roberta-base-wechsel-german / trainer_state.json
benjamin's picture
initial commit
f63ec76
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 74.98500299940012,
"global_step": 250000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.15,
"learning_rate": 2.0000000000000003e-06,
"loss": 6.3557,
"step": 500
},
{
"epoch": 0.3,
"learning_rate": 4.000000000000001e-06,
"loss": 5.4171,
"step": 1000
},
{
"epoch": 0.45,
"learning_rate": 6e-06,
"loss": 4.8594,
"step": 1500
},
{
"epoch": 0.6,
"learning_rate": 8.000000000000001e-06,
"loss": 4.488,
"step": 2000
},
{
"epoch": 0.75,
"learning_rate": 1e-05,
"loss": 4.2125,
"step": 2500
},
{
"epoch": 0.9,
"learning_rate": 1.2e-05,
"loss": 3.9836,
"step": 3000
},
{
"epoch": 1.05,
"learning_rate": 1.4000000000000001e-05,
"loss": 3.794,
"step": 3500
},
{
"epoch": 1.2,
"learning_rate": 1.6000000000000003e-05,
"loss": 3.6179,
"step": 4000
},
{
"epoch": 1.35,
"learning_rate": 1.8e-05,
"loss": 3.4697,
"step": 4500
},
{
"epoch": 1.5,
"learning_rate": 2e-05,
"loss": 3.3321,
"step": 5000
},
{
"epoch": 1.65,
"learning_rate": 2.2000000000000003e-05,
"loss": 3.2206,
"step": 5500
},
{
"epoch": 1.8,
"learning_rate": 2.4e-05,
"loss": 3.1178,
"step": 6000
},
{
"epoch": 1.95,
"learning_rate": 2.6000000000000002e-05,
"loss": 3.0262,
"step": 6500
},
{
"epoch": 2.1,
"learning_rate": 2.8000000000000003e-05,
"loss": 2.9351,
"step": 7000
},
{
"epoch": 2.25,
"learning_rate": 3e-05,
"loss": 2.8609,
"step": 7500
},
{
"epoch": 2.4,
"learning_rate": 3.2000000000000005e-05,
"loss": 2.793,
"step": 8000
},
{
"epoch": 2.55,
"learning_rate": 3.4000000000000007e-05,
"loss": 2.7255,
"step": 8500
},
{
"epoch": 2.7,
"learning_rate": 3.6e-05,
"loss": 2.6665,
"step": 9000
},
{
"epoch": 2.85,
"learning_rate": 3.8e-05,
"loss": 2.6135,
"step": 9500
},
{
"epoch": 3.0,
"learning_rate": 4e-05,
"loss": 2.5613,
"step": 10000
},
{
"epoch": 3.15,
"learning_rate": 4.2e-05,
"loss": 2.5102,
"step": 10500
},
{
"epoch": 3.3,
"learning_rate": 4.4000000000000006e-05,
"loss": 2.4629,
"step": 11000
},
{
"epoch": 3.45,
"learning_rate": 4.600000000000001e-05,
"loss": 2.419,
"step": 11500
},
{
"epoch": 3.6,
"learning_rate": 4.8e-05,
"loss": 2.3811,
"step": 12000
},
{
"epoch": 3.75,
"learning_rate": 5e-05,
"loss": 2.3446,
"step": 12500
},
{
"epoch": 3.75,
"eval_loss": 2.2050914764404297,
"eval_runtime": 371.9663,
"eval_samples_per_second": 458.074,
"eval_steps_per_second": 3.581,
"step": 12500
},
{
"epoch": 3.9,
"learning_rate": 5.2000000000000004e-05,
"loss": 2.3102,
"step": 13000
},
{
"epoch": 4.05,
"learning_rate": 5.4000000000000005e-05,
"loss": 2.2772,
"step": 13500
},
{
"epoch": 4.2,
"learning_rate": 5.6000000000000006e-05,
"loss": 2.2466,
"step": 14000
},
{
"epoch": 4.35,
"learning_rate": 5.8e-05,
"loss": 2.2166,
"step": 14500
},
{
"epoch": 4.5,
"learning_rate": 6e-05,
"loss": 2.1951,
"step": 15000
},
{
"epoch": 4.65,
"learning_rate": 6.2e-05,
"loss": 2.1697,
"step": 15500
},
{
"epoch": 4.8,
"learning_rate": 6.400000000000001e-05,
"loss": 2.1481,
"step": 16000
},
{
"epoch": 4.95,
"learning_rate": 6.6e-05,
"loss": 2.1281,
"step": 16500
},
{
"epoch": 5.1,
"learning_rate": 6.800000000000001e-05,
"loss": 2.1042,
"step": 17000
},
{
"epoch": 5.25,
"learning_rate": 7e-05,
"loss": 2.0863,
"step": 17500
},
{
"epoch": 5.4,
"learning_rate": 7.2e-05,
"loss": 2.0667,
"step": 18000
},
{
"epoch": 5.55,
"learning_rate": 7.4e-05,
"loss": 2.0539,
"step": 18500
},
{
"epoch": 5.7,
"learning_rate": 7.6e-05,
"loss": 2.0358,
"step": 19000
},
{
"epoch": 5.85,
"learning_rate": 7.800000000000001e-05,
"loss": 2.0244,
"step": 19500
},
{
"epoch": 6.0,
"learning_rate": 8e-05,
"loss": 2.0081,
"step": 20000
},
{
"epoch": 6.15,
"learning_rate": 8.2e-05,
"loss": 1.9923,
"step": 20500
},
{
"epoch": 6.3,
"learning_rate": 8.4e-05,
"loss": 1.9846,
"step": 21000
},
{
"epoch": 6.45,
"learning_rate": 8.6e-05,
"loss": 1.968,
"step": 21500
},
{
"epoch": 6.6,
"learning_rate": 8.800000000000001e-05,
"loss": 1.9588,
"step": 22000
},
{
"epoch": 6.75,
"learning_rate": 9e-05,
"loss": 1.9472,
"step": 22500
},
{
"epoch": 6.9,
"learning_rate": 9.200000000000001e-05,
"loss": 1.9378,
"step": 23000
},
{
"epoch": 7.05,
"learning_rate": 9.4e-05,
"loss": 1.9285,
"step": 23500
},
{
"epoch": 7.2,
"learning_rate": 9.6e-05,
"loss": 1.9167,
"step": 24000
},
{
"epoch": 7.35,
"learning_rate": 9.8e-05,
"loss": 1.9053,
"step": 24500
},
{
"epoch": 7.5,
"learning_rate": 0.0001,
"loss": 1.8975,
"step": 25000
},
{
"epoch": 7.5,
"eval_loss": 1.7942529916763306,
"eval_runtime": 357.1217,
"eval_samples_per_second": 477.115,
"eval_steps_per_second": 3.73,
"step": 25000
},
{
"epoch": 7.65,
"learning_rate": 9.977777777777779e-05,
"loss": 1.8928,
"step": 25500
},
{
"epoch": 7.8,
"learning_rate": 9.955555555555556e-05,
"loss": 1.8814,
"step": 26000
},
{
"epoch": 7.95,
"learning_rate": 9.933333333333334e-05,
"loss": 1.8761,
"step": 26500
},
{
"epoch": 8.1,
"learning_rate": 9.911111111111112e-05,
"loss": 1.8629,
"step": 27000
},
{
"epoch": 8.25,
"learning_rate": 9.888888888888889e-05,
"loss": 1.8538,
"step": 27500
},
{
"epoch": 8.4,
"learning_rate": 9.866666666666668e-05,
"loss": 1.8486,
"step": 28000
},
{
"epoch": 8.55,
"learning_rate": 9.844444444444444e-05,
"loss": 1.8439,
"step": 28500
},
{
"epoch": 8.7,
"learning_rate": 9.822222222222223e-05,
"loss": 1.835,
"step": 29000
},
{
"epoch": 8.85,
"learning_rate": 9.8e-05,
"loss": 1.8286,
"step": 29500
},
{
"epoch": 9.0,
"learning_rate": 9.777777777777778e-05,
"loss": 1.8235,
"step": 30000
},
{
"epoch": 9.15,
"learning_rate": 9.755555555555555e-05,
"loss": 1.8151,
"step": 30500
},
{
"epoch": 9.3,
"learning_rate": 9.733333333333335e-05,
"loss": 1.81,
"step": 31000
},
{
"epoch": 9.45,
"learning_rate": 9.711111111111111e-05,
"loss": 1.8049,
"step": 31500
},
{
"epoch": 9.6,
"learning_rate": 9.68888888888889e-05,
"loss": 1.7986,
"step": 32000
},
{
"epoch": 9.75,
"learning_rate": 9.666666666666667e-05,
"loss": 1.797,
"step": 32500
},
{
"epoch": 9.9,
"learning_rate": 9.644444444444445e-05,
"loss": 1.7878,
"step": 33000
},
{
"epoch": 10.05,
"learning_rate": 9.622222222222222e-05,
"loss": 1.7842,
