 13000
},
{
"epoch": 5.09,
"learning_rate": 5.4000000000000005e-05,
"loss": 3.6604,
"step": 13500
},
{
"epoch": 5.28,
"learning_rate": 5.6000000000000006e-05,
"loss": 3.6173,
"step": 14000
},
{
"epoch": 5.47,
"learning_rate": 5.8e-05,
"loss": 3.5777,
"step": 14500
},
{
"epoch": 5.66,
"learning_rate": 6e-05,
"loss": 3.5396,
"step": 15000
},
{
"epoch": 5.85,
"learning_rate": 6.2e-05,
"loss": 3.5062,
"step": 15500
},
{
"epoch": 6.04,
"learning_rate": 6.400000000000001e-05,
"loss": 3.4766,
"step": 16000
},
{
"epoch": 6.22,
"learning_rate": 6.6e-05,
"loss": 3.4386,
"step": 16500
},
{
"epoch": 6.41,
"learning_rate": 6.800000000000001e-05,
"loss": 3.4131,
"step": 17000
},
{
"epoch": 6.6,
"learning_rate": 7e-05,
"loss": 3.388,
"step": 17500
},
{
"epoch": 6.79,
"learning_rate": 7.2e-05,
"loss": 3.3536,
"step": 18000
},
{
"epoch": 6.98,
"learning_rate": 7.4e-05,
"loss": 3.3314,
"step": 18500
},
{
"epoch": 7.17,
"learning_rate": 7.6e-05,
"loss": 3.3095,
"step": 19000
},
{
"epoch": 7.36,
"learning_rate": 7.800000000000001e-05,
"loss": 3.2802,
"step": 19500
},
{
"epoch": 7.54,
"learning_rate": 8e-05,
"loss": 3.2583,
"step": 20000
},
{
"epoch": 7.73,
"learning_rate": 8.2e-05,
"loss": 3.2374,
"step": 20500
},
{
"epoch": 7.92,
"learning_rate": 8.4e-05,
"loss": 3.2223,
"step": 21000
},
{
"epoch": 8.11,
"learning_rate": 8.6e-05,
"loss": 3.1986,
"step": 21500
},
{
"epoch": 8.3,
"learning_rate": 8.800000000000001e-05,
"loss": 3.1782,
"step": 22000
},
{
"epoch": 8.49,
"learning_rate": 9e-05,
"loss": 3.1585,
"step": 22500
},
{
"epoch": 8.68,
"learning_rate": 9.200000000000001e-05,
"loss": 3.1438,
"step": 23000
},
{
"epoch": 8.86,
"learning_rate": 9.4e-05,
"loss": 3.1313,
"step": 23500
},
{
"epoch": 9.05,
"learning_rate": 9.6e-05,
"loss": 3.1168,
"step": 24000
},
{
"epoch": 9.24,
"learning_rate": 9.8e-05,
"loss": 3.092,
"step": 24500
},
{
"epoch": 9.43,
"learning_rate": 0.0001,
"loss": 3.0788,
"step": 25000
},
{
"epoch": 9.43,
"eval_loss": 2.930635690689087,
"eval_runtime": 283.6019,
"eval_samples_per_second": 486.076,
"eval_steps_per_second": 3.798,
"step": 25000
},
{
"epoch": 9.62,
"learning_rate": 9.977777777777779e-05,
"loss": 3.0672,
"step": 25500
},
{
"epoch": 9.81,
"learning_rate": 9.955555555555556e-05,
"loss": 3.0534,
"step": 26000
},
{
"epoch": 10.0,
"learning_rate": 9.933333333333334e-05,
"loss": 3.0395,
"step": 26500
},
{
"epoch": 10.18,
"learning_rate": 9.911111111111112e-05,
"loss": 3.0276,
"step": 27000
},
{
"epoch": 10.37,
"learning_rate": 9.888888888888889e-05,
"loss": 3.0069,
"step": 27500
},
{
"epoch": 10.56,
"learning_rate": 9.866666666666668e-05,
"loss": 2.9995,
"step": 28000
},
{
"epoch": 10.75,
"learning_rate": 9.844444444444444e-05,
"loss": 2.9893,
"step": 28500
},
{
"epoch": 10.94,
"learning_rate": 9.822222222222223e-05,
"loss": 2.9776,
"step": 29000
},
{
"epoch": 11.13,
"learning_rate": 9.8e-05,
"loss": 2.9693,
"step": 29500
},
{
"epoch": 11.32,
"learning_rate": 9.777777777777778e-05,
"loss": 2.9528,
"step": 30000
},
{
"epoch": 11.5,
"learning_rate": 9.755555555555555e-05,
"loss": 2.9467,
"step": 30500
},
{
"epoch": 11.69,
"learning_rate": 9.733333333333335e-05,
"loss": 2.94,
"step": 31000
},
{
"epoch": 11.88,
"learning_rate": 9.711111111111111e-05,
"loss": 2.9276,
"step": 31500
},
{
"epoch": 12.07,
"learning_rate": 9.68888888888889e-05,
"loss": 2.9234,
"step": 32000
},
{
"epoch": 12.26,
"learning_rate": 9.666666666666667e-05,
"loss": 2.9092,
"step": 32500
},
{
"epoch": 12.45,
"learning_rate": 9.644444444444445e-05,
"loss": 2.9046,
"step": 33000
},
{
"epoch": 12.64,
"learning_rate": 9.622222222222222e-05,
"loss": 2.8948,
"step": 33500
},
{
"epoch": 12.83,
"learning_rate": 9.6e-05,
"loss": 2.8869,
"step": 34000
},
{
"epoch": 13.01,
"learning_rate": 9.577777777777777e-05,
"loss": 2.884,
"step": 34500
},
{
"epoch": 13.2,
"learning_rate": 9.555555555555557e-05,
"loss": 2.8717,
"step": 35000
},
{
"epoch": 13.39,
"learning_rate": 9.533333333333334e-05,
"loss": 2.8653,
"step": 35500
},
{
"epoch": 13.58,
"learning_rate": 9.511111111111112e-05,
"loss": 2.8565,
"step": 36000
},
{
"epoch": 13.77,
"learning_rate": 9.488888888888889e-05,
"loss": 2.8524,
"step": 36500
},
{
"epoch": 13.96,
"learning_rate": 9.466666666666667e-05,
"loss": 2.8449,
"step": 37000
},
{
"epoch": 14.15,
"learning_rate": 9.444444444444444e-05,
"loss": 2.8406,
"step": 37500
},
{
"epoch": 14.15,
"eval_loss": 2.706800937652588,
"eval_runtime": 283.8148,
"eval_samples_per_second": 485.711,
"eval_steps_per_second": 3.795,
"step": 37500
},
{
"epoch": 14.33,
"learning_rate": 9.422222222222223e-05,
"loss": 2.835,
"step": 38000
},
{
"epoch": 14.52,
"learning_rate": 9.4e-05,
"loss": 2.8281,
"step": 38500
},
{
"epoch": 14.71,
"learning_rate": 9.377777777777779e-05,
"loss": 2.8208,
"step": 39000
},
{
"epoch": 14.9,
"learning_rate": 9.355555555555556e-05,
"loss": 2.8225,
"step": 39500
},
{
"epoch": 15.09,
"learning_rate": 9.333333333333334e-05,
"loss": 2.8122,
"step": 40000
},
{
"epoch": 15.28,
"learning_rate": 9.311111111111111e-05,
"loss": 2.8051,
"step": 40500
},
{
"epoch": 15.47,
"learning_rate": 9.28888888888889e-05,
"loss": 2.8047,
"step": 41000
},
{
"epoch": 15.65,
"learning_rate": 9.266666666666666e-05,
"loss": 2.7949,
"step": 41500
},
{
"epoch": 15.84,
"learning_rate": 9.244444444444445e-05,
"loss": 2.7948,
"step": 42000
},
{
"epoch": 16.03,
"learning_rate": 9.222222222222223e-05,
"loss": 2.7895,
"step": 42500
},
{
"epoch": 16.22,
"learning_rate": 9.200000000000001e-05,
"loss": 2.784,
"step": 43000
},
{
"epoch": 16.41,
"learning_rate": 9.177777777777778e-05,
"loss": 2.7806,
"step": 43500
},
{
"epoch": 16.6,
"learning_rate": 9.155555555555557e-05,
"loss": 2.7774,
"step": 44000
},
{
"epoch": 16.79,
"learning_rate": 9.133333333333334e-05,
"loss": 2.769,
"step": 44500
},
{
