gpt2-wechsel-french / trainer_state.json
benjamin's picture
initial commit
2c0ac76
raw
history blame
92.7 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 69.61844761573268,
"global_step": 250000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.14,
"learning_rate": 1e-05,
"learning_rate_embeddings": 1e-05,
"loss": 6.6253,
"step": 500
},
{
"epoch": 0.28,
"learning_rate": 2e-05,
"learning_rate_embeddings": 2e-05,
"loss": 5.3003,
"step": 1000
},
{
"epoch": 0.42,
"learning_rate": 3e-05,
"learning_rate_embeddings": 3e-05,
"loss": 4.8291,
"step": 1500
},
{
"epoch": 0.56,
"learning_rate": 4e-05,
"learning_rate_embeddings": 4e-05,
"loss": 4.5601,
"step": 2000
},
{
"epoch": 0.7,
"learning_rate": 5e-05,
"learning_rate_embeddings": 5e-05,
"loss": 4.3749,
"step": 2500
},
{
"epoch": 0.84,
"learning_rate": 6e-05,
"learning_rate_embeddings": 6e-05,
"loss": 4.2365,
"step": 3000
},
{
"epoch": 0.97,
"learning_rate": 7.000000000000001e-05,
"learning_rate_embeddings": 7.000000000000001e-05,
"loss": 4.114,
"step": 3500
},
{
"epoch": 1.11,
"learning_rate": 8e-05,
"learning_rate_embeddings": 8e-05,
"loss": 4.0063,
"step": 4000
},
{
"epoch": 1.25,
"learning_rate": 8.999999999999999e-05,
"learning_rate_embeddings": 8.999999999999999e-05,
"loss": 3.9135,
"step": 4500
},
{
"epoch": 1.39,
"learning_rate": 0.0001,
"learning_rate_embeddings": 0.0001,
"loss": 3.8331,
"step": 5000
},
{
"epoch": 1.53,
"learning_rate": 0.00011,
"learning_rate_embeddings": 0.00011,
"loss": 3.7766,
"step": 5500
},
{
"epoch": 1.67,
"learning_rate": 0.00012,
"learning_rate_embeddings": 0.00012,
"loss": 3.7228,
"step": 6000
},
{
"epoch": 1.81,
"learning_rate": 0.00013000000000000002,
"learning_rate_embeddings": 0.00013000000000000002,
"loss": 3.6715,
"step": 6500
},
{
"epoch": 1.95,
"learning_rate": 0.00014000000000000001,
"learning_rate_embeddings": 0.00014000000000000001,
"loss": 3.634,
"step": 7000
},
{
"epoch": 2.09,
"learning_rate": 0.00015,
"learning_rate_embeddings": 0.00015,
"loss": 3.5923,
"step": 7500
},
{
"epoch": 2.23,
"learning_rate": 0.00016,
"learning_rate_embeddings": 0.00016,
"loss": 3.5615,
"step": 8000
},
{
"epoch": 2.37,
"learning_rate": 0.00017,
"learning_rate_embeddings": 0.00017,
"loss": 3.5314,
"step": 8500
},
{
"epoch": 2.51,
"learning_rate": 0.00017999999999999998,
"learning_rate_embeddings": 0.00017999999999999998,
"loss": 3.506,
"step": 9000
},
{
"epoch": 2.65,
"learning_rate": 0.00019,
"learning_rate_embeddings": 0.00019,
"loss": 3.4803,
"step": 9500
},
{
"epoch": 2.78,
"learning_rate": 0.0002,
"learning_rate_embeddings": 0.0002,
"loss": 3.4648,
"step": 10000
},
{
"epoch": 2.92,
"learning_rate": 0.00021,
"learning_rate_embeddings": 0.00021,
"loss": 3.4497,
"step": 10500
},
{
"epoch": 3.06,
"learning_rate": 0.00022,
"learning_rate_embeddings": 0.00022,
"loss": 3.4254,
"step": 11000
},
{
"epoch": 3.2,
"learning_rate": 0.00023,
"learning_rate_embeddings": 0.00023,
"loss": 3.3993,
"step": 11500
},
{
"epoch": 3.34,
"learning_rate": 0.00024,
"learning_rate_embeddings": 0.00024,
"loss": 3.3893,
"step": 12000
},
{
"epoch": 3.48,
"learning_rate": 0.00025,
"learning_rate_embeddings": 0.00025,
"loss": 3.3673,
"step": 12500
},
{
"epoch": 3.48,
"eval_loss": 3.3232665061950684,
"eval_runtime": 384.8896,