"step": 33500
},
{
"epoch": 10.2,
"learning_rate": 9.6e-05,
"loss": 1.7774,
"step": 34000
},
{
"epoch": 10.35,
"learning_rate": 9.577777777777777e-05,
"loss": 1.7719,
"step": 34500
},
{
"epoch": 10.5,
"learning_rate": 9.555555555555557e-05,
"loss": 1.769,
"step": 35000
},
{
"epoch": 10.65,
"learning_rate": 9.533333333333334e-05,
"loss": 1.7678,
"step": 35500
},
{
"epoch": 10.8,
"learning_rate": 9.511111111111112e-05,
"loss": 1.7598,
"step": 36000
},
{
"epoch": 10.95,
"learning_rate": 9.488888888888889e-05,
"loss": 1.759,
"step": 36500
},
{
"epoch": 11.1,
"learning_rate": 9.466666666666667e-05,
"loss": 1.7488,
"step": 37000
},
{
"epoch": 11.25,
"learning_rate": 9.444444444444444e-05,
"loss": 1.7479,
"step": 37500
},
{
"epoch": 11.25,
"eval_loss": 1.6556859016418457,
"eval_runtime": 356.8188,
"eval_samples_per_second": 477.52,
"eval_steps_per_second": 3.733,
"step": 37500
},
{
"epoch": 11.4,
"learning_rate": 9.422222222222223e-05,
"loss": 1.7473,
"step": 38000
},
{
"epoch": 11.55,
"learning_rate": 9.4e-05,
"loss": 1.7423,
"step": 38500
},
{
"epoch": 11.7,
"learning_rate": 9.377777777777779e-05,
"loss": 1.7374,
"step": 39000
},
{
"epoch": 11.85,
"learning_rate": 9.355555555555556e-05,
"loss": 1.7334,
"step": 39500
},
{
"epoch": 12.0,
"learning_rate": 9.333333333333334e-05,
"loss": 1.7315,
"step": 40000
},
{
"epoch": 12.15,
"learning_rate": 9.311111111111111e-05,
"loss": 1.725,
"step": 40500
},
{
"epoch": 12.3,
"learning_rate": 9.28888888888889e-05,
"loss": 1.722,
"step": 41000
},
{
"epoch": 12.45,
"learning_rate": 9.266666666666666e-05,
"loss": 1.7219,
"step": 41500
},
{
"epoch": 12.6,
"learning_rate": 9.244444444444445e-05,
"loss": 1.7211,
"step": 42000
},
{
"epoch": 12.75,
"learning_rate": 9.222222222222223e-05,
"loss": 1.716,
"step": 42500
},
{
"epoch": 12.9,
"learning_rate": 9.200000000000001e-05,
"loss": 1.7118,
"step": 43000
},
{
"epoch": 13.05,
"learning_rate": 9.177777777777778e-05,
"loss": 1.7106,
"step": 43500
},
{
"epoch": 13.2,
"learning_rate": 9.155555555555557e-05,
"loss": 1.7033,
"step": 44000
},
{
"epoch": 13.35,
"learning_rate": 9.133333333333334e-05,
"loss": 1.703,
"step": 44500
},
{
"epoch": 13.5,
"learning_rate": 9.111111111111112e-05,
"loss": 1.7002,
"step": 45000
},
{
"epoch": 13.65,
"learning_rate": 9.088888888888889e-05,
"loss": 1.7012,
"step": 45500
},
{
"epoch": 13.8,
"learning_rate": 9.066666666666667e-05,
"loss": 1.6943,
"step": 46000
},
{
"epoch": 13.95,
"learning_rate": 9.044444444444445e-05,
"loss": 1.6932,
"step": 46500
},
{
"epoch": 14.1,
"learning_rate": 9.022222222222224e-05,
"loss": 1.6912,
"step": 47000
},
{
"epoch": 14.25,
"learning_rate": 9e-05,
"loss": 1.687,
"step": 47500
},
{
"epoch": 14.4,
"learning_rate": 8.977777777777779e-05,
"loss": 1.6875,
"step": 48000
},
{
"epoch": 14.55,
"learning_rate": 8.955555555555556e-05,
"loss": 1.6849,
"step": 48500
},
{
"epoch": 14.7,
"learning_rate": 8.933333333333334e-05,
"loss": 1.6788,
"step": 49000
},
{
"epoch": 14.85,
"learning_rate": 8.911111111111111e-05,
"loss": 1.6801,
"step": 49500
},
{
"epoch": 15.0,
"learning_rate": 8.888888888888889e-05,
"loss": 1.6774,
"step": 50000
},
{
"epoch": 15.0,
"eval_loss": 1.5897281169891357,
"eval_runtime": 357.0501,
"eval_samples_per_second": 477.21,
"eval_steps_per_second": 3.731,
"step": 50000
},
{
"epoch": 15.15,
"learning_rate": 8.866666666666668e-05,
"loss": 1.6729,
"step": 50500
},
{
"epoch": 15.3,
"learning_rate": 8.844444444444445e-05,
"loss": 1.672,
"step": 51000
},
{
"epoch": 15.45,
"learning_rate": 8.822222222222223e-05,
"loss": 1.6724,
"step": 51500
},
{
"epoch": 15.6,
"learning_rate": 8.800000000000001e-05,
"loss": 1.6688,
"step": 52000
},
{
"epoch": 15.75,
"learning_rate": 8.777777777777778e-05,
"loss": 1.6658,
"step": 52500
},
{
"epoch": 15.9,
"learning_rate": 8.755555555555556e-05,
"loss": 1.6633,
"step": 53000
},
{
"epoch": 16.05,
"learning_rate": 8.733333333333333e-05,
"loss": 1.6618,
"step": 53500
},
{
"epoch": 16.2,
"learning_rate": 8.711111111111112e-05,
"loss": 1.6586,
"step": 54000
},
{
"epoch": 16.35,
"learning_rate": 8.68888888888889e-05,
"loss": 1.6573,
"step": 54500
},
{
"epoch": 16.5,
"learning_rate": 8.666666666666667e-05,
"loss": 1.6554,
"step": 55000
},
{
"epoch": 16.65,
"learning_rate": 8.644444444444445e-05,
"loss": 1.6576,
"step": 55500
},
{
"epoch": 16.8,
"learning_rate": 8.622222222222222e-05,
"loss": 1.6539,
"step": 56000
},
{
"epoch": 16.95,
"learning_rate": 8.6e-05,
"loss": 1.6523,
"step": 56500
},
{
"epoch": 17.1,
"learning_rate": 8.577777777777777e-05,
"loss": 1.6473,
"step": 57000
},
{
"epoch": 17.25,
"learning_rate": 8.555555555555556e-05,
"loss": 1.6475,
"step": 57500
},
{
"epoch": 17.4,
"learning_rate": 8.533333333333334e-05,
"loss": 1.647,
"step": 58000
},
{
"epoch": 17.55,
"learning_rate": 8.511111111111112e-05,
"loss": 1.6421,
"step": 58500
},
{
"epoch": 17.7,
"learning_rate": 8.488888888888889e-05,
"loss": 1.6444,
"step": 59000
},
{
"epoch": 17.85,
"learning_rate": 8.466666666666667e-05,
"loss": 1.6446,
"step": 59500
},
{
"epoch": 18.0,
"learning_rate": 8.444444444444444e-05,
"loss": 1.6423,
"step": 60000
},
{
"epoch": 18.15,
"learning_rate": 8.422222222222223e-05,
"loss": 1.6394,
"step": 60500
},
{
"epoch": 18.3,
"learning_rate": 8.4e-05,
"loss": 1.6348,
"step": 61000
},
{
"epoch": 18.45,
"learning_rate": 8.377777777777778e-05,
"loss": 1.633,
"step": 61500
},
{
"epoch": 18.6,
"learning_rate": 8.355555555555556e-05,
"loss": 1.6364,
"step": 62000
},
{
"epoch": 18.75,
"learning_rate": 8.333333333333334e-05,
"loss": 1.6314,
"step": 62500
},
{
"epoch": 18.75,
"eval_loss": 1.5525341033935547,
"eval_runtime": 356.5096,
"eval_samples_per_second": 477.934,
"eval_steps_per_second": 3.736,
"step": 62500
},
{
"epoch": 18.9,
"learning_rate": 8.311111111111111e-05,
"loss": 1.6316,
"step": 63000
},
{
"epoch": 19.05,
"learning_rate": 8.28888888888889e-05,
"loss": 1.6315,
"step": 63500
},
{
"epoch": 19.2,
"learning_rate": 8.266666666666667e-05,
"loss": 1.6288,
"step": 64000
},
{
"epoch": 19.35,
"learning_rate": 8.244444444444445e-05,
"loss": 1.6277,
"step": 64500
},
{
"epoch": 19.5,