"epoch": 16.97,
"learning_rate": 9.111111111111112e-05,
"loss": 2.7634,
"step": 45000
},
{
"epoch": 17.16,
"learning_rate": 9.088888888888889e-05,
"loss": 2.7645,
"step": 45500
},
{
"epoch": 17.35,
"learning_rate": 9.066666666666667e-05,
"loss": 2.7576,
"step": 46000
},
{
"epoch": 17.54,
"learning_rate": 9.044444444444445e-05,
"loss": 2.7533,
"step": 46500
},
{
"epoch": 17.73,
"learning_rate": 9.022222222222224e-05,
"loss": 2.7536,
"step": 47000
},
{
"epoch": 17.92,
"learning_rate": 9e-05,
"loss": 2.7447,
"step": 47500
},
{
"epoch": 18.11,
"learning_rate": 8.977777777777779e-05,
"loss": 2.7401,
"step": 48000
},
{
"epoch": 18.29,
"learning_rate": 8.955555555555556e-05,
"loss": 2.7365,
"step": 48500
},
{
"epoch": 18.48,
"learning_rate": 8.933333333333334e-05,
"loss": 2.7349,
"step": 49000
},
{
"epoch": 18.67,
"learning_rate": 8.911111111111111e-05,
"loss": 2.7308,
"step": 49500
},
{
"epoch": 18.86,
"learning_rate": 8.888888888888889e-05,
"loss": 2.7305,
"step": 50000
},
{
"epoch": 18.86,
"eval_loss": 2.608755111694336,
"eval_runtime": 283.7638,
"eval_samples_per_second": 485.798,
"eval_steps_per_second": 3.795,
"step": 50000
},
{
"epoch": 19.05,
"learning_rate": 8.866666666666668e-05,
"loss": 2.7327,
"step": 50500
},
{
"epoch": 19.24,
"learning_rate": 8.844444444444445e-05,
"loss": 2.722,
"step": 51000
},
{
"epoch": 19.43,
"learning_rate": 8.822222222222223e-05,
"loss": 2.7219,
"step": 51500
},
{
"epoch": 19.62,
"learning_rate": 8.800000000000001e-05,
"loss": 2.71,
"step": 52000
},
{
"epoch": 19.8,
"learning_rate": 8.777777777777778e-05,
"loss": 2.7107,
"step": 52500
},
{
"epoch": 19.99,
"learning_rate": 8.755555555555556e-05,
"loss": 2.7127,
"step": 53000
},
{
"epoch": 20.18,
"learning_rate": 8.733333333333333e-05,
"loss": 2.7044,
"step": 53500
},
{
"epoch": 20.37,
"learning_rate": 8.711111111111112e-05,
"loss": 2.7036,
"step": 54000
},
{
"epoch": 20.56,
"learning_rate": 8.68888888888889e-05,
"loss": 2.7002,
"step": 54500
},
{
"epoch": 20.75,
"learning_rate": 8.666666666666667e-05,
"loss": 2.6973,
"step": 55000
},
{
"epoch": 20.94,
"learning_rate": 8.644444444444445e-05,
"loss": 2.6931,
"step": 55500
},
{
"epoch": 21.12,
"learning_rate": 8.622222222222222e-05,
"loss": 2.6938,
"step": 56000
},
{
"epoch": 21.31,
"learning_rate": 8.6e-05,
"loss": 2.6837,
"step": 56500
},
{
"epoch": 21.5,
"learning_rate": 8.577777777777777e-05,
"loss": 2.6851,
"step": 57000
},
{
"epoch": 21.69,
"learning_rate": 8.555555555555556e-05,
"loss": 2.6848,
"step": 57500
},
{
"epoch": 21.88,
"learning_rate": 8.533333333333334e-05,
"loss": 2.6845,
"step": 58000
},
{
"epoch": 22.07,
"learning_rate": 8.511111111111112e-05,
"loss": 2.6839,
"step": 58500
},
{
"epoch": 22.26,
"learning_rate": 8.488888888888889e-05,
"loss": 2.6746,
"step": 59000
},
{
"epoch": 22.44,
"learning_rate": 8.466666666666667e-05,
"loss": 2.6736,
"step": 59500
},
{
"epoch": 22.63,
"learning_rate": 8.444444444444444e-05,
"loss": 2.6688,
"step": 60000
},
{
"epoch": 22.82,
"learning_rate": 8.422222222222223e-05,
"loss": 2.6699,
"step": 60500
},
{
"epoch": 23.01,
"learning_rate": 8.4e-05,
"loss": 2.673,
"step": 61000
},
{
"epoch": 23.2,
"learning_rate": 8.377777777777778e-05,
"loss": 2.6606,
"step": 61500
},
{
"epoch": 23.39,
"learning_rate": 8.355555555555556e-05,
"loss": 2.6595,
"step": 62000
},
{
"epoch": 23.58,
"learning_rate": 8.333333333333334e-05,
"loss": 2.6634,
"step": 62500
},
{
"epoch": 23.58,
"eval_loss": 2.541694164276123,
"eval_runtime": 283.9564,
"eval_samples_per_second": 485.469,
"eval_steps_per_second": 3.793,
"step": 62500
},
{
"epoch": 23.76,
"learning_rate": 8.311111111111111e-05,
"loss": 2.6599,
"step": 63000
},
{
"epoch": 23.95,
"learning_rate": 8.28888888888889e-05,
"loss": 2.6509,
"step": 63500
},
{
"epoch": 24.14,
"learning_rate": 8.266666666666667e-05,
"loss": 2.6573,
"step": 64000
},
{
"epoch": 24.33,
"learning_rate": 8.244444444444445e-05,
"loss": 2.6497,
"step": 64500
},
{
"epoch": 24.52,
"learning_rate": 8.222222222222222e-05,
"loss": 2.6492,
"step": 65000
},
{
"epoch": 24.71,
"learning_rate": 8.2e-05,
"loss": 2.646,
"step": 65500
},
{
"epoch": 24.9,
"learning_rate": 8.177777777777778e-05,
"loss": 2.6444,
"step": 66000
},
{
"epoch": 25.08,
"learning_rate": 8.155555555555557e-05,
"loss": 2.6465,
"step": 66500
},
{
"epoch": 25.27,
"learning_rate": 8.133333333333334e-05,
"loss": 2.6412,
"step": 67000
},
{
"epoch": 25.46,
"learning_rate": 8.111111111111112e-05,
"loss": 2.6364,
"step": 67500
},
{
"epoch": 25.65,
"learning_rate": 8.088888888888889e-05,
"loss": 2.6377,
"step": 68000
},
{
"epoch": 25.84,
"learning_rate": 8.066666666666667e-05,
"loss": 2.6331,
"step": 68500
},
{
"epoch": 26.03,
"learning_rate": 8.044444444444444e-05,
"loss": 2.6374,
"step": 69000
},
{
"epoch": 26.22,
"learning_rate": 8.022222222222222e-05,
"loss": 2.6269,
"step": 69500
},
{
"epoch": 26.41,
"learning_rate": 8e-05,
"loss": 2.625,
"step": 70000
},
{
"epoch": 26.59,
"learning_rate": 7.977777777777779e-05,
"loss": 2.6279,
"step": 70500
},
{
"epoch": 26.78,
"learning_rate": 7.955555555555556e-05,
"loss": 2.6278,
"step": 71000
},
{
"epoch": 26.97,
"learning_rate": 7.933333333333334e-05,
"loss": 2.6267,
"step": 71500
},
{
"epoch": 27.16,
"learning_rate": 7.911111111111111e-05,
"loss": 2.6273,
"step": 72000
},
{
"epoch": 27.35,
"learning_rate": 7.88888888888889e-05,
"loss": 2.6175,
"step": 72500
},
{
"epoch": 27.54,
"learning_rate": 7.866666666666666e-05,
"loss": 2.6149,
"step": 73000
},
{
"epoch": 27.73,
"learning_rate": 7.844444444444446e-05,
"loss": 2.6183,
"step": 73500
},
{
"epoch": 27.91,
"learning_rate": 7.822222222222223e-05,
"loss": 2.6107,
"step": 74000
},
{
"epoch": 28.1,
"learning_rate": 7.800000000000001e-05,
"loss": 2.6181,
"step": 74500
},
{
"epoch": 28.29,
"learning_rate": 7.777777777777778e-05,