"eval_samples_per_second": 479.493,
"eval_steps_per_second": 3.747,
"step": 12500
},
{
"epoch": 3.62,
"learning_rate": 0.00026000000000000003,
"learning_rate_embeddings": 0.00026000000000000003,
"loss": 3.3559,
"step": 13000
},
{
"epoch": 3.76,
"learning_rate": 0.00027,
"learning_rate_embeddings": 0.00027,
"loss": 3.3442,
"step": 13500
},
{
"epoch": 3.9,
"learning_rate": 0.00028000000000000003,
"learning_rate_embeddings": 0.00028000000000000003,
"loss": 3.3282,
"step": 14000
},
{
"epoch": 4.04,
"learning_rate": 0.00029,
"learning_rate_embeddings": 0.00029,
"loss": 3.3189,
"step": 14500
},
{
"epoch": 4.18,
"learning_rate": 0.0003,
"learning_rate_embeddings": 0.0003,
"loss": 3.2965,
"step": 15000
},
{
"epoch": 4.32,
"learning_rate": 0.00031,
"learning_rate_embeddings": 0.00031,
"loss": 3.2934,
"step": 15500
},
{
"epoch": 4.46,
"learning_rate": 0.00032,
"learning_rate_embeddings": 0.00032,
"loss": 3.281,
"step": 16000
},
{
"epoch": 4.59,
"learning_rate": 0.00033,
"learning_rate_embeddings": 0.00033,
"loss": 3.2739,
"step": 16500
},
{
"epoch": 4.73,
"learning_rate": 0.00034,
"learning_rate_embeddings": 0.00034,
"loss": 3.2687,
"step": 17000
},
{
"epoch": 4.87,
"learning_rate": 0.00035,
"learning_rate_embeddings": 0.00035,
"loss": 3.2584,
"step": 17500
},
{
"epoch": 5.01,
"learning_rate": 0.00035999999999999997,
"learning_rate_embeddings": 0.00035999999999999997,
"loss": 3.2436,
"step": 18000
},
{
"epoch": 5.15,
"learning_rate": 0.00037,
"learning_rate_embeddings": 0.00037,
"loss": 3.2294,
"step": 18500
},
{
"epoch": 5.29,
"learning_rate": 0.00038,
"learning_rate_embeddings": 0.00038,
"loss": 3.2221,
"step": 19000
},
{
"epoch": 5.43,
"learning_rate": 0.00039000000000000005,
"learning_rate_embeddings": 0.00039000000000000005,
"loss": 3.2163,
"step": 19500
},
{
"epoch": 5.57,
"learning_rate": 0.0004,
"learning_rate_embeddings": 0.0004,
"loss": 3.2084,
"step": 20000
},
{
"epoch": 5.71,
"learning_rate": 0.00041,
"learning_rate_embeddings": 0.00041,
"loss": 3.209,
"step": 20500
},
{
"epoch": 5.85,
"learning_rate": 0.00042,
"learning_rate_embeddings": 0.00042,
"loss": 3.2081,
"step": 21000
},
{
"epoch": 5.99,
"learning_rate": 0.00043,
"learning_rate_embeddings": 0.00043,
"loss": 3.1962,
"step": 21500
},
{
"epoch": 6.13,
"learning_rate": 0.00044,
"learning_rate_embeddings": 0.00044,
"loss": 3.1895,
"step": 22000
},
{
"epoch": 6.27,
"learning_rate": 0.00045000000000000004,
"learning_rate_embeddings": 0.00045000000000000004,
"loss": 3.1731,
"step": 22500
},
{
"epoch": 6.4,
"learning_rate": 0.00046,
"learning_rate_embeddings": 0.00046,
"loss": 3.1771,
"step": 23000
},
{
"epoch": 6.54,
"learning_rate": 0.00047,
"learning_rate_embeddings": 0.00047,
"loss": 3.1732,
"step": 23500
},
{
"epoch": 6.68,
"learning_rate": 0.00048,
"learning_rate_embeddings": 0.00048,
"loss": 3.1668,
"step": 24000
},
{
"epoch": 6.82,
"learning_rate": 0.00049,
"learning_rate_embeddings": 0.00049,
"loss": 3.1598,
"step": 24500
},
{
"epoch": 6.96,
"learning_rate": 0.0005,
"learning_rate_embeddings": 0.0005,
"loss": 3.1646,
"step": 25000
},
{
"epoch": 6.96,
"eval_loss": 3.1556143760681152,
"eval_runtime": 371.5552,
"eval_samples_per_second": 496.701,
"eval_steps_per_second": 3.881,
"step": 25000
},
{
"epoch": 7.1,
"learning_rate": 0.0004988888888888889,