"learning_rate": 8.222222222222222e-05,
"loss": 1.6258,
"step": 65000
},
{
"epoch": 19.65,
"learning_rate": 8.2e-05,
"loss": 1.6223,
"step": 65500
},
{
"epoch": 19.8,
"learning_rate": 8.177777777777778e-05,
"loss": 1.6217,
"step": 66000
},
{
"epoch": 19.95,
"learning_rate": 8.155555555555557e-05,
"loss": 1.6254,
"step": 66500
},
{
"epoch": 20.1,
"learning_rate": 8.133333333333334e-05,
"loss": 1.6196,
"step": 67000
},
{
"epoch": 20.25,
"learning_rate": 8.111111111111112e-05,
"loss": 1.6172,
"step": 67500
},
{
"epoch": 20.4,
"learning_rate": 8.088888888888889e-05,
"loss": 1.6162,
"step": 68000
},
{
"epoch": 20.55,
"learning_rate": 8.066666666666667e-05,
"loss": 1.6136,
"step": 68500
},
{
"epoch": 20.7,
"learning_rate": 8.044444444444444e-05,
"loss": 1.6162,
"step": 69000
},
{
"epoch": 20.85,
"learning_rate": 8.022222222222222e-05,
"loss": 1.6153,
"step": 69500
},
{
"epoch": 21.0,
"learning_rate": 8e-05,
"loss": 1.613,
"step": 70000
},
{
"epoch": 21.15,
"learning_rate": 7.977777777777779e-05,
"loss": 1.6098,
"step": 70500
},
{
"epoch": 21.3,
"learning_rate": 7.955555555555556e-05,
"loss": 1.6115,
"step": 71000
},
{
"epoch": 21.45,
"learning_rate": 7.933333333333334e-05,
"loss": 1.6077,
"step": 71500
},
{
"epoch": 21.6,
"learning_rate": 7.911111111111111e-05,
"loss": 1.6077,
"step": 72000
},
{
"epoch": 21.75,
"learning_rate": 7.88888888888889e-05,
"loss": 1.607,
"step": 72500
},
{
"epoch": 21.9,
"learning_rate": 7.866666666666666e-05,
"loss": 1.607,
"step": 73000
},
{
"epoch": 22.05,
"learning_rate": 7.844444444444446e-05,
"loss": 1.6067,
"step": 73500
},
{
"epoch": 22.2,
"learning_rate": 7.822222222222223e-05,
"loss": 1.6024,
"step": 74000
},
{
"epoch": 22.35,
"learning_rate": 7.800000000000001e-05,
"loss": 1.6028,
"step": 74500
},
{
"epoch": 22.5,
"learning_rate": 7.777777777777778e-05,
"loss": 1.6017,
"step": 75000
},
{
"epoch": 22.5,
"eval_loss": 1.525025725364685,
"eval_runtime": 356.6664,
"eval_samples_per_second": 477.724,
"eval_steps_per_second": 3.735,
"step": 75000
},
{
"epoch": 22.65,
"learning_rate": 7.755555555555556e-05,
"loss": 1.5995,
"step": 75500
},
{
"epoch": 22.8,
"learning_rate": 7.733333333333333e-05,
"loss": 1.596,
"step": 76000
},
{
"epoch": 22.95,
"learning_rate": 7.711111111111112e-05,
"loss": 1.6002,
"step": 76500
},
{
"epoch": 23.1,
"learning_rate": 7.688888888888889e-05,
"loss": 1.5939,
"step": 77000
},
{
"epoch": 23.25,
"learning_rate": 7.666666666666667e-05,
"loss": 1.5929,
"step": 77500
},
{
"epoch": 23.4,
"learning_rate": 7.644444444444445e-05,
"loss": 1.5948,
"step": 78000
},
{
"epoch": 23.55,
"learning_rate": 7.622222222222223e-05,
"loss": 1.5921,
"step": 78500
},
{
"epoch": 23.7,
"learning_rate": 7.6e-05,
"loss": 1.5917,
"step": 79000
},
{
"epoch": 23.85,
"learning_rate": 7.577777777777779e-05,
"loss": 1.5953,
"step": 79500
},
{
"epoch": 24.0,
"learning_rate": 7.555555555555556e-05,
"loss": 1.5909,
"step": 80000
},
{
"epoch": 24.15,
"learning_rate": 7.533333333333334e-05,
"loss": 1.5886,
"step": 80500
},
{
"epoch": 24.3,
"learning_rate": 7.511111111111111e-05,
"loss": 1.5873,
"step": 81000
},
{
"epoch": 24.45,
"learning_rate": 7.488888888888889e-05,
"loss": 1.5863,
"step": 81500
},
{
"epoch": 24.6,
"learning_rate": 7.466666666666667e-05,
"loss": 1.5861,
"step": 82000
},
{
"epoch": 24.75,
"learning_rate": 7.444444444444444e-05,
"loss": 1.5845,
"step": 82500
},
{
"epoch": 24.9,
"learning_rate": 7.422222222222223e-05,
"loss": 1.5847,
"step": 83000
},
{
"epoch": 25.04,
"learning_rate": 7.4e-05,
"loss": 1.5832,
"step": 83500
},
{
"epoch": 25.19,
"learning_rate": 7.377777777777778e-05,
"loss": 1.5804,
"step": 84000
},
{
"epoch": 25.34,
"learning_rate": 7.355555555555556e-05,
"loss": 1.5804,
"step": 84500
},
{
"epoch": 25.49,
"learning_rate": 7.333333333333333e-05,
"loss": 1.5809,
"step": 85000
},
{
"epoch": 25.64,
"learning_rate": 7.311111111111111e-05,
"loss": 1.5785,
"step": 85500
},
{
"epoch": 25.79,
"learning_rate": 7.28888888888889e-05,
"loss": 1.5804,
"step": 86000
},
{
"epoch": 25.94,
"learning_rate": 7.266666666666667e-05,
"loss": 1.5796,
"step": 86500
},
{
"epoch": 26.09,
"learning_rate": 7.244444444444445e-05,
"loss": 1.5777,
"step": 87000
},
{
"epoch": 26.24,
"learning_rate": 7.222222222222222e-05,
"loss": 1.5761,
"step": 87500
},
{
"epoch": 26.24,
"eval_loss": 1.5014694929122925,
"eval_runtime": 356.8479,
"eval_samples_per_second": 477.481,
"eval_steps_per_second": 3.733,
"step": 87500
},
{
"epoch": 26.39,
"learning_rate": 7.2e-05,
"loss": 1.5762,
"step": 88000
},
{
"epoch": 26.54,
"learning_rate": 7.177777777777777e-05,
"loss": 1.5745,
"step": 88500
},
{
"epoch": 26.69,
"learning_rate": 7.155555555555555e-05,
"loss": 1.5726,
"step": 89000
},
{
"epoch": 26.84,
"learning_rate": 7.133333333333334e-05,
"loss": 1.5732,
"step": 89500
},
{
"epoch": 26.99,
"learning_rate": 7.111111111111112e-05,
"loss": 1.5723,
"step": 90000
},
{
"epoch": 27.14,
"learning_rate": 7.088888888888889e-05,
"loss": 1.57,
"step": 90500
},
{
"epoch": 27.29,
"learning_rate": 7.066666666666667e-05,
"loss": 1.5677,
"step": 91000
},
{
"epoch": 27.44,
"learning_rate": 7.044444444444444e-05,
"loss": 1.5684,
"step": 91500
},
{
"epoch": 27.59,
"learning_rate": 7.022222222222222e-05,
"loss": 1.5693,
"step": 92000
},
{
"epoch": 27.74,
"learning_rate": 7e-05,
"loss": 1.5676,
"step": 92500
},
{
"epoch": 27.89,
"learning_rate": 6.977777777777779e-05,
"loss": 1.5673,
"step": 93000
},
{
"epoch": 28.04,
"learning_rate": 6.955555555555556e-05,
"loss": 1.568,
"step": 93500
},
{
"epoch": 28.19,
"learning_rate": 6.933333333333334e-05,
"loss": 1.5646,
"step": 94000
},
{
"epoch": 28.34,
"learning_rate": 6.911111111111111e-05,
"loss": 1.5651,
"step": 94500
},
{
"epoch": 28.49,
"learning_rate": 6.88888888888889e-05,
"loss": 1.5658,
"step": 95000
},
{
"epoch": 28.64,
"learning_rate": 6.866666666666666e-05,
"loss": 1.563,
"step": 95500
},
{
"epoch": 28.79,
"learning_rate": 6.844444444444445e-05,
"loss": 1.5613,
"step": 96000
},
{
"epoch": 28.94,
"learning_rate": 6.822222222222222e-05,
"loss": 1.562,
"step": 96500
},
{
"epoch": 29.09,
"learning_rate": 6.800000000000001e-05,