"loss": 2.6134,
"step": 75000
},
{
"epoch": 28.29,
"eval_loss": 2.4989469051361084,
"eval_runtime": 283.5232,
"eval_samples_per_second": 486.211,
"eval_steps_per_second": 3.799,
"step": 75000
},
{
"epoch": 28.48,
"learning_rate": 7.755555555555556e-05,
"loss": 2.6136,
"step": 75500
},
{
"epoch": 28.67,
"learning_rate": 7.733333333333333e-05,
"loss": 2.6061,
"step": 76000
},
{
"epoch": 28.86,
"learning_rate": 7.711111111111112e-05,
"loss": 2.6038,
"step": 76500
},
{
"epoch": 29.05,
"learning_rate": 7.688888888888889e-05,
"loss": 2.6056,
"step": 77000
},
{
"epoch": 29.23,
"learning_rate": 7.666666666666667e-05,
"loss": 2.6044,
"step": 77500
},
{
"epoch": 29.42,
"learning_rate": 7.644444444444445e-05,
"loss": 2.6012,
"step": 78000
},
{
"epoch": 29.61,
"learning_rate": 7.622222222222223e-05,
"loss": 2.5976,
"step": 78500
},
{
"epoch": 29.8,
"learning_rate": 7.6e-05,
"loss": 2.5993,
"step": 79000
},
{
"epoch": 29.99,
"learning_rate": 7.577777777777779e-05,
"loss": 2.5964,
"step": 79500
},
{
"epoch": 30.18,
"learning_rate": 7.555555555555556e-05,
"loss": 2.5953,
"step": 80000
},
{
"epoch": 30.37,
"learning_rate": 7.533333333333334e-05,
"loss": 2.5916,
"step": 80500
},
{
"epoch": 30.55,
"learning_rate": 7.511111111111111e-05,
"loss": 2.6004,
"step": 81000
},
{
"epoch": 30.74,
"learning_rate": 7.488888888888889e-05,
"loss": 2.5902,
"step": 81500
},
{
"epoch": 30.93,
"learning_rate": 7.466666666666667e-05,
"loss": 2.5936,
"step": 82000
},
{
"epoch": 31.12,
"learning_rate": 7.444444444444444e-05,
"loss": 2.5939,
"step": 82500
},
{
"epoch": 31.31,
"learning_rate": 7.422222222222223e-05,
"loss": 2.5893,
"step": 83000
},
{
"epoch": 31.5,
"learning_rate": 7.4e-05,
"loss": 2.5834,
"step": 83500
},
{
"epoch": 31.69,
"learning_rate": 7.377777777777778e-05,
"loss": 2.5853,
"step": 84000
},
{
"epoch": 31.87,
"learning_rate": 7.355555555555556e-05,
"loss": 2.5825,
"step": 84500
},
{
"epoch": 32.06,
"learning_rate": 7.333333333333333e-05,
"loss": 2.5865,
"step": 85000
},
{
"epoch": 32.25,
"learning_rate": 7.311111111111111e-05,
"loss": 2.5756,
"step": 85500
},
{
"epoch": 32.44,
"learning_rate": 7.28888888888889e-05,
"loss": 2.5774,
"step": 86000
},
{
"epoch": 32.63,
"learning_rate": 7.266666666666667e-05,
"loss": 2.5792,
"step": 86500
},
{
"epoch": 32.82,
"learning_rate": 7.244444444444445e-05,
"loss": 2.5753,
"step": 87000
},
{
"epoch": 33.01,
"learning_rate": 7.222222222222222e-05,
"loss": 2.5812,
"step": 87500
},
{
"epoch": 33.01,
"eval_loss": 2.469001531600952,
"eval_runtime": 283.5082,
"eval_samples_per_second": 486.236,
"eval_steps_per_second": 3.799,
"step": 87500
},
{
"epoch": 33.19,
"learning_rate": 7.2e-05,
"loss": 2.571,
"step": 88000
},
{
"epoch": 33.38,
"learning_rate": 7.177777777777777e-05,
"loss": 2.5736,
"step": 88500
},
{
"epoch": 33.57,
"learning_rate": 7.155555555555555e-05,
"loss": 2.5712,
"step": 89000
},
{
"epoch": 33.76,
"learning_rate": 7.133333333333334e-05,
"loss": 2.5728,
"step": 89500
},
{
"epoch": 33.95,
"learning_rate": 7.111111111111112e-05,
"loss": 2.5676,
"step": 90000
},
{
"epoch": 34.14,
"learning_rate": 7.088888888888889e-05,
"loss": 2.5682,
"step": 90500
},
{
"epoch": 34.33,
"learning_rate": 7.066666666666667e-05,
"loss": 2.5662,
"step": 91000
},
{
"epoch": 34.52,
"learning_rate": 7.044444444444444e-05,
"loss": 2.5671,
"step": 91500
},
{
"epoch": 34.7,
"learning_rate": 7.022222222222222e-05,
"loss": 2.5667,
"step": 92000
},
{
"epoch": 34.89,
"learning_rate": 7e-05,
"loss": 2.5619,
"step": 92500
},
{
"epoch": 35.08,
"learning_rate": 6.977777777777779e-05,
"loss": 2.5639,
"step": 93000
},
{
"epoch": 35.27,
"learning_rate": 6.955555555555556e-05,
"loss": 2.5581,
"step": 93500
},
{
"epoch": 35.46,
"learning_rate": 6.933333333333334e-05,
"loss": 2.5586,
"step": 94000
},
{
"epoch": 35.65,
"learning_rate": 6.911111111111111e-05,
"loss": 2.5612,
"step": 94500
},
{
"epoch": 35.84,
"learning_rate": 6.88888888888889e-05,
"loss": 2.5552,
"step": 95000
},
{
"epoch": 36.02,
"learning_rate": 6.866666666666666e-05,
"loss": 2.5598,
"step": 95500
},
{
"epoch": 36.21,
"learning_rate": 6.844444444444445e-05,
"loss": 2.5479,
"step": 96000
},
{
"epoch": 36.4,
"learning_rate": 6.822222222222222e-05,
"loss": 2.5545,
"step": 96500
},
{
"epoch": 36.59,
"learning_rate": 6.800000000000001e-05,
"loss": 2.5506,
"step": 97000
},
{
"epoch": 36.78,
"learning_rate": 6.777777777777778e-05,
"loss": 2.5526,
"step": 97500
},
{
"epoch": 36.97,
"learning_rate": 6.755555555555557e-05,
"loss": 2.552,
"step": 98000
},
{
"epoch": 37.16,
"learning_rate": 6.733333333333333e-05,
"loss": 2.5493,
"step": 98500
},
{
"epoch": 37.34,
"learning_rate": 6.711111111111112e-05,
"loss": 2.5474,
"step": 99000
},
{
"epoch": 37.53,
"learning_rate": 6.688888888888889e-05,
"loss": 2.5475,
"step": 99500
},
{
"epoch": 37.72,
"learning_rate": 6.666666666666667e-05,
"loss": 2.5447,
"step": 100000
},
{
"epoch": 37.72,
"eval_loss": 2.4395639896392822,
"eval_runtime": 283.6209,
"eval_samples_per_second": 486.043,
"eval_steps_per_second": 3.797,
"step": 100000
},
{
"epoch": 37.91,
"learning_rate": 6.644444444444444e-05,
"loss": 2.5444,
"step": 100500
},
{
"epoch": 38.1,
"learning_rate": 6.622222222222224e-05,
"loss": 2.5467,
"step": 101000
},
{
"epoch": 38.29,
"learning_rate": 6.6e-05,
"loss": 2.537,
"step": 101500
},
{
"epoch": 38.48,
"learning_rate": 6.577777777777779e-05,
"loss": 2.5344,
"step": 102000
},
{
"epoch": 38.66,
"learning_rate": 6.555555555555556e-05,
"loss": 2.5431,
"step": 102500
},
{
"epoch": 38.85,
"learning_rate": 6.533333333333334e-05,
"loss": 2.5435,
"step": 103000
},
{
"epoch": 39.04,
"learning_rate": 6.511111111111111e-05,
"loss": 2.5434,
"step": 103500
},
{
"epoch": 39.23,
"learning_rate": 6.488888888888889e-05,
"loss": 2.5297,
"step": 104000
},
{
"epoch": 39.42,