"learning_rate_embeddings": 0.0004988888888888889,
"loss": 3.1431,
"step": 25500
},
{
"epoch": 7.24,
"learning_rate": 0.0004977777777777778,
"learning_rate_embeddings": 0.0004977777777777778,
"loss": 3.1398,
"step": 26000
},
{
"epoch": 7.38,
"learning_rate": 0.0004966666666666666,
"learning_rate_embeddings": 0.0004966666666666666,
"loss": 3.1355,
"step": 26500
},
{
"epoch": 7.52,
"learning_rate": 0.0004955555555555556,
"learning_rate_embeddings": 0.0004955555555555556,
"loss": 3.1349,
"step": 27000
},
{
"epoch": 7.66,
"learning_rate": 0.0004944444444444445,
"learning_rate_embeddings": 0.0004944444444444445,
"loss": 3.1308,
"step": 27500
},
{
"epoch": 7.8,
"learning_rate": 0.0004933333333333334,
"learning_rate_embeddings": 0.0004933333333333334,
"loss": 3.1305,
"step": 28000
},
{
"epoch": 7.94,
"learning_rate": 0.0004922222222222222,
"learning_rate_embeddings": 0.0004922222222222222,
"loss": 3.1323,
"step": 28500
},
{
"epoch": 8.08,
"learning_rate": 0.0004911111111111111,
"learning_rate_embeddings": 0.0004911111111111111,
"loss": 3.1179,
"step": 29000
},
{
"epoch": 8.21,
"learning_rate": 0.00049,
"learning_rate_embeddings": 0.00049,
"loss": 3.1083,
"step": 29500
},
{
"epoch": 8.35,
"learning_rate": 0.0004888888888888889,
"learning_rate_embeddings": 0.0004888888888888889,
"loss": 3.1094,
"step": 30000
},
{
"epoch": 8.49,
"learning_rate": 0.0004877777777777778,
"learning_rate_embeddings": 0.0004877777777777778,
"loss": 3.1131,
"step": 30500
},
{
"epoch": 8.63,
"learning_rate": 0.0004866666666666667,
"learning_rate_embeddings": 0.0004866666666666667,
"loss": 3.1127,
"step": 31000
},
{
"epoch": 8.77,
"learning_rate": 0.0004855555555555556,
"learning_rate_embeddings": 0.0004855555555555556,
"loss": 3.1104,
"step": 31500
},
{
"epoch": 8.91,
"learning_rate": 0.00048444444444444446,
"learning_rate_embeddings": 0.00048444444444444446,
"loss": 3.1043,
"step": 32000
},
{
"epoch": 9.05,
"learning_rate": 0.00048333333333333334,
"learning_rate_embeddings": 0.00048333333333333334,
"loss": 3.0959,
"step": 32500
},
{
"epoch": 9.19,
"learning_rate": 0.0004822222222222222,
"learning_rate_embeddings": 0.0004822222222222222,
"loss": 3.0801,
"step": 33000
},
{
"epoch": 9.33,
"learning_rate": 0.0004811111111111111,
"learning_rate_embeddings": 0.0004811111111111111,
"loss": 3.0935,
"step": 33500
},
{
"epoch": 9.47,
"learning_rate": 0.00048,
"learning_rate_embeddings": 0.00048,
"loss": 3.084,
"step": 34000
},
{
"epoch": 9.61,
"learning_rate": 0.0004788888888888889,
"learning_rate_embeddings": 0.0004788888888888889,
"loss": 3.0918,
"step": 34500
},
{
"epoch": 9.75,
"learning_rate": 0.0004777777777777778,
"learning_rate_embeddings": 0.0004777777777777778,
"loss": 3.0901,
"step": 35000
},
{
"epoch": 9.89,
"learning_rate": 0.0004766666666666667,
"learning_rate_embeddings": 0.0004766666666666667,
"loss": 3.084,
"step": 35500
},
{
"epoch": 10.03,
"learning_rate": 0.00047555555555555556,
"learning_rate_embeddings": 0.00047555555555555556,
"loss": 3.0878,
"step": 36000
},
{
"epoch": 10.16,
"learning_rate": 0.00047444444444444444,
"learning_rate_embeddings": 0.00047444444444444444,
"loss": 3.0629,
"step": 36500
},
{
"epoch": 10.3,
"learning_rate": 0.00047333333333333336,
"learning_rate_embeddings": 0.00047333333333333336,