"loss": 1.5602,
"step": 97000
},
{
"epoch": 29.24,
"learning_rate": 6.777777777777778e-05,
"loss": 1.5587,
"step": 97500
},
{
"epoch": 29.39,
"learning_rate": 6.755555555555557e-05,
"loss": 1.5586,
"step": 98000
},
{
"epoch": 29.54,
"learning_rate": 6.733333333333333e-05,
"loss": 1.5573,
"step": 98500
},
{
"epoch": 29.69,
"learning_rate": 6.711111111111112e-05,
"loss": 1.5574,
"step": 99000
},
{
"epoch": 29.84,
"learning_rate": 6.688888888888889e-05,
"loss": 1.5582,
"step": 99500
},
{
"epoch": 29.99,
"learning_rate": 6.666666666666667e-05,
"loss": 1.5571,
"step": 100000
},
{
"epoch": 29.99,
"eval_loss": 1.486186146736145,
"eval_runtime": 356.5839,
"eval_samples_per_second": 477.834,
"eval_steps_per_second": 3.735,
"step": 100000
},
{
"epoch": 30.14,
"learning_rate": 6.644444444444444e-05,
"loss": 1.5551,
"step": 100500
},
{
"epoch": 30.29,
"learning_rate": 6.622222222222224e-05,
"loss": 1.555,
"step": 101000
},
{
"epoch": 30.44,
"learning_rate": 6.6e-05,
"loss": 1.5542,
"step": 101500
},
{
"epoch": 30.59,
"learning_rate": 6.577777777777779e-05,
"loss": 1.5542,
"step": 102000
},
{
"epoch": 30.74,
"learning_rate": 6.555555555555556e-05,
"loss": 1.5527,
"step": 102500
},
{
"epoch": 30.89,
"learning_rate": 6.533333333333334e-05,
"loss": 1.5508,
"step": 103000
},
{
"epoch": 31.04,
"learning_rate": 6.511111111111111e-05,
"loss": 1.5514,
"step": 103500
},
{
"epoch": 31.19,
"learning_rate": 6.488888888888889e-05,
"loss": 1.5496,
"step": 104000
},
{
"epoch": 31.34,
"learning_rate": 6.466666666666666e-05,
"loss": 1.5478,
"step": 104500
},
{
"epoch": 31.49,
"learning_rate": 6.444444444444446e-05,
"loss": 1.5493,
"step": 105000
},
{
"epoch": 31.64,
"learning_rate": 6.422222222222223e-05,
"loss": 1.5473,
"step": 105500
},
{
"epoch": 31.79,
"learning_rate": 6.400000000000001e-05,
"loss": 1.5489,
"step": 106000
},
{
"epoch": 31.94,
"learning_rate": 6.377777777777778e-05,
"loss": 1.5475,
"step": 106500
},
{
"epoch": 32.09,
"learning_rate": 6.355555555555556e-05,
"loss": 1.5448,
"step": 107000
},
{
"epoch": 32.24,
"learning_rate": 6.333333333333333e-05,
"loss": 1.546,
"step": 107500
},
{
"epoch": 32.39,
"learning_rate": 6.311111111111112e-05,
"loss": 1.5472,
"step": 108000
},
{
"epoch": 32.54,
"learning_rate": 6.28888888888889e-05,
"loss": 1.5464,
"step": 108500
},
{
"epoch": 32.69,
"learning_rate": 6.266666666666667e-05,
"loss": 1.5435,
"step": 109000
},
{
"epoch": 32.84,
"learning_rate": 6.244444444444445e-05,
"loss": 1.5436,
"step": 109500
},
{
"epoch": 32.99,
"learning_rate": 6.222222222222222e-05,
"loss": 1.5449,
"step": 110000
},
{
"epoch": 33.14,
"learning_rate": 6.2e-05,
"loss": 1.5398,
"step": 110500
},
{
"epoch": 33.29,
"learning_rate": 6.177777777777779e-05,
"loss": 1.5387,
"step": 111000
},
{
"epoch": 33.44,
"learning_rate": 6.155555555555555e-05,
"loss": 1.5404,
"step": 111500
},
{
"epoch": 33.59,
"learning_rate": 6.133333333333334e-05,
"loss": 1.5397,
"step": 112000
},
{
"epoch": 33.74,
"learning_rate": 6.111111111111112e-05,
"loss": 1.5411,
"step": 112500
},
{
"epoch": 33.74,
"eval_loss": 1.4691314697265625,
"eval_runtime": 356.7348,
"eval_samples_per_second": 477.632,
"eval_steps_per_second": 3.734,
"step": 112500
},
{
"epoch": 33.89,
"learning_rate": 6.08888888888889e-05,
"loss": 1.5408,
"step": 113000
},
{
"epoch": 34.04,
"learning_rate": 6.066666666666667e-05,
"loss": 1.5398,
"step": 113500
},
{
"epoch": 34.19,
"learning_rate": 6.044444444444445e-05,
"loss": 1.5373,
"step": 114000
},
{
"epoch": 34.34,
"learning_rate": 6.0222222222222225e-05,
"loss": 1.5347,
"step": 114500
},
{
"epoch": 34.49,
"learning_rate": 6e-05,
"loss": 1.536,
"step": 115000
},
{
"epoch": 34.64,
"learning_rate": 5.977777777777778e-05,
"loss": 1.5363,
"step": 115500
},
{
"epoch": 34.79,
"learning_rate": 5.9555555555555554e-05,
"loss": 1.5343,
"step": 116000
},
{
"epoch": 34.94,
"learning_rate": 5.9333333333333343e-05,
"loss": 1.5377,
"step": 116500
},
{
"epoch": 35.09,
"learning_rate": 5.911111111111112e-05,
"loss": 1.5349,
"step": 117000
},
{
"epoch": 35.24,
"learning_rate": 5.8888888888888896e-05,
"loss": 1.5325,
"step": 117500
},
{
"epoch": 35.39,
"learning_rate": 5.866666666666667e-05,
"loss": 1.5325,
"step": 118000
},
{
"epoch": 35.54,
"learning_rate": 5.844444444444445e-05,
"loss": 1.5312,
"step": 118500
},
{
"epoch": 35.69,
"learning_rate": 5.8222222222222224e-05,
"loss": 1.5342,
"step": 119000
},
{
"epoch": 35.84,
"learning_rate": 5.8e-05,
"loss": 1.5304,
"step": 119500
},
{
"epoch": 35.99,
"learning_rate": 5.7777777777777776e-05,
"loss": 1.5312,
"step": 120000
},
{
"epoch": 36.14,
"learning_rate": 5.755555555555556e-05,
"loss": 1.5294,
"step": 120500
},
{
"epoch": 36.29,
"learning_rate": 5.7333333333333336e-05,
"loss": 1.5301,
"step": 121000
},
{
"epoch": 36.44,
"learning_rate": 5.711111111111112e-05,
"loss": 1.5298,
"step": 121500
},
{
"epoch": 36.59,
"learning_rate": 5.6888888888888895e-05,
"loss": 1.5284,
"step": 122000
},
{
"epoch": 36.74,
"learning_rate": 5.666666666666667e-05,
"loss": 1.5303,
"step": 122500
},
{
"epoch": 36.89,
"learning_rate": 5.644444444444445e-05,
"loss": 1.5285,
"step": 123000
},
{
"epoch": 37.04,
"learning_rate": 5.622222222222222e-05,
"loss": 1.527,
"step": 123500
},
{
"epoch": 37.19,
"learning_rate": 5.6000000000000006e-05,
"loss": 1.524,
"step": 124000
},
{
"epoch": 37.34,
"learning_rate": 5.577777777777778e-05,
"loss": 1.5259,
"step": 124500
},
{
"epoch": 37.49,
"learning_rate": 5.555555555555556e-05,
"loss": 1.5258,
"step": 125000
},
{
"epoch": 37.49,
"eval_loss": 1.4563013315200806,
"eval_runtime": 356.8442,
"eval_samples_per_second": 477.486,
"eval_steps_per_second": 3.733,
"step": 125000
},
{
"epoch": 37.64,
"learning_rate": 5.5333333333333334e-05,
"loss": 1.5242,
"step": 125500
},
{
"epoch": 37.79,
"learning_rate": 5.511111111111111e-05,
"loss": 1.5262,
"step": 126000
},
{
"epoch": 37.94,
"learning_rate": 5.488888888888889e-05,
"loss": 1.5257,
"step": 126500
},
{
"epoch": 38.09,
"learning_rate": 5.466666666666666e-05,
"loss": 1.5231,
"step": 127000
},
{
"epoch": 38.24,
"learning_rate": 5.4444444444444446e-05,