"learning_rate": 6.466666666666666e-05,
"loss": 2.5408,
"step": 104500
},
{
"epoch": 39.61,
"learning_rate": 6.444444444444446e-05,
"loss": 2.5298,
"step": 105000
},
{
"epoch": 39.8,
"learning_rate": 6.422222222222223e-05,
"loss": 2.5358,
"step": 105500
},
{
"epoch": 39.98,
"learning_rate": 6.400000000000001e-05,
"loss": 2.5379,
"step": 106000
},
{
"epoch": 40.17,
"learning_rate": 6.377777777777778e-05,
"loss": 2.5357,
"step": 106500
},
{
"epoch": 40.36,
"learning_rate": 6.355555555555556e-05,
"loss": 2.5268,
"step": 107000
},
{
"epoch": 40.55,
"learning_rate": 6.333333333333333e-05,
"loss": 2.5263,
"step": 107500
},
{
"epoch": 40.74,
"learning_rate": 6.311111111111112e-05,
"loss": 2.5329,
"step": 108000
},
{
"epoch": 40.93,
"learning_rate": 6.28888888888889e-05,
"loss": 2.533,
"step": 108500
},
{
"epoch": 41.12,
"learning_rate": 6.266666666666667e-05,
"loss": 2.5282,
"step": 109000
},
{
"epoch": 41.31,
"learning_rate": 6.244444444444445e-05,
"loss": 2.5227,
"step": 109500
},
{
"epoch": 41.49,
"learning_rate": 6.222222222222222e-05,
"loss": 2.5279,
"step": 110000
},
{
"epoch": 41.68,
"learning_rate": 6.2e-05,
"loss": 2.5272,
"step": 110500
},
{
"epoch": 41.87,
"learning_rate": 6.177777777777779e-05,
"loss": 2.522,
"step": 111000
},
{
"epoch": 42.06,
"learning_rate": 6.155555555555555e-05,
"loss": 2.5253,
"step": 111500
},
{
"epoch": 42.25,
"learning_rate": 6.133333333333334e-05,
"loss": 2.5215,
"step": 112000
},
{
"epoch": 42.44,
"learning_rate": 6.111111111111112e-05,
"loss": 2.5205,
"step": 112500
},
{
"epoch": 42.44,
"eval_loss": 2.419991970062256,
"eval_runtime": 283.6709,
"eval_samples_per_second": 485.957,
"eval_steps_per_second": 3.797,
"step": 112500
},
{
"epoch": 42.63,
"learning_rate": 6.08888888888889e-05,
"loss": 2.5194,
"step": 113000
},
{
"epoch": 42.81,
"learning_rate": 6.066666666666667e-05,
"loss": 2.5179,
"step": 113500
},
{
"epoch": 43.0,
"learning_rate": 6.044444444444445e-05,
"loss": 2.526,
"step": 114000
},
{
"epoch": 43.19,
"learning_rate": 6.0222222222222225e-05,
"loss": 2.5172,
"step": 114500
},
{
"epoch": 43.38,
"learning_rate": 6e-05,
"loss": 2.5114,
"step": 115000
},
{
"epoch": 43.57,
"learning_rate": 5.977777777777778e-05,
"loss": 2.5179,
"step": 115500
},
{
"epoch": 43.76,
"learning_rate": 5.9555555555555554e-05,
"loss": 2.5189,
"step": 116000
},
{
"epoch": 43.95,
"learning_rate": 5.9333333333333343e-05,
"loss": 2.5202,
"step": 116500
},
{
"epoch": 44.13,
"learning_rate": 5.911111111111112e-05,
"loss": 2.517,
"step": 117000
},
{
"epoch": 44.32,
"learning_rate": 5.8888888888888896e-05,
"loss": 2.5151,
"step": 117500
},
{
"epoch": 44.51,
"learning_rate": 5.866666666666667e-05,
"loss": 2.5101,
"step": 118000
},
{
"epoch": 44.7,
"learning_rate": 5.844444444444445e-05,
"loss": 2.511,
"step": 118500
},
{
"epoch": 44.89,
"learning_rate": 5.8222222222222224e-05,
"loss": 2.5105,
"step": 119000
},
{
"epoch": 45.08,
"learning_rate": 5.8e-05,
"loss": 2.5107,
"step": 119500
},
{
"epoch": 45.27,
"learning_rate": 5.7777777777777776e-05,
"loss": 2.5089,
"step": 120000
},
{
"epoch": 45.45,
"learning_rate": 5.755555555555556e-05,
"loss": 2.5051,
"step": 120500
},
{
"epoch": 45.64,
"learning_rate": 5.7333333333333336e-05,
"loss": 2.5083,
"step": 121000
},
{
"epoch": 45.83,
"learning_rate": 5.711111111111112e-05,
"loss": 2.5072,
"step": 121500
},
{
"epoch": 46.02,
"learning_rate": 5.6888888888888895e-05,
"loss": 2.5107,
"step": 122000
},
{
"epoch": 46.21,
"learning_rate": 5.666666666666667e-05,
"loss": 2.5036,
"step": 122500
},
{
"epoch": 46.4,
"learning_rate": 5.644444444444445e-05,
"loss": 2.5051,
"step": 123000
},
{
"epoch": 46.59,
"learning_rate": 5.622222222222222e-05,
"loss": 2.5039,
"step": 123500
},
{
"epoch": 46.77,
"learning_rate": 5.6000000000000006e-05,
"loss": 2.4972,
"step": 124000
},
{
"epoch": 46.96,
"learning_rate": 5.577777777777778e-05,
"loss": 2.5073,
"step": 124500
},
{
"epoch": 47.15,
"learning_rate": 5.555555555555556e-05,
"loss": 2.5052,
"step": 125000
},
{
"epoch": 47.15,
"eval_loss": 2.4018545150756836,
"eval_runtime": 283.8972,
"eval_samples_per_second": 485.57,
"eval_steps_per_second": 3.794,
"step": 125000
},
{
"epoch": 47.34,
"learning_rate": 5.5333333333333334e-05,
"loss": 2.5015,
"step": 125500
},
{
"epoch": 47.53,
"learning_rate": 5.511111111111111e-05,
"loss": 2.4983,
"step": 126000
},
{
"epoch": 47.72,
"learning_rate": 5.488888888888889e-05,
"loss": 2.5022,
"step": 126500
},
{
"epoch": 47.91,
"learning_rate": 5.466666666666666e-05,
"loss": 2.4932,
"step": 127000
},
{
"epoch": 48.1,
"learning_rate": 5.4444444444444446e-05,
"loss": 2.5007,
"step": 127500
},
{
"epoch": 48.28,
"learning_rate": 5.422222222222223e-05,
"loss": 2.4944,
"step": 128000
},
{
"epoch": 48.47,
"learning_rate": 5.4000000000000005e-05,
"loss": 2.4964,
"step": 128500
},
{
"epoch": 48.66,
"learning_rate": 5.377777777777778e-05,
"loss": 2.4976,
"step": 129000
},
{
"epoch": 48.85,
"learning_rate": 5.355555555555556e-05,
"loss": 2.4948,
"step": 129500
},
{
"epoch": 49.04,
"learning_rate": 5.333333333333333e-05,
"loss": 2.5018,
"step": 130000
},
{
"epoch": 49.23,
"learning_rate": 5.311111111111111e-05,
"loss": 2.4919,
"step": 130500
},
{
"epoch": 49.42,
"learning_rate": 5.2888888888888885e-05,
"loss": 2.4925,
"step": 131000
},
{
"epoch": 49.6,
"learning_rate": 5.266666666666666e-05,
"loss": 2.491,
"step": 131500
},
{
"epoch": 49.79,
"learning_rate": 5.244444444444445e-05,
"loss": 2.4932,
"step": 132000
},
{
"epoch": 49.98,
"learning_rate": 5.222222222222223e-05,
"loss": 2.4892,
"step": 132500
},
{
"epoch": 50.17,
"learning_rate": 5.2000000000000004e-05,
"loss": 2.4877,
"step": 133000
},
{
"epoch": 50.36,
"learning_rate": 5.177777777777778e-05,
"loss": 2.4899,
"step": 133500
},
{
"epoch": 50.55,
"learning_rate": 5.1555555555555556e-05,