"loss": 3.0741,
"step": 37000
},
{
"epoch": 10.44,
"learning_rate": 0.00047222222222222224,
"learning_rate_embeddings": 0.00047222222222222224,
"loss": 3.0791,
"step": 37500
},
{
"epoch": 10.44,
"eval_loss": 3.100255250930786,
"eval_runtime": 371.6431,
"eval_samples_per_second": 496.584,
"eval_steps_per_second": 3.88,
"step": 37500
},
{
"epoch": 10.58,
"learning_rate": 0.0004711111111111111,
"learning_rate_embeddings": 0.0004711111111111111,
"loss": 3.0781,
"step": 38000
},
{
"epoch": 10.72,
"learning_rate": 0.00047,
"learning_rate_embeddings": 0.00047,
"loss": 3.0746,
"step": 38500
},
{
"epoch": 10.86,
"learning_rate": 0.0004688888888888889,
"learning_rate_embeddings": 0.0004688888888888889,
"loss": 3.0734,
"step": 39000
},
{
"epoch": 11.0,
"learning_rate": 0.0004677777777777778,
"learning_rate_embeddings": 0.0004677777777777778,
"loss": 3.0779,
"step": 39500
},
{
"epoch": 11.14,
"learning_rate": 0.00046666666666666666,
"learning_rate_embeddings": 0.00046666666666666666,
"loss": 3.0547,
"step": 40000
},
{
"epoch": 11.28,
"learning_rate": 0.0004655555555555556,
"learning_rate_embeddings": 0.0004655555555555556,
"loss": 3.0579,
"step": 40500
},
{
"epoch": 11.42,
"learning_rate": 0.00046444444444444446,
"learning_rate_embeddings": 0.00046444444444444446,
"loss": 3.0598,
"step": 41000
},
{
"epoch": 11.56,
"learning_rate": 0.00046333333333333334,
"learning_rate_embeddings": 0.00046333333333333334,
"loss": 3.0549,
"step": 41500
},
{
"epoch": 11.7,
"learning_rate": 0.0004622222222222222,
"learning_rate_embeddings": 0.0004622222222222222,
"loss": 3.0635,
"step": 42000
},
{
"epoch": 11.84,
"learning_rate": 0.00046111111111111114,
"learning_rate_embeddings": 0.00046111111111111114,
"loss": 3.0648,
"step": 42500
},
{
"epoch": 11.97,
"learning_rate": 0.00046,
"learning_rate_embeddings": 0.00046,
"loss": 3.0614,
"step": 43000
},
{
"epoch": 12.11,
"learning_rate": 0.0004588888888888889,
"learning_rate_embeddings": 0.0004588888888888889,
"loss": 3.046,
"step": 43500
},
{
"epoch": 12.25,
"learning_rate": 0.0004577777777777778,
"learning_rate_embeddings": 0.0004577777777777778,
"loss": 3.0464,
"step": 44000
},
{
"epoch": 12.39,
"learning_rate": 0.0004566666666666667,
"learning_rate_embeddings": 0.0004566666666666667,
"loss": 3.0481,
"step": 44500
},
{
"epoch": 12.53,
"learning_rate": 0.00045555555555555556,
"learning_rate_embeddings": 0.00045555555555555556,
"loss": 3.0445,
"step": 45000
},
{
"epoch": 12.67,
"learning_rate": 0.00045444444444444444,
"learning_rate_embeddings": 0.00045444444444444444,
"loss": 3.0547,
"step": 45500
},
{
"epoch": 12.81,
"learning_rate": 0.0004533333333333333,
"learning_rate_embeddings": 0.0004533333333333333,
"loss": 3.0611,
"step": 46000
},
{
"epoch": 12.95,
"learning_rate": 0.00045222222222222224,
"learning_rate_embeddings": 0.00045222222222222224,
"loss": 3.055,
"step": 46500
},
{
"epoch": 13.09,
"learning_rate": 0.0004511111111111111,
"learning_rate_embeddings": 0.0004511111111111111,
"loss": 3.0476,
"step": 47000
},
{
"epoch": 13.23,
"learning_rate": 0.00045000000000000004,
"learning_rate_embeddings": 0.00045000000000000004,
"loss": 3.0308,
"step": 47500
},
{
"epoch": 13.37,
"learning_rate": 0.0004488888888888889,
"learning_rate_embeddings": 0.0004488888888888889,