"loss": 1.5204,
"step": 127500
},
{
"epoch": 38.39,
"learning_rate": 5.422222222222223e-05,
"loss": 1.5227,
"step": 128000
},
{
"epoch": 38.54,
"learning_rate": 5.4000000000000005e-05,
"loss": 1.5226,
"step": 128500
},
{
"epoch": 38.69,
"learning_rate": 5.377777777777778e-05,
"loss": 1.5212,
"step": 129000
},
{
"epoch": 38.84,
"learning_rate": 5.355555555555556e-05,
"loss": 1.525,
"step": 129500
},
{
"epoch": 38.99,
"learning_rate": 5.333333333333333e-05,
"loss": 1.5222,
"step": 130000
},
{
"epoch": 39.14,
"learning_rate": 5.311111111111111e-05,
"loss": 1.5176,
"step": 130500
},
{
"epoch": 39.29,
"learning_rate": 5.2888888888888885e-05,
"loss": 1.5166,
"step": 131000
},
{
"epoch": 39.44,
"learning_rate": 5.266666666666666e-05,
"loss": 1.5186,
"step": 131500
},
{
"epoch": 39.59,
"learning_rate": 5.244444444444445e-05,
"loss": 1.5176,
"step": 132000
},
{
"epoch": 39.74,
"learning_rate": 5.222222222222223e-05,
"loss": 1.5199,
"step": 132500
},
{
"epoch": 39.89,
"learning_rate": 5.2000000000000004e-05,
"loss": 1.5188,
"step": 133000
},
{
"epoch": 40.04,
"learning_rate": 5.177777777777778e-05,
"loss": 1.5177,
"step": 133500
},
{
"epoch": 40.19,
"learning_rate": 5.1555555555555556e-05,
"loss": 1.5147,
"step": 134000
},
{
"epoch": 40.34,
"learning_rate": 5.133333333333333e-05,
"loss": 1.5163,
"step": 134500
},
{
"epoch": 40.49,
"learning_rate": 5.111111111111111e-05,
"loss": 1.5149,
"step": 135000
},
{
"epoch": 40.64,
"learning_rate": 5.0888888888888884e-05,
"loss": 1.5142,
"step": 135500
},
{
"epoch": 40.79,
"learning_rate": 5.0666666666666674e-05,
"loss": 1.5133,
"step": 136000
},
{
"epoch": 40.94,
"learning_rate": 5.044444444444445e-05,
"loss": 1.5135,
"step": 136500
},
{
"epoch": 41.09,
"learning_rate": 5.0222222222222226e-05,
"loss": 1.5129,
"step": 137000
},
{
"epoch": 41.24,
"learning_rate": 5e-05,
"loss": 1.5114,
"step": 137500
},
{
"epoch": 41.24,
"eval_loss": 1.4439390897750854,
"eval_runtime": 356.5866,
"eval_samples_per_second": 477.831,
"eval_steps_per_second": 3.735,
"step": 137500
},
{
"epoch": 41.39,
"learning_rate": 4.977777777777778e-05,
"loss": 1.5106,
"step": 138000
},
{
"epoch": 41.54,
"learning_rate": 4.955555555555556e-05,
"loss": 1.5115,
"step": 138500
},
{
"epoch": 41.69,
"learning_rate": 4.933333333333334e-05,
"loss": 1.5114,
"step": 139000
},
{
"epoch": 41.84,
"learning_rate": 4.9111111111111114e-05,
"loss": 1.5111,
"step": 139500
},
{
"epoch": 41.99,
"learning_rate": 4.888888888888889e-05,
"loss": 1.5118,
"step": 140000
},
{
"epoch": 42.14,
"learning_rate": 4.866666666666667e-05,
"loss": 1.5073,
"step": 140500
},
{
"epoch": 42.29,
"learning_rate": 4.844444444444445e-05,
"loss": 1.5102,
"step": 141000
},
{
"epoch": 42.44,
"learning_rate": 4.8222222222222225e-05,
"loss": 1.508,
"step": 141500
},
{
"epoch": 42.59,
"learning_rate": 4.8e-05,
"loss": 1.5081,
"step": 142000
},
{
"epoch": 42.74,
"learning_rate": 4.7777777777777784e-05,
"loss": 1.5086,
"step": 142500
},
{
"epoch": 42.89,
"learning_rate": 4.755555555555556e-05,
"loss": 1.5093,
"step": 143000
},
{
"epoch": 43.04,
"learning_rate": 4.7333333333333336e-05,
"loss": 1.5064,
"step": 143500
},
{
"epoch": 43.19,
"learning_rate": 4.711111111111111e-05,
"loss": 1.5079,
"step": 144000
},
{
"epoch": 43.34,
"learning_rate": 4.6888888888888895e-05,
"loss": 1.5058,
"step": 144500
},
{
"epoch": 43.49,
"learning_rate": 4.666666666666667e-05,
"loss": 1.5073,
"step": 145000
},
{
"epoch": 43.64,
"learning_rate": 4.644444444444445e-05,
"loss": 1.5055,
"step": 145500
},
{
"epoch": 43.79,
"learning_rate": 4.6222222222222224e-05,
"loss": 1.5056,
"step": 146000
},
{
"epoch": 43.94,
"learning_rate": 4.600000000000001e-05,
"loss": 1.5069,
"step": 146500
},
{
"epoch": 44.09,
"learning_rate": 4.577777777777778e-05,
"loss": 1.5028,
"step": 147000
},
{
"epoch": 44.24,
"learning_rate": 4.555555555555556e-05,
"loss": 1.502,
"step": 147500
},
{
"epoch": 44.39,
"learning_rate": 4.5333333333333335e-05,
"loss": 1.5019,
"step": 148000
},
{
"epoch": 44.54,
"learning_rate": 4.511111111111112e-05,
"loss": 1.5024,
"step": 148500
},
{
"epoch": 44.69,
"learning_rate": 4.4888888888888894e-05,
"loss": 1.503,
"step": 149000
},
{
"epoch": 44.84,
"learning_rate": 4.466666666666667e-05,
"loss": 1.5041,
"step": 149500
},
{
"epoch": 44.99,
"learning_rate": 4.4444444444444447e-05,
"loss": 1.5015,
"step": 150000
},
{
"epoch": 44.99,
"eval_loss": 1.436716914176941,
"eval_runtime": 356.4605,
"eval_samples_per_second": 478.0,
"eval_steps_per_second": 3.737,
"step": 150000
},
{
"epoch": 45.14,
"learning_rate": 4.422222222222222e-05,
"loss": 1.4997,
"step": 150500
},
{
"epoch": 45.29,
"learning_rate": 4.4000000000000006e-05,
"loss": 1.4993,
"step": 151000
},
{
"epoch": 45.44,
"learning_rate": 4.377777777777778e-05,
"loss": 1.4993,
"step": 151500
},
{
"epoch": 45.59,
"learning_rate": 4.355555555555556e-05,
"loss": 1.5004,
"step": 152000
},
{
"epoch": 45.74,
"learning_rate": 4.3333333333333334e-05,
"loss": 1.5004,
"step": 152500
},
{
"epoch": 45.89,
"learning_rate": 4.311111111111111e-05,
"loss": 1.4977,
"step": 153000
},
{
"epoch": 46.04,
"learning_rate": 4.2888888888888886e-05,
"loss": 1.497,
"step": 153500
},
{
"epoch": 46.19,
"learning_rate": 4.266666666666667e-05,
"loss": 1.4996,
"step": 154000
},
{
"epoch": 46.34,
"learning_rate": 4.2444444444444445e-05,
"loss": 1.4968,
"step": 154500
},
{
"epoch": 46.49,
"learning_rate": 4.222222222222222e-05,
"loss": 1.4967,
"step": 155000
},
{
"epoch": 46.64,
"learning_rate": 4.2e-05,
"loss": 1.4952,
"step": 155500
},
{
"epoch": 46.79,
"learning_rate": 4.177777777777778e-05,
"loss": 1.4965,
"step": 156000
},
{
"epoch": 46.94,
"learning_rate": 4.155555555555556e-05,
"loss": 1.4986,
"step": 156500
},
{
"epoch": 47.09,
"learning_rate": 4.133333333333333e-05,
"loss": 1.495,
"step": 157000
},
{
"epoch": 47.24,
"learning_rate": 4.111111111111111e-05,
"loss": 1.4936,
"step": 157500
},
{
"epoch": 47.39,
"learning_rate": 4.088888888888889e-05,
"loss": 1.4948,
"step": 158000
},
{
"epoch": 47.54,
"learning_rate": 4.066666666666667e-05,
"loss": 1.4933,
"step": 158500
},
{