"loss": 2.49,
"step": 134000
},
{
"epoch": 50.74,
"learning_rate": 5.133333333333333e-05,
"loss": 2.4843,
"step": 134500
},
{
"epoch": 50.92,
"learning_rate": 5.111111111111111e-05,
"loss": 2.4901,
"step": 135000
},
{
"epoch": 51.11,
"learning_rate": 5.0888888888888884e-05,
"loss": 2.4919,
"step": 135500
},
{
"epoch": 51.3,
"learning_rate": 5.0666666666666674e-05,
"loss": 2.4862,
"step": 136000
},
{
"epoch": 51.49,
"learning_rate": 5.044444444444445e-05,
"loss": 2.4813,
"step": 136500
},
{
"epoch": 51.68,
"learning_rate": 5.0222222222222226e-05,
"loss": 2.4818,
"step": 137000
},
{
"epoch": 51.87,
"learning_rate": 5e-05,
"loss": 2.4834,
"step": 137500
},
{
"epoch": 51.87,
"eval_loss": 2.3841142654418945,
"eval_runtime": 283.7333,
"eval_samples_per_second": 485.851,
"eval_steps_per_second": 3.796,
"step": 137500
},
{
"epoch": 52.06,
"learning_rate": 4.977777777777778e-05,
"loss": 2.4823,
"step": 138000
},
{
"epoch": 52.24,
"learning_rate": 4.955555555555556e-05,
"loss": 2.4773,
"step": 138500
},
{
"epoch": 52.43,
"learning_rate": 4.933333333333334e-05,
"loss": 2.4818,
"step": 139000
},
{
"epoch": 52.62,
"learning_rate": 4.9111111111111114e-05,
"loss": 2.4807,
"step": 139500
},
{
"epoch": 52.81,
"learning_rate": 4.888888888888889e-05,
"loss": 2.4801,
"step": 140000
},
{
"epoch": 53.0,
"learning_rate": 4.866666666666667e-05,
"loss": 2.4808,
"step": 140500
},
{
"epoch": 53.19,
"learning_rate": 4.844444444444445e-05,
"loss": 2.4788,
"step": 141000
},
{
"epoch": 53.38,
"learning_rate": 4.8222222222222225e-05,
"loss": 2.478,
"step": 141500
},
{
"epoch": 53.56,
"learning_rate": 4.8e-05,
"loss": 2.4758,
"step": 142000
},
{
"epoch": 53.75,
"learning_rate": 4.7777777777777784e-05,
"loss": 2.4792,
"step": 142500
},
{
"epoch": 53.94,
"learning_rate": 4.755555555555556e-05,
"loss": 2.4781,
"step": 143000
},
{
"epoch": 54.13,
"learning_rate": 4.7333333333333336e-05,
"loss": 2.4773,
"step": 143500
},
{
"epoch": 54.32,
"learning_rate": 4.711111111111111e-05,
"loss": 2.4725,
"step": 144000
},
{
"epoch": 54.51,
"learning_rate": 4.6888888888888895e-05,
"loss": 2.4748,
"step": 144500
},
{
"epoch": 54.7,
"learning_rate": 4.666666666666667e-05,
"loss": 2.4751,
"step": 145000
},
{
"epoch": 54.88,
"learning_rate": 4.644444444444445e-05,
"loss": 2.47,
"step": 145500
},
{
"epoch": 55.07,
"learning_rate": 4.6222222222222224e-05,
"loss": 2.4796,
"step": 146000
},
{
"epoch": 55.26,
"learning_rate": 4.600000000000001e-05,
"loss": 2.4717,
"step": 146500
},
{
"epoch": 55.45,
"learning_rate": 4.577777777777778e-05,
"loss": 2.4759,
"step": 147000
},
{
"epoch": 55.64,
"learning_rate": 4.555555555555556e-05,
"loss": 2.4696,
"step": 147500
},
{
"epoch": 55.83,
"learning_rate": 4.5333333333333335e-05,
"loss": 2.4707,
"step": 148000
},
{
"epoch": 56.02,
"learning_rate": 4.511111111111112e-05,
"loss": 2.4664,
"step": 148500
},
{
"epoch": 56.21,
"learning_rate": 4.4888888888888894e-05,
"loss": 2.4663,
"step": 149000
},
{
"epoch": 56.39,
"learning_rate": 4.466666666666667e-05,
"loss": 2.4679,
"step": 149500
},
{
"epoch": 56.58,
"learning_rate": 4.4444444444444447e-05,
"loss": 2.4694,
"step": 150000
},
{
"epoch": 56.58,
"eval_loss": 2.371152877807617,
"eval_runtime": 283.5132,
"eval_samples_per_second": 486.228,
"eval_steps_per_second": 3.799,
"step": 150000
},
{
"epoch": 56.77,
"learning_rate": 4.422222222222222e-05,
"loss": 2.4619,
"step": 150500
},
{
"epoch": 56.96,
"learning_rate": 4.4000000000000006e-05,
"loss": 2.4686,
"step": 151000
},
{
"epoch": 57.15,
"learning_rate": 4.377777777777778e-05,
"loss": 2.4715,
"step": 151500
},
{
"epoch": 57.34,
"learning_rate": 4.355555555555556e-05,
"loss": 2.4619,
"step": 152000
},
{
"epoch": 57.53,
"learning_rate": 4.3333333333333334e-05,
"loss": 2.463,
"step": 152500
},
{
"epoch": 57.71,
"learning_rate": 4.311111111111111e-05,
"loss": 2.4609,
"step": 153000
},
{
"epoch": 57.9,
"learning_rate": 4.2888888888888886e-05,
"loss": 2.4646,
"step": 153500
},
{
"epoch": 58.09,
"learning_rate": 4.266666666666667e-05,
"loss": 2.4658,
"step": 154000
},
{
"epoch": 58.28,
"learning_rate": 4.2444444444444445e-05,
"loss": 2.4614,
"step": 154500
},
{
"epoch": 58.47,
"learning_rate": 4.222222222222222e-05,
"loss": 2.4646,
"step": 155000
},
{
"epoch": 58.66,
"learning_rate": 4.2e-05,
"loss": 2.4629,
"step": 155500
},
{
"epoch": 58.85,
"learning_rate": 4.177777777777778e-05,
"loss": 2.4587,
"step": 156000
},
{
"epoch": 59.03,
"learning_rate": 4.155555555555556e-05,
"loss": 2.466,
"step": 156500
},
{
"epoch": 59.22,
"learning_rate": 4.133333333333333e-05,
"loss": 2.452,
"step": 157000
},
{
"epoch": 59.41,
"learning_rate": 4.111111111111111e-05,
"loss": 2.4608,
"step": 157500
},
{
"epoch": 59.6,
"learning_rate": 4.088888888888889e-05,
"loss": 2.4629,
"step": 158000
},
{
"epoch": 59.79,
"learning_rate": 4.066666666666667e-05,
"loss": 2.4605,
"step": 158500
},
{
"epoch": 59.98,
"learning_rate": 4.0444444444444444e-05,
"loss": 2.4576,
"step": 159000
},
{
"epoch": 60.17,
"learning_rate": 4.022222222222222e-05,
"loss": 2.4599,
"step": 159500
},
{
"epoch": 60.35,
"learning_rate": 4e-05,
"loss": 2.4567,
"step": 160000
},
{
"epoch": 60.54,
"learning_rate": 3.977777777777778e-05,
"loss": 2.4604,
"step": 160500
},
{
"epoch": 60.73,
"learning_rate": 3.9555555555555556e-05,
"loss": 2.4565,
"step": 161000
},
{
"epoch": 60.92,
"learning_rate": 3.933333333333333e-05,
"loss": 2.4559,
"step": 161500
},
{
"epoch": 61.11,
"learning_rate": 3.9111111111111115e-05,
"loss": 2.4587,
"step": 162000
},
{
"epoch": 61.3,
"learning_rate": 3.888888888888889e-05,
"loss": 2.4509,
"step": 162500
},
{
"epoch": 61.3,
"eval_loss": 2.3613131046295166,
"eval_runtime": 283.6567,
"eval_samples_per_second": 485.982,
"eval_steps_per_second": 3.797,
"step": 162500
},
{
"epoch": 61.49,