"loss": 3.0415,
"step": 48000
},
{
"epoch": 13.51,
"learning_rate": 0.0004477777777777778,
"learning_rate_embeddings": 0.0004477777777777778,
"loss": 3.0434,
"step": 48500
},
{
"epoch": 13.65,
"learning_rate": 0.00044666666666666666,
"learning_rate_embeddings": 0.00044666666666666666,
"loss": 3.0444,
"step": 49000
},
{
"epoch": 13.78,
"learning_rate": 0.00044555555555555554,
"learning_rate_embeddings": 0.00044555555555555554,
"loss": 3.0389,
"step": 49500
},
{
"epoch": 13.92,
"learning_rate": 0.0004444444444444444,
"learning_rate_embeddings": 0.0004444444444444444,
"loss": 3.0458,
"step": 50000
},
{
"epoch": 13.92,
"eval_loss": 3.0721282958984375,
"eval_runtime": 371.4664,
"eval_samples_per_second": 496.82,
"eval_steps_per_second": 3.882,
"step": 50000
},
{
"epoch": 14.06,
"learning_rate": 0.00044333333333333334,
"learning_rate_embeddings": 0.00044333333333333334,
"loss": 3.0367,
"step": 50500
},
{
"epoch": 14.2,
"learning_rate": 0.00044222222222222227,
"learning_rate_embeddings": 0.00044222222222222227,
"loss": 3.0245,
"step": 51000
},
{
"epoch": 14.34,
"learning_rate": 0.00044111111111111114,
"learning_rate_embeddings": 0.00044111111111111114,
"loss": 3.0234,
"step": 51500
},
{
"epoch": 14.48,
"learning_rate": 0.00044,
"learning_rate_embeddings": 0.00044,
"loss": 3.0328,
"step": 52000
},
{
"epoch": 14.62,
"learning_rate": 0.0004388888888888889,
"learning_rate_embeddings": 0.0004388888888888889,
"loss": 3.0301,
"step": 52500
},
{
"epoch": 14.76,
"learning_rate": 0.00043777777777777776,
"learning_rate_embeddings": 0.00043777777777777776,
"loss": 3.0275,
"step": 53000
},
{
"epoch": 14.9,
"learning_rate": 0.00043666666666666664,
"learning_rate_embeddings": 0.00043666666666666664,
"loss": 3.0303,
"step": 53500
},
{
"epoch": 15.04,
"learning_rate": 0.0004355555555555555,
"learning_rate_embeddings": 0.0004355555555555555,
"loss": 3.0339,
"step": 54000
},
{
"epoch": 15.18,
"learning_rate": 0.0004344444444444445,
"learning_rate_embeddings": 0.0004344444444444445,
"loss": 3.0195,
"step": 54500
},
{
"epoch": 15.32,
"learning_rate": 0.00043333333333333337,
"learning_rate_embeddings": 0.00043333333333333337,
"loss": 3.0201,
"step": 55000
},
{
"epoch": 15.46,
"learning_rate": 0.00043222222222222224,
"learning_rate_embeddings": 0.00043222222222222224,
"loss": 3.0305,
"step": 55500
},
{
"epoch": 15.59,
"learning_rate": 0.0004311111111111111,
"learning_rate_embeddings": 0.0004311111111111111,
"loss": 3.0345,
"step": 56000
},
{
"epoch": 15.73,
"learning_rate": 0.00043,
"learning_rate_embeddings": 0.00043,
"loss": 3.0209,
"step": 56500
},
{
"epoch": 15.87,
"learning_rate": 0.00042888888888888886,
"learning_rate_embeddings": 0.00042888888888888886,
"loss": 3.0264,
"step": 57000
},
{
"epoch": 16.01,
"learning_rate": 0.0004277777777777778,
"learning_rate_embeddings": 0.0004277777777777778,
"loss": 3.0239,
"step": 57500
},
{
"epoch": 16.15,
"learning_rate": 0.0004266666666666667,
"learning_rate_embeddings": 0.0004266666666666667,
"loss": 3.0056,
"step": 58000
},
{
"epoch": 16.29,
"learning_rate": 0.0004255555555555556,
"learning_rate_embeddings": 0.0004255555555555556,
"loss": 3.0169,
"step": 58500
},
{
"epoch": 16.43,
"learning_rate": 0.00042444444444444447,