"epoch": 47.69,
"learning_rate": 4.0444444444444444e-05,
"loss": 1.493,
"step": 159000
},
{
"epoch": 47.84,
"learning_rate": 4.022222222222222e-05,
"loss": 1.4944,
"step": 159500
},
{
"epoch": 47.99,
"learning_rate": 4e-05,
"loss": 1.4968,
"step": 160000
},
{
"epoch": 48.14,
"learning_rate": 3.977777777777778e-05,
"loss": 1.491,
"step": 160500
},
{
"epoch": 48.29,
"learning_rate": 3.9555555555555556e-05,
"loss": 1.4908,
"step": 161000
},
{
"epoch": 48.44,
"learning_rate": 3.933333333333333e-05,
"loss": 1.4916,
"step": 161500
},
{
"epoch": 48.59,
"learning_rate": 3.9111111111111115e-05,
"loss": 1.4937,
"step": 162000
},
{
"epoch": 48.74,
"learning_rate": 3.888888888888889e-05,
"loss": 1.4923,
"step": 162500
},
{
"epoch": 48.74,
"eval_loss": 1.4284732341766357,
"eval_runtime": 356.8446,
"eval_samples_per_second": 477.485,
"eval_steps_per_second": 3.733,
"step": 162500
},
{
"epoch": 48.89,
"learning_rate": 3.866666666666667e-05,
"loss": 1.4927,
"step": 163000
},
{
"epoch": 49.04,
"learning_rate": 3.844444444444444e-05,
"loss": 1.4895,
"step": 163500
},
{
"epoch": 49.19,
"learning_rate": 3.8222222222222226e-05,
"loss": 1.4884,
"step": 164000
},
{
"epoch": 49.34,
"learning_rate": 3.8e-05,
"loss": 1.4897,
"step": 164500
},
{
"epoch": 49.49,
"learning_rate": 3.777777777777778e-05,
"loss": 1.4909,
"step": 165000
},
{
"epoch": 49.64,
"learning_rate": 3.7555555555555554e-05,
"loss": 1.4906,
"step": 165500
},
{
"epoch": 49.79,
"learning_rate": 3.733333333333334e-05,
"loss": 1.4859,
"step": 166000
},
{
"epoch": 49.94,
"learning_rate": 3.7111111111111113e-05,
"loss": 1.4898,
"step": 166500
},
{
"epoch": 50.09,
"learning_rate": 3.688888888888889e-05,
"loss": 1.4881,
"step": 167000
},
{
"epoch": 50.24,
"learning_rate": 3.6666666666666666e-05,
"loss": 1.4886,
"step": 167500
},
{
"epoch": 50.39,
"learning_rate": 3.644444444444445e-05,
"loss": 1.4875,
"step": 168000
},
{
"epoch": 50.54,
"learning_rate": 3.6222222222222225e-05,
"loss": 1.4879,
"step": 168500
},
{
"epoch": 50.69,
"learning_rate": 3.6e-05,
"loss": 1.4867,
"step": 169000
},
{
"epoch": 50.84,
"learning_rate": 3.577777777777778e-05,
"loss": 1.4866,
"step": 169500
},
{
"epoch": 50.99,
"learning_rate": 3.555555555555556e-05,
"loss": 1.4862,
"step": 170000
},
{
"epoch": 51.14,
"learning_rate": 3.5333333333333336e-05,
"loss": 1.4839,
"step": 170500
},
{
"epoch": 51.29,
"learning_rate": 3.511111111111111e-05,
"loss": 1.4848,
"step": 171000
},
{
"epoch": 51.44,
"learning_rate": 3.4888888888888895e-05,
"loss": 1.4861,
"step": 171500
},
{
"epoch": 51.59,
"learning_rate": 3.466666666666667e-05,
"loss": 1.4846,
"step": 172000
},
{
"epoch": 51.74,
"learning_rate": 3.444444444444445e-05,
"loss": 1.4853,
"step": 172500
},
{
"epoch": 51.89,
"learning_rate": 3.4222222222222224e-05,
"loss": 1.4852,
"step": 173000
},
{
"epoch": 52.04,
"learning_rate": 3.4000000000000007e-05,
"loss": 1.4826,
"step": 173500
},
{
"epoch": 52.19,
"learning_rate": 3.377777777777778e-05,
"loss": 1.4829,
"step": 174000
},
{
"epoch": 52.34,
"learning_rate": 3.355555555555556e-05,
"loss": 1.4815,
"step": 174500
},
{
"epoch": 52.49,
"learning_rate": 3.3333333333333335e-05,
"loss": 1.4809,
"step": 175000
},
{
"epoch": 52.49,
"eval_loss": 1.4197227954864502,
"eval_runtime": 356.5538,
"eval_samples_per_second": 477.875,
"eval_steps_per_second": 3.736,
"step": 175000
},
{
"epoch": 52.64,
"learning_rate": 3.311111111111112e-05,
"loss": 1.4826,
"step": 175500
},
{
"epoch": 52.79,
"learning_rate": 3.2888888888888894e-05,
"loss": 1.4834,
"step": 176000
},
{
"epoch": 52.94,
"learning_rate": 3.266666666666667e-05,
"loss": 1.4816,
"step": 176500
},
{
"epoch": 53.09,
"learning_rate": 3.2444444444444446e-05,
"loss": 1.4801,
"step": 177000
},
{
"epoch": 53.24,
"learning_rate": 3.222222222222223e-05,
"loss": 1.4807,
"step": 177500
},
{
"epoch": 53.39,
"learning_rate": 3.2000000000000005e-05,
"loss": 1.4798,
"step": 178000
},
{
"epoch": 53.54,
"learning_rate": 3.177777777777778e-05,
"loss": 1.4792,
"step": 178500
},
{
"epoch": 53.69,
"learning_rate": 3.155555555555556e-05,
"loss": 1.4795,
"step": 179000
},
{
"epoch": 53.84,
"learning_rate": 3.1333333333333334e-05,
"loss": 1.4796,
"step": 179500
},
{
"epoch": 53.99,
"learning_rate": 3.111111111111111e-05,
"loss": 1.4815,
"step": 180000
},
{
"epoch": 54.14,
"learning_rate": 3.088888888888889e-05,
"loss": 1.4782,
"step": 180500
},
{
"epoch": 54.29,
"learning_rate": 3.066666666666667e-05,
"loss": 1.4784,
"step": 181000
},
{
"epoch": 54.44,
"learning_rate": 3.044444444444445e-05,
"loss": 1.478,
"step": 181500
},
{
"epoch": 54.59,
"learning_rate": 3.0222222222222225e-05,
"loss": 1.4767,
"step": 182000
},
{
"epoch": 54.74,
"learning_rate": 3e-05,
"loss": 1.4771,
"step": 182500
},
{
"epoch": 54.89,
"learning_rate": 2.9777777777777777e-05,
"loss": 1.4781,
"step": 183000
},
{
"epoch": 55.04,
"learning_rate": 2.955555555555556e-05,
"loss": 1.477,
"step": 183500
},
{
"epoch": 55.19,
"learning_rate": 2.9333333333333336e-05,
"loss": 1.4764,
"step": 184000
},
{
"epoch": 55.34,
"learning_rate": 2.9111111111111112e-05,
"loss": 1.4765,
"step": 184500
},
{
"epoch": 55.49,
"learning_rate": 2.8888888888888888e-05,
"loss": 1.4765,
"step": 185000
},
{
"epoch": 55.64,
"learning_rate": 2.8666666666666668e-05,
"loss": 1.4749,
"step": 185500
},
{
"epoch": 55.79,
"learning_rate": 2.8444444444444447e-05,
"loss": 1.4762,
"step": 186000
},
{
"epoch": 55.94,
"learning_rate": 2.8222222222222223e-05,
"loss": 1.4754,
"step": 186500
},
{
"epoch": 56.09,
"learning_rate": 2.8000000000000003e-05,
"loss": 1.4719,
"step": 187000
},
{
"epoch": 56.24,
"learning_rate": 2.777777777777778e-05,
"loss": 1.473,
"step": 187500
},
{
"epoch": 56.24,
"eval_loss": 1.4130820035934448,
"eval_runtime": 356.901,
"eval_samples_per_second": 477.41,
"eval_steps_per_second": 3.732,
"step": 187500
},
{
"epoch": 56.39,
"learning_rate": 2.7555555555555555e-05,
"loss": 1.4724,
"step": 188000
},
{
"epoch": 56.54,
"learning_rate": 2.733333333333333e-05,
"loss": 1.4697,
"step": 188500
},
{
"epoch": 56.69,
"learning_rate": 2.7111111111111114e-05,
"loss": 1.4728,
"step": 189000