"learning_rate": 3.866666666666667e-05,
"loss": 2.4554,
"step": 163000
},
{
"epoch": 61.67,
"learning_rate": 3.844444444444444e-05,
"loss": 2.4562,
"step": 163500
},
{
"epoch": 61.86,
"learning_rate": 3.8222222222222226e-05,
"loss": 2.4511,
"step": 164000
},
{
"epoch": 62.05,
"learning_rate": 3.8e-05,
"loss": 2.4578,
"step": 164500
},
{
"epoch": 62.24,
"learning_rate": 3.777777777777778e-05,
"loss": 2.4513,
"step": 165000
},
{
"epoch": 62.43,
"learning_rate": 3.7555555555555554e-05,
"loss": 2.4509,
"step": 165500
},
{
"epoch": 62.62,
"learning_rate": 3.733333333333334e-05,
"loss": 2.4498,
"step": 166000
},
{
"epoch": 62.81,
"learning_rate": 3.7111111111111113e-05,
"loss": 2.4499,
"step": 166500
},
{
"epoch": 62.99,
"learning_rate": 3.688888888888889e-05,
"loss": 2.4543,
"step": 167000
},
{
"epoch": 63.18,
"learning_rate": 3.6666666666666666e-05,
"loss": 2.4498,
"step": 167500
},
{
"epoch": 63.37,
"learning_rate": 3.644444444444445e-05,
"loss": 2.4463,
"step": 168000
},
{
"epoch": 63.56,
"learning_rate": 3.6222222222222225e-05,
"loss": 2.4491,
"step": 168500
},
{
"epoch": 63.75,
"learning_rate": 3.6e-05,
"loss": 2.4532,
"step": 169000
},
{
"epoch": 63.94,
"learning_rate": 3.577777777777778e-05,
"loss": 2.4441,
"step": 169500
},
{
"epoch": 64.13,
"learning_rate": 3.555555555555556e-05,
"loss": 2.45,
"step": 170000
},
{
"epoch": 64.32,
"learning_rate": 3.5333333333333336e-05,
"loss": 2.4453,
"step": 170500
},
{
"epoch": 64.5,
"learning_rate": 3.511111111111111e-05,
"loss": 2.4463,
"step": 171000
},
{
"epoch": 64.69,
"learning_rate": 3.4888888888888895e-05,
"loss": 2.4457,
"step": 171500
},
{
"epoch": 64.88,
"learning_rate": 3.466666666666667e-05,
"loss": 2.4477,
"step": 172000
},
{
"epoch": 65.07,
"learning_rate": 3.444444444444445e-05,
"loss": 2.4467,
"step": 172500
},
{
"epoch": 65.26,
"learning_rate": 3.4222222222222224e-05,
"loss": 2.447,
"step": 173000
},
{
"epoch": 65.45,
"learning_rate": 3.4000000000000007e-05,
"loss": 2.443,
"step": 173500
},
{
"epoch": 65.64,
"learning_rate": 3.377777777777778e-05,
"loss": 2.443,
"step": 174000
},
{
"epoch": 65.82,
"learning_rate": 3.355555555555556e-05,
"loss": 2.4441,
"step": 174500
},
{
"epoch": 66.01,
"learning_rate": 3.3333333333333335e-05,
"loss": 2.4469,
"step": 175000
},
{
"epoch": 66.01,
"eval_loss": 2.3445606231689453,
"eval_runtime": 283.6225,
"eval_samples_per_second": 486.04,
"eval_steps_per_second": 3.797,
"step": 175000
},
{
"epoch": 66.2,
"learning_rate": 3.311111111111112e-05,
"loss": 2.4367,
"step": 175500
},
{
"epoch": 66.39,
"learning_rate": 3.2888888888888894e-05,
"loss": 2.4414,
"step": 176000
},
{
"epoch": 66.58,
"learning_rate": 3.266666666666667e-05,
"loss": 2.4424,
"step": 176500
},
{
"epoch": 66.77,
"learning_rate": 3.2444444444444446e-05,
"loss": 2.4405,
"step": 177000
},
{
"epoch": 66.96,
"learning_rate": 3.222222222222223e-05,
"loss": 2.4415,
"step": 177500
},
{
"epoch": 67.14,
"learning_rate": 3.2000000000000005e-05,
"loss": 2.4398,
"step": 178000
},
{
"epoch": 67.33,
"learning_rate": 3.177777777777778e-05,
"loss": 2.4406,
"step": 178500
},
{
"epoch": 67.52,
"learning_rate": 3.155555555555556e-05,
"loss": 2.4375,
"step": 179000
},
{
"epoch": 67.71,
"learning_rate": 3.1333333333333334e-05,
"loss": 2.4365,
"step": 179500
},
{
"epoch": 67.9,
"learning_rate": 3.111111111111111e-05,
"loss": 2.4415,
"step": 180000
},
{
"epoch": 68.09,
"learning_rate": 3.088888888888889e-05,
"loss": 2.4391,
"step": 180500
},
{
"epoch": 68.28,
"learning_rate": 3.066666666666667e-05,
"loss": 2.437,
"step": 181000
},
{
"epoch": 68.46,
"learning_rate": 3.044444444444445e-05,
"loss": 2.4357,
"step": 181500
},
{
"epoch": 68.65,
"learning_rate": 3.0222222222222225e-05,
"loss": 2.4375,
"step": 182000
},
{
"epoch": 68.84,
"learning_rate": 3e-05,
"loss": 2.4356,
"step": 182500
},
{
"epoch": 69.03,
"learning_rate": 2.9777777777777777e-05,
"loss": 2.4364,
"step": 183000
},
{
"epoch": 69.22,
"learning_rate": 2.955555555555556e-05,
"loss": 2.4335,
"step": 183500
},
{
"epoch": 69.41,
"learning_rate": 2.9333333333333336e-05,
"loss": 2.4325,
"step": 184000
},
{
"epoch": 69.6,
"learning_rate": 2.9111111111111112e-05,
"loss": 2.434,
"step": 184500
},
{
"epoch": 69.78,
"learning_rate": 2.8888888888888888e-05,
"loss": 2.4397,
"step": 185000
},
{
"epoch": 69.97,
"learning_rate": 2.8666666666666668e-05,
"loss": 2.4375,
"step": 185500
},
{
"epoch": 70.16,
"learning_rate": 2.8444444444444447e-05,
"loss": 2.4338,
"step": 186000
},
{
"epoch": 70.35,
"learning_rate": 2.8222222222222223e-05,
"loss": 2.4317,
"step": 186500
},
{
"epoch": 70.54,
"learning_rate": 2.8000000000000003e-05,
"loss": 2.4321,
"step": 187000
},
{
"epoch": 70.73,
"learning_rate": 2.777777777777778e-05,
"loss": 2.4296,
"step": 187500
},
{
"epoch": 70.73,
"eval_loss": 2.3378496170043945,
"eval_runtime": 283.6193,
"eval_samples_per_second": 486.046,
"eval_steps_per_second": 3.797,
"step": 187500
},
{
"epoch": 70.92,
"learning_rate": 2.7555555555555555e-05,
"loss": 2.4349,
"step": 188000
},
{
"epoch": 71.11,
"learning_rate": 2.733333333333333e-05,
"loss": 2.4326,
"step": 188500
},
{
"epoch": 71.29,
"learning_rate": 2.7111111111111114e-05,
"loss": 2.4284,
"step": 189000
},
{
"epoch": 71.48,
"learning_rate": 2.688888888888889e-05,
"loss": 2.4299,
"step": 189500
},
{
"epoch": 71.67,
"learning_rate": 2.6666666666666667e-05,
"loss": 2.4317,
"step": 190000
},
{
"epoch": 71.86,
"learning_rate": 2.6444444444444443e-05,
"loss": 2.4288,
"step": 190500
},
{
"epoch": 72.05,
"learning_rate": 2.6222222222222226e-05,
"loss": 2.4322,
"step": 191000
},
{
"epoch": 72.24,
"learning_rate": 2.6000000000000002e-05,
"loss": 2.4317,
"step": 191500
},
{
"epoch": 72.43,
"learning_rate": 2.5777777777777778e-05,
"loss": 2.4262,
"step": 192000
},
{
"epoch": 72.61,
"learning_rate": 2.5555555555555554e-05,