},
{
"epoch": 56.84,
"learning_rate": 2.688888888888889e-05,
"loss": 1.4738,
"step": 189500
},
{
"epoch": 56.99,
"learning_rate": 2.6666666666666667e-05,
"loss": 1.4726,
"step": 190000
},
{
"epoch": 57.14,
"learning_rate": 2.6444444444444443e-05,
"loss": 1.4712,
"step": 190500
},
{
"epoch": 57.29,
"learning_rate": 2.6222222222222226e-05,
"loss": 1.47,
"step": 191000
},
{
"epoch": 57.44,
"learning_rate": 2.6000000000000002e-05,
"loss": 1.4709,
"step": 191500
},
{
"epoch": 57.59,
"learning_rate": 2.5777777777777778e-05,
"loss": 1.469,
"step": 192000
},
{
"epoch": 57.74,
"learning_rate": 2.5555555555555554e-05,
"loss": 1.472,
"step": 192500
},
{
"epoch": 57.89,
"learning_rate": 2.5333333333333337e-05,
"loss": 1.473,
"step": 193000
},
{
"epoch": 58.04,
"learning_rate": 2.5111111111111113e-05,
"loss": 1.4681,
"step": 193500
},
{
"epoch": 58.19,
"learning_rate": 2.488888888888889e-05,
"loss": 1.4679,
"step": 194000
},
{
"epoch": 58.34,
"learning_rate": 2.466666666666667e-05,
"loss": 1.47,
"step": 194500
},
{
"epoch": 58.49,
"learning_rate": 2.4444444444444445e-05,
"loss": 1.4668,
"step": 195000
},
{
"epoch": 58.64,
"learning_rate": 2.4222222222222224e-05,
"loss": 1.4678,
"step": 195500
},
{
"epoch": 58.79,
"learning_rate": 2.4e-05,
"loss": 1.4681,
"step": 196000
},
{
"epoch": 58.94,
"learning_rate": 2.377777777777778e-05,
"loss": 1.4694,
"step": 196500
},
{
"epoch": 59.09,
"learning_rate": 2.3555555555555556e-05,
"loss": 1.4651,
"step": 197000
},
{
"epoch": 59.24,
"learning_rate": 2.3333333333333336e-05,
"loss": 1.4694,
"step": 197500
},
{
"epoch": 59.39,
"learning_rate": 2.3111111111111112e-05,
"loss": 1.4692,
"step": 198000
},
{
"epoch": 59.54,
"learning_rate": 2.288888888888889e-05,
"loss": 1.4675,
"step": 198500
},
{
"epoch": 59.69,
"learning_rate": 2.2666666666666668e-05,
"loss": 1.4653,
"step": 199000
},
{
"epoch": 59.84,
"learning_rate": 2.2444444444444447e-05,
"loss": 1.465,
"step": 199500
},
{
"epoch": 59.99,
"learning_rate": 2.2222222222222223e-05,
"loss": 1.4695,
"step": 200000
},
{
"epoch": 59.99,
"eval_loss": 1.4062610864639282,
"eval_runtime": 356.555,
"eval_samples_per_second": 477.873,
"eval_steps_per_second": 3.736,
"step": 200000
},
{
"epoch": 60.14,
"learning_rate": 2.2000000000000003e-05,
"loss": 1.4618,
"step": 200500
},
{
"epoch": 60.29,
"learning_rate": 2.177777777777778e-05,
"loss": 1.467,
"step": 201000
},
{
"epoch": 60.44,
"learning_rate": 2.1555555555555555e-05,
"loss": 1.467,
"step": 201500
},
{
"epoch": 60.59,
"learning_rate": 2.1333333333333335e-05,
"loss": 1.4646,
"step": 202000
},
{
"epoch": 60.74,
"learning_rate": 2.111111111111111e-05,
"loss": 1.464,
"step": 202500
},
{
"epoch": 60.89,
"learning_rate": 2.088888888888889e-05,
"loss": 1.4644,
"step": 203000
},
{
"epoch": 61.04,
"learning_rate": 2.0666666666666666e-05,
"loss": 1.4632,
"step": 203500
},
{
"epoch": 61.19,
"learning_rate": 2.0444444444444446e-05,
"loss": 1.4625,
"step": 204000
},
{
"epoch": 61.34,
"learning_rate": 2.0222222222222222e-05,
"loss": 1.4625,
"step": 204500
},
{
"epoch": 61.49,
"learning_rate": 2e-05,
"loss": 1.4634,
"step": 205000
},
{
"epoch": 61.64,
"learning_rate": 1.9777777777777778e-05,
"loss": 1.4651,
"step": 205500
},
{
"epoch": 61.79,
"learning_rate": 1.9555555555555557e-05,
"loss": 1.4617,
"step": 206000
},
{
"epoch": 61.94,
"learning_rate": 1.9333333333333333e-05,
"loss": 1.4633,
"step": 206500
},
{
"epoch": 62.09,
"learning_rate": 1.9111111111111113e-05,
"loss": 1.4604,
"step": 207000
},
{
"epoch": 62.24,
"learning_rate": 1.888888888888889e-05,
"loss": 1.4589,
"step": 207500
},
{
"epoch": 62.39,
"learning_rate": 1.866666666666667e-05,
"loss": 1.4619,
"step": 208000
},
{
"epoch": 62.54,
"learning_rate": 1.8444444444444445e-05,
"loss": 1.4625,
"step": 208500
},
{
"epoch": 62.69,
"learning_rate": 1.8222222222222224e-05,
"loss": 1.4613,
"step": 209000
},
{
"epoch": 62.84,
"learning_rate": 1.8e-05,
"loss": 1.4587,
"step": 209500
},
{
"epoch": 62.99,
"learning_rate": 1.777777777777778e-05,
"loss": 1.459,
"step": 210000
},
{
"epoch": 63.14,
"learning_rate": 1.7555555555555556e-05,
"loss": 1.4563,
"step": 210500
},
{
"epoch": 63.29,
"learning_rate": 1.7333333333333336e-05,
"loss": 1.4609,
"step": 211000
},
{
"epoch": 63.44,
"learning_rate": 1.7111111111111112e-05,
"loss": 1.4587,
"step": 211500
},
{
"epoch": 63.59,
"learning_rate": 1.688888888888889e-05,
"loss": 1.46,
"step": 212000
},
{
"epoch": 63.74,
"learning_rate": 1.6666666666666667e-05,
"loss": 1.46,
"step": 212500
},
{
"epoch": 63.74,
"eval_loss": 1.3980076313018799,
"eval_runtime": 356.9089,
"eval_samples_per_second": 477.399,
"eval_steps_per_second": 3.732,
"step": 212500
},
{
"epoch": 63.89,
"learning_rate": 1.6444444444444447e-05,
"loss": 1.4591,
"step": 213000
},
{
"epoch": 64.04,
"learning_rate": 1.6222222222222223e-05,
"loss": 1.4595,
"step": 213500
},
{
"epoch": 64.19,
"learning_rate": 1.6000000000000003e-05,
"loss": 1.4565,
"step": 214000
},
{
"epoch": 64.34,
"learning_rate": 1.577777777777778e-05,
"loss": 1.4577,
"step": 214500
},
{
"epoch": 64.49,
"learning_rate": 1.5555555555555555e-05,
"loss": 1.4576,
"step": 215000
},
{
"epoch": 64.64,
"learning_rate": 1.5333333333333334e-05,
"loss": 1.4565,
"step": 215500
},
{
"epoch": 64.79,
"learning_rate": 1.5111111111111112e-05,
"loss": 1.4582,
"step": 216000
},
{
"epoch": 64.94,
"learning_rate": 1.4888888888888888e-05,
"loss": 1.4564,
"step": 216500
},
{
"epoch": 65.09,
"learning_rate": 1.4666666666666668e-05,
"loss": 1.4555,
"step": 217000
},
{
"epoch": 65.24,
"learning_rate": 1.4444444444444444e-05,
"loss": 1.457,
"step": 217500
},
{
"epoch": 65.39,
"learning_rate": 1.4222222222222224e-05,
"loss": 1.4554,
"step": 218000
},
{
"epoch": 65.54,
"learning_rate": 1.4000000000000001e-05,
"loss": 1.4575,
"step": 218500
},
{
"epoch": 65.69,
"learning_rate": 1.3777777777777778e-05,
"loss": 1.4558,
"step": 219000
},
{
"epoch": 65.84,
"learning_rate": 1.3555555555555557e-05,
"loss": 1.4566,
"step": 219500
},
{
"epoch": 65.99,
"learning_rate": 1.3333333333333333e-05,
"loss": 1.4551,
"step": 220000
},
{
"epoch": 66.14,