"loss": 2.4239,
"step": 192500
},
{
"epoch": 72.8,
"learning_rate": 2.5333333333333337e-05,
"loss": 2.4279,
"step": 193000
},
{
"epoch": 72.99,
"learning_rate": 2.5111111111111113e-05,
"loss": 2.4271,
"step": 193500
},
{
"epoch": 73.18,
"learning_rate": 2.488888888888889e-05,
"loss": 2.4349,
"step": 194000
},
{
"epoch": 73.37,
"learning_rate": 2.466666666666667e-05,
"loss": 2.4285,
"step": 194500
},
{
"epoch": 73.56,
"learning_rate": 2.4444444444444445e-05,
"loss": 2.4217,
"step": 195000
},
{
"epoch": 73.75,
"learning_rate": 2.4222222222222224e-05,
"loss": 2.4215,
"step": 195500
},
{
"epoch": 73.93,
"learning_rate": 2.4e-05,
"loss": 2.4267,
"step": 196000
},
{
"epoch": 74.12,
"learning_rate": 2.377777777777778e-05,
"loss": 2.4281,
"step": 196500
},
{
"epoch": 74.31,
"learning_rate": 2.3555555555555556e-05,
"loss": 2.4238,
"step": 197000
},
{
"epoch": 74.5,
"learning_rate": 2.3333333333333336e-05,
"loss": 2.4243,
"step": 197500
},
{
"epoch": 74.69,
"learning_rate": 2.3111111111111112e-05,
"loss": 2.4238,
"step": 198000
},
{
"epoch": 74.88,
"learning_rate": 2.288888888888889e-05,
"loss": 2.4231,
"step": 198500
},
{
"epoch": 75.07,
"learning_rate": 2.2666666666666668e-05,
"loss": 2.4231,
"step": 199000
},
{
"epoch": 75.25,
"learning_rate": 2.2444444444444447e-05,
"loss": 2.4238,
"step": 199500
},
{
"epoch": 75.44,
"learning_rate": 2.2222222222222223e-05,
"loss": 2.4213,
"step": 200000
},
{
"epoch": 75.44,
"eval_loss": 2.328047513961792,
"eval_runtime": 283.8545,
"eval_samples_per_second": 485.643,
"eval_steps_per_second": 3.794,
"step": 200000
},
{
"epoch": 75.63,
"learning_rate": 2.2000000000000003e-05,
"loss": 2.4195,
"step": 200500
},
{
"epoch": 75.82,
"learning_rate": 2.177777777777778e-05,
"loss": 2.427,
"step": 201000
},
{
"epoch": 76.01,
"learning_rate": 2.1555555555555555e-05,
"loss": 2.4211,
"step": 201500
},
{
"epoch": 76.2,
"learning_rate": 2.1333333333333335e-05,
"loss": 2.421,
"step": 202000
},
{
"epoch": 76.39,
"learning_rate": 2.111111111111111e-05,
"loss": 2.4212,
"step": 202500
},
{
"epoch": 76.57,
"learning_rate": 2.088888888888889e-05,
"loss": 2.4211,
"step": 203000
},
{
"epoch": 76.76,
"learning_rate": 2.0666666666666666e-05,
"loss": 2.4172,
"step": 203500
},
{
"epoch": 76.95,
"learning_rate": 2.0444444444444446e-05,
"loss": 2.4197,
"step": 204000
},
{
"epoch": 77.14,
"learning_rate": 2.0222222222222222e-05,
"loss": 2.4202,
"step": 204500
},
{
"epoch": 77.33,
"learning_rate": 2e-05,
"loss": 2.4198,
"step": 205000
},
{
"epoch": 77.52,
"learning_rate": 1.9777777777777778e-05,
"loss": 2.4182,
"step": 205500
},
{
"epoch": 77.71,
"learning_rate": 1.9555555555555557e-05,
"loss": 2.4151,
"step": 206000
},
{
"epoch": 77.89,
"learning_rate": 1.9333333333333333e-05,
"loss": 2.417,
"step": 206500
},
{
"epoch": 78.08,
"learning_rate": 1.9111111111111113e-05,
"loss": 2.4166,
"step": 207000
},
{
"epoch": 78.27,
"learning_rate": 1.888888888888889e-05,
"loss": 2.4173,
"step": 207500
},
{
"epoch": 78.46,
"learning_rate": 1.866666666666667e-05,
"loss": 2.413,
"step": 208000
},
{
"epoch": 78.65,
"learning_rate": 1.8444444444444445e-05,
"loss": 2.4116,
"step": 208500
},
{
"epoch": 78.84,
"learning_rate": 1.8222222222222224e-05,
"loss": 2.4129,
"step": 209000
},
{
"epoch": 79.03,
"learning_rate": 1.8e-05,
"loss": 2.4186,
"step": 209500
},
{
"epoch": 79.22,
"learning_rate": 1.777777777777778e-05,
"loss": 2.4105,
"step": 210000
},
{
"epoch": 79.4,
"learning_rate": 1.7555555555555556e-05,
"loss": 2.4136,
"step": 210500
},
{
"epoch": 79.59,
"learning_rate": 1.7333333333333336e-05,
"loss": 2.4117,
"step": 211000
},
{
"epoch": 79.78,
"learning_rate": 1.7111111111111112e-05,
"loss": 2.412,
"step": 211500
},
{
"epoch": 79.97,
"learning_rate": 1.688888888888889e-05,
"loss": 2.414,
"step": 212000
},
{
"epoch": 80.16,
"learning_rate": 1.6666666666666667e-05,
"loss": 2.4166,
"step": 212500
},
{
"epoch": 80.16,
"eval_loss": 2.32003116607666,
"eval_runtime": 283.7234,
"eval_samples_per_second": 485.867,
"eval_steps_per_second": 3.796,
"step": 212500
},
{
"epoch": 80.35,
"learning_rate": 1.6444444444444447e-05,
"loss": 2.4142,
"step": 213000
},
{
"epoch": 80.54,
"learning_rate": 1.6222222222222223e-05,
"loss": 2.4085,
"step": 213500
},
{
"epoch": 80.72,
"learning_rate": 1.6000000000000003e-05,
"loss": 2.4121,
"step": 214000
},
{
"epoch": 80.91,
"learning_rate": 1.577777777777778e-05,
"loss": 2.4077,
"step": 214500
},
{
"epoch": 81.1,
"learning_rate": 1.5555555555555555e-05,
"loss": 2.4129,
"step": 215000
},
{
"epoch": 81.29,
"learning_rate": 1.5333333333333334e-05,
"loss": 2.4103,
"step": 215500
},
{
"epoch": 81.48,
"learning_rate": 1.5111111111111112e-05,
"loss": 2.4116,
"step": 216000
},
{
"epoch": 81.67,
"learning_rate": 1.4888888888888888e-05,
"loss": 2.4073,
"step": 216500
},
{
"epoch": 81.86,
"learning_rate": 1.4666666666666668e-05,
"loss": 2.4111,
"step": 217000
},
{
"epoch": 82.04,
"learning_rate": 1.4444444444444444e-05,
"loss": 2.4099,
"step": 217500
},
{
"epoch": 82.23,
"learning_rate": 1.4222222222222224e-05,
"loss": 2.4051,
"step": 218000
},
{
"epoch": 82.42,
"learning_rate": 1.4000000000000001e-05,
"loss": 2.406,
"step": 218500
},
{
"epoch": 82.61,
"learning_rate": 1.3777777777777778e-05,
"loss": 2.4115,
"step": 219000
},
{
"epoch": 82.8,
"learning_rate": 1.3555555555555557e-05,
"loss": 2.4027,
"step": 219500
},
{
"epoch": 82.99,
"learning_rate": 1.3333333333333333e-05,
"loss": 2.4124,
"step": 220000
},
{
"epoch": 83.18,
"learning_rate": 1.3111111111111113e-05,
"loss": 2.4104,
"step": 220500
},
{
"epoch": 83.36,
"learning_rate": 1.2888888888888889e-05,
"loss": 2.405,
"step": 221000
},
{
"epoch": 83.55,
"learning_rate": 1.2666666666666668e-05,
"loss": 2.4078,
"step": 221500
},
{
"epoch": 83.74,
"learning_rate": 1.2444444444444445e-05,
"loss": 2.4091,
"step": 222000
},
{
"epoch": 83.93,
"learning_rate": 1.2222222222222222e-05,
"loss": 2.4013,
"step": 222500
},
{
"epoch": 84.12,
"learning_rate": 1.2e-05,
"loss": 2.4117,
"step": 223000
},
{
"epoch": 84.31,
"learning_rate": 1.1777777777777778e-05,
"loss": 2.408,
"step": 223500
},
{
"epoch": 84.5,
"learning_rate": 1.1555555555555556e-05,
"loss": 2.4035,
"step": 224000
},
{
"epoch": 84.68,
"learning_rate": 1.1333333333333334e-05,
"loss": 2.4005,
"step": 224500
},
{
"epoch": 84.87,
"learning_rate": 1.1111111111111112e-05,
"loss": 2.3995,
"step": 225000
},
{
"epoch": 84.87,
"eval_loss": 2.3128256797790527,
"eval_runtime": 283.7082,
"eval_samples_per_second": 485.894,
"eval_steps_per_second": 3.796,
"step": 225000
},
{
"epoch": 85.06,
"learning_rate": 1.088888888888889e-05,
"loss": 2.41,
"step": 225500
},
{
"epoch": 85.25,
"learning_rate": 1.0666666666666667e-05,
"loss": 2.4006,
"step": 226000
},
{
"epoch": 85.44,
"learning_rate": 1.0444444444444445e-05,
"loss": 2.4071,
"step": 226500
},
{
"epoch": 85.63,
"learning_rate": 1.0222222222222223e-05,
"loss": 2.4026,
"step": 227000
},
{
"epoch": 85.82,
"learning_rate": 1e-05,
"loss": 2.4054,
"step": 227500
},
{
"epoch": 86.01,
"learning_rate": 9.777777777777779e-06,
"loss": 2.406,
"step": 228000
},
{
"epoch": 86.19,
"learning_rate": 9.555555555555556e-06,
"loss": 2.4008,
"step": 228500
},
{
"epoch": 86.38,
"learning_rate": 9.333333333333334e-06,
"loss": 2.4,
"step": 229000
},
{
"epoch": 86.57,
"learning_rate": 9.111111111111112e-06,
"loss": 2.4014,
"step": 229500
},
{
"epoch": 86.76,
"learning_rate": 8.88888888888889e-06,
"loss": 2.4026,
"step": 230000
},
{
"epoch": 86.95,
"learning_rate": 8.666666666666668e-06,
"loss": 2.4038,
"step": 230500
},
{
"epoch": 87.14,
"learning_rate": 8.444444444444446e-06,
"loss": 2.4036,
"step": 231000
},
{
"epoch": 87.33,
"learning_rate": 8.222222222222223e-06,
"loss": 2.4024,
"step": 231500
},
{
"epoch": 87.51,
"learning_rate": 8.000000000000001e-06,
"loss": 2.4045,
"step": 232000
},
{
"epoch": 87.7,
"learning_rate": 7.777777777777777e-06,
"loss": 2.4001,
"step": 232500
},
{
"epoch": 87.89,
"learning_rate": 7.555555555555556e-06,
"loss": 2.4001,
"step": 233000
},
{
"epoch": 88.08,
"learning_rate": 7.333333333333334e-06,
"loss": 2.4025,
"step": 233500
},
{
"epoch": 88.27,
"learning_rate": 7.111111111111112e-06,
"loss": 2.3957,
"step": 234000
},
{
"epoch": 88.46,
"learning_rate": 6.888888888888889e-06,
"loss": 2.3971,
"step": 234500
},
{
"epoch": 88.65,
"learning_rate": 6.666666666666667e-06,
"loss": 2.4007,
"step": 235000
},
{
"epoch": 88.83,
"learning_rate": 6.4444444444444445e-06,
"loss": 2.4002,
"step": 235500
},
{
"epoch": 89.02,
"learning_rate": 6.222222222222222e-06,
"loss": 2.3995,
"step": 236000
},
{
"epoch": 89.21,
"learning_rate": 6e-06,
"loss": 2.3985,
"step": 236500
},
{
"epoch": 89.4,
"learning_rate": 5.777777777777778e-06,
"loss": 2.4028,
"step": 237000
},
{
"epoch": 89.59,
"learning_rate": 5.555555555555556e-06,
"loss": 2.395,
"step": 237500
},
{
"epoch": 89.59,
"eval_loss": 2.3064496517181396,
"eval_runtime": 283.9176,
"eval_samples_per_second": 485.535,
"eval_steps_per_second": 3.793,
"step": 237500
},
{
"epoch": 89.78,
"learning_rate": 5.333333333333334e-06,
"loss": 2.3968,
"step": 238000
},
{
"epoch": 89.97,
"learning_rate": 5.1111111111111115e-06,
"loss": 2.3952,
"step": 238500
},
{
"epoch": 90.15,
"learning_rate": 4.888888888888889e-06,
"loss": 2.4006,
"step": 239000
},
{
"epoch": 90.34,
"learning_rate": 4.666666666666667e-06,
"loss": 2.3952,
"step": 239500
},
{
"epoch": 90.53,
"learning_rate": 4.444444444444445e-06,
"loss": 2.3949,
"step": 240000
},
{
"epoch": 90.72,
"learning_rate": 4.222222222222223e-06,
"loss": 2.3966,
"step": 240500
},
{
"epoch": 90.91,
"learning_rate": 4.000000000000001e-06,
"loss": 2.3955,
"step": 241000
},
{
"epoch": 91.1,
"learning_rate": 3.777777777777778e-06,
"loss": 2.4001,
"step": 241500
},
{
"epoch": 91.29,
"learning_rate": 3.555555555555556e-06,
"loss": 2.3956,
"step": 242000
},
{
"epoch": 91.47,
"learning_rate": 3.3333333333333333e-06,
"loss": 2.3951,
"step": 242500
},
{
"epoch": 91.66,
"learning_rate": 3.111111111111111e-06,
"loss": 2.4001,
"step": 243000
},
{
"epoch": 91.85,
"learning_rate": 2.888888888888889e-06,
"loss": 2.3945,
"step": 243500
},
{
"epoch": 92.04,
"learning_rate": 2.666666666666667e-06,
"loss": 2.395,
"step": 244000
},
{
"epoch": 92.23,
"learning_rate": 2.4444444444444447e-06,
"loss": 2.3963,
"step": 244500
},
{
"epoch": 92.42,
"learning_rate": 2.2222222222222225e-06,
"loss": 2.3944,
"step": 245000
},
{
"epoch": 92.61,
"learning_rate": 2.0000000000000003e-06,
"loss": 2.3951,
"step": 245500
},
{
"epoch": 92.79,
"learning_rate": 1.777777777777778e-06,
"loss": 2.3897,
"step": 246000
},
{
"epoch": 92.98,
"learning_rate": 1.5555555555555556e-06,
"loss": 2.3926,
"step": 246500
},
{
"epoch": 93.17,
"learning_rate": 1.3333333333333334e-06,
"loss": 2.3968,
"step": 247000
},
{
"epoch": 93.36,
"learning_rate": 1.1111111111111112e-06,
"loss": 2.3924,
"step": 247500
},
{
"epoch": 93.55,
"learning_rate": 8.88888888888889e-07,
"loss": 2.3908,
"step": 248000
},
{
"epoch": 93.74,
"learning_rate": 6.666666666666667e-07,
"loss": 2.3932,
"step": 248500
},
{
"epoch": 93.93,
"learning_rate": 4.444444444444445e-07,
"loss": 2.3942,
"step": 249000
},
{
"epoch": 94.12,
"learning_rate": 2.2222222222222224e-07,
"loss": 2.398,
"step": 249500
},
{
"epoch": 94.3,
"learning_rate": 0.0,
"loss": 2.3932,
"step": 250000
},
{
"epoch": 94.3,
"eval_loss": 2.302946090698242,
"eval_runtime": 283.7177,
"eval_samples_per_second": 485.877,
"eval_steps_per_second": 3.796,
"step": 250000
}
],
"max_steps": 250000,
"num_train_epochs": 95,
"total_flos": 4.2134352665063916e+18,
"trial_name": null,
"trial_params": null
}