"learning_rate": 1.3111111111111113e-05,
"loss": 1.4547,
"step": 220500
},
{
"epoch": 66.29,
"learning_rate": 1.2888888888888889e-05,
"loss": 1.4553,
"step": 221000
},
{
"epoch": 66.44,
"learning_rate": 1.2666666666666668e-05,
"loss": 1.4525,
"step": 221500
},
{
"epoch": 66.59,
"learning_rate": 1.2444444444444445e-05,
"loss": 1.4544,
"step": 222000
},
{
"epoch": 66.74,
"learning_rate": 1.2222222222222222e-05,
"loss": 1.4528,
"step": 222500
},
{
"epoch": 66.89,
"learning_rate": 1.2e-05,
"loss": 1.4531,
"step": 223000
},
{
"epoch": 67.04,
"learning_rate": 1.1777777777777778e-05,
"loss": 1.4533,
"step": 223500
},
{
"epoch": 67.19,
"learning_rate": 1.1555555555555556e-05,
"loss": 1.452,
"step": 224000
},
{
"epoch": 67.34,
"learning_rate": 1.1333333333333334e-05,
"loss": 1.4526,
"step": 224500
},
{
"epoch": 67.49,
"learning_rate": 1.1111111111111112e-05,
"loss": 1.451,
"step": 225000
},
{
"epoch": 67.49,
"eval_loss": 1.3923670053482056,
"eval_runtime": 356.8892,
"eval_samples_per_second": 477.426,
"eval_steps_per_second": 3.732,
"step": 225000
},
{
"epoch": 67.64,
"learning_rate": 1.088888888888889e-05,
"loss": 1.4535,
"step": 225500
},
{
"epoch": 67.79,
"learning_rate": 1.0666666666666667e-05,
"loss": 1.4527,
"step": 226000
},
{
"epoch": 67.94,
"learning_rate": 1.0444444444444445e-05,
"loss": 1.4516,
"step": 226500
},
{
"epoch": 68.09,
"learning_rate": 1.0222222222222223e-05,
"loss": 1.4526,
"step": 227000
},
{
"epoch": 68.24,
"learning_rate": 1e-05,
"loss": 1.4517,
"step": 227500
},
{
"epoch": 68.39,
"learning_rate": 9.777777777777779e-06,
"loss": 1.4511,
"step": 228000
},
{
"epoch": 68.54,
"learning_rate": 9.555555555555556e-06,
"loss": 1.4507,
"step": 228500
},
{
"epoch": 68.69,
"learning_rate": 9.333333333333334e-06,
"loss": 1.4486,
"step": 229000
},
{
"epoch": 68.84,
"learning_rate": 9.111111111111112e-06,
"loss": 1.4531,
"step": 229500
},
{
"epoch": 68.99,
"learning_rate": 8.88888888888889e-06,
"loss": 1.4512,
"step": 230000
},
{
"epoch": 69.14,
"learning_rate": 8.666666666666668e-06,
"loss": 1.4494,
"step": 230500
},
{
"epoch": 69.29,
"learning_rate": 8.444444444444446e-06,
"loss": 1.4524,
"step": 231000
},
{
"epoch": 69.44,
"learning_rate": 8.222222222222223e-06,
"loss": 1.4509,
"step": 231500
},
{
"epoch": 69.59,
"learning_rate": 8.000000000000001e-06,
"loss": 1.45,
"step": 232000
},
{
"epoch": 69.74,
"learning_rate": 7.777777777777777e-06,
"loss": 1.4468,
"step": 232500
},
{
"epoch": 69.89,
"learning_rate": 7.555555555555556e-06,
"loss": 1.4463,
"step": 233000
},
{
"epoch": 70.04,
"learning_rate": 7.333333333333334e-06,
"loss": 1.4479,
"step": 233500
},
{
"epoch": 70.19,
"learning_rate": 7.111111111111112e-06,
"loss": 1.45,
"step": 234000
},
{
"epoch": 70.34,
"learning_rate": 6.888888888888889e-06,
"loss": 1.4478,
"step": 234500
},
{
"epoch": 70.49,
"learning_rate": 6.666666666666667e-06,
"loss": 1.448,
"step": 235000
},
{
"epoch": 70.64,
"learning_rate": 6.4444444444444445e-06,
"loss": 1.4477,
"step": 235500
},
{
"epoch": 70.79,
"learning_rate": 6.222222222222222e-06,
"loss": 1.4467,
"step": 236000
},
{
"epoch": 70.94,
"learning_rate": 6e-06,
"loss": 1.447,
"step": 236500
},
{
"epoch": 71.09,
"learning_rate": 5.777777777777778e-06,
"loss": 1.4475,
"step": 237000
},
{
"epoch": 71.24,
"learning_rate": 5.555555555555556e-06,
"loss": 1.4435,
"step": 237500
},
{
"epoch": 71.24,
"eval_loss": 1.3885681629180908,
"eval_runtime": 356.9911,
"eval_samples_per_second": 477.289,
"eval_steps_per_second": 3.731,
"step": 237500
},
{
"epoch": 71.39,
"learning_rate": 5.333333333333334e-06,
"loss": 1.4468,
"step": 238000
},
{
"epoch": 71.54,
"learning_rate": 5.1111111111111115e-06,
"loss": 1.446,
"step": 238500
},
{
"epoch": 71.69,
"learning_rate": 4.888888888888889e-06,
"loss": 1.4476,
"step": 239000
},
{
"epoch": 71.84,
"learning_rate": 4.666666666666667e-06,
"loss": 1.4479,
"step": 239500
},
{
"epoch": 71.99,
"learning_rate": 4.444444444444445e-06,
"loss": 1.4454,
"step": 240000
},
{
"epoch": 72.14,
"learning_rate": 4.222222222222223e-06,
"loss": 1.4464,
"step": 240500
},
{
"epoch": 72.29,
"learning_rate": 4.000000000000001e-06,
"loss": 1.445,
"step": 241000
},
{
"epoch": 72.44,
"learning_rate": 3.777777777777778e-06,
"loss": 1.4462,
"step": 241500
},
{
"epoch": 72.59,
"learning_rate": 3.555555555555556e-06,
"loss": 1.4457,
"step": 242000
},
{
"epoch": 72.74,
"learning_rate": 3.3333333333333333e-06,
"loss": 1.4465,
"step": 242500
},
{
"epoch": 72.89,
"learning_rate": 3.111111111111111e-06,
"loss": 1.4463,
"step": 243000
},
{
"epoch": 73.04,
"learning_rate": 2.888888888888889e-06,
"loss": 1.4453,
"step": 243500
},
{
"epoch": 73.19,
"learning_rate": 2.666666666666667e-06,
"loss": 1.444,
"step": 244000
},
{
"epoch": 73.34,
"learning_rate": 2.4444444444444447e-06,
"loss": 1.4426,
"step": 244500
},
{
"epoch": 73.49,
"learning_rate": 2.2222222222222225e-06,
"loss": 1.4431,
"step": 245000
},
{
"epoch": 73.64,
"learning_rate": 2.0000000000000003e-06,
"loss": 1.4438,
"step": 245500
},
{
"epoch": 73.79,
"learning_rate": 1.777777777777778e-06,
"loss": 1.4437,
"step": 246000
},
{
"epoch": 73.94,
"learning_rate": 1.5555555555555556e-06,
"loss": 1.444,
"step": 246500
},
{
"epoch": 74.09,
"learning_rate": 1.3333333333333334e-06,
"loss": 1.444,
"step": 247000
},
{
"epoch": 74.24,
"learning_rate": 1.1111111111111112e-06,
"loss": 1.4457,
"step": 247500
},
{
"epoch": 74.39,
"learning_rate": 8.88888888888889e-07,
"loss": 1.4434,
"step": 248000
},
{
"epoch": 74.54,
"learning_rate": 6.666666666666667e-07,
"loss": 1.4429,
"step": 248500
},
{
"epoch": 74.69,
"learning_rate": 4.444444444444445e-07,
"loss": 1.4426,
"step": 249000
},
{
"epoch": 74.84,
"learning_rate": 2.2222222222222224e-07,
"loss": 1.4413,
"step": 249500
},
{
"epoch": 74.99,
"learning_rate": 0.0,
"loss": 1.4436,
"step": 250000
},
{
"epoch": 74.99,
"eval_loss": 1.3831435441970825,
"eval_runtime": 356.4648,
"eval_samples_per_second": 477.994,
"eval_steps_per_second": 3.737,
"step": 250000
}
],
"max_steps": 250000,
"num_train_epochs": 75,
"total_flos": 4.212247412736e+18,
"trial_name": null,
"trial_params": null
}