"learning_rate_embeddings": 0.00042444444444444447,
"loss": 3.0115,
"step": 59000
},
{
"epoch": 16.57,
"learning_rate": 0.00042333333333333334,
"learning_rate_embeddings": 0.00042333333333333334,
"loss": 3.022,
"step": 59500
},
{
"epoch": 16.71,
"learning_rate": 0.0004222222222222222,
"learning_rate_embeddings": 0.0004222222222222222,
"loss": 3.0212,
"step": 60000
},
{
"epoch": 16.85,
"learning_rate": 0.0004211111111111111,
"learning_rate_embeddings": 0.0004211111111111111,
"loss": 3.0194,
"step": 60500
},
{
"epoch": 16.99,
"learning_rate": 0.00042,
"learning_rate_embeddings": 0.00042,
"loss": 3.0207,
"step": 61000
},
{
"epoch": 17.13,
"learning_rate": 0.0004188888888888889,
"learning_rate_embeddings": 0.0004188888888888889,
"loss": 3.0067,
"step": 61500
},
{
"epoch": 17.27,
"learning_rate": 0.0004177777777777778,
"learning_rate_embeddings": 0.0004177777777777778,
"loss": 3.0035,
"step": 62000
},
{
"epoch": 17.4,
"learning_rate": 0.0004166666666666667,
"learning_rate_embeddings": 0.0004166666666666667,
"loss": 3.0143,
"step": 62500
},
{
"epoch": 17.4,
"eval_loss": 3.0584957599639893,
"eval_runtime": 372.4044,
"eval_samples_per_second": 495.569,
"eval_steps_per_second": 3.872,
"step": 62500
},
{
"epoch": 17.54,
"learning_rate": 0.00041555555555555557,
"learning_rate_embeddings": 0.00041555555555555557,
"loss": 3.0167,
"step": 63000
},
{
"epoch": 17.68,
"learning_rate": 0.00041444444444444444,
"learning_rate_embeddings": 0.00041444444444444444,
"loss": 3.0174,
"step": 63500
},
{
"epoch": 17.82,
"learning_rate": 0.0004133333333333333,
"learning_rate_embeddings": 0.0004133333333333333,
"loss": 3.009,
"step": 64000
},
{
"epoch": 17.96,
"learning_rate": 0.00041222222222222224,
"learning_rate_embeddings": 0.00041222222222222224,
"loss": 3.0174,
"step": 64500
},
{
"epoch": 18.1,
"learning_rate": 0.0004111111111111111,
"learning_rate_embeddings": 0.0004111111111111111,
"loss": 2.9896,
"step": 65000
},
{
"epoch": 18.24,
"learning_rate": 0.00041,
"learning_rate_embeddings": 0.00041,
"loss": 2.9986,
"step": 65500
},
{
"epoch": 18.38,
"learning_rate": 0.0004088888888888889,
"learning_rate_embeddings": 0.0004088888888888889,
"loss": 2.9994,
"step": 66000
},
{
"epoch": 18.52,
"learning_rate": 0.0004077777777777778,
"learning_rate_embeddings": 0.0004077777777777778,
"loss": 3.0095,
"step": 66500
},
{
"epoch": 18.66,
"learning_rate": 0.00040666666666666667,
"learning_rate_embeddings": 0.00040666666666666667,
"loss": 3.0088,
"step": 67000
},
{
"epoch": 18.8,
"learning_rate": 0.00040555555555555554,
"learning_rate_embeddings": 0.00040555555555555554,
"loss": 3.0075,
"step": 67500
},
{
"epoch": 18.94,
"learning_rate": 0.00040444444444444447,
"learning_rate_embeddings": 0.00040444444444444447,
"loss": 3.0176,
"step": 68000
},
{
"epoch": 19.08,
"learning_rate": 0.00040333333333333334,
"learning_rate_embeddings": 0.00040333333333333334,
"loss": 2.9948,
"step": 68500
},
{
"epoch": 19.21,
"learning_rate": 0.0004022222222222222,
"learning_rate_embeddings": 0.0004022222222222222,
"loss": 2.9915,
"step": 69000
},
{
"epoch": 19.35,
"learning_rate": 0.0004011111111111111,
"learning_rate_embeddings": 0.0004011111111111111,
"loss": 2.9945,
"step": 69500
},
{
"epoch": 19.49,
"learning_rate": 0.0004,
"learning_rate_embeddings": 0.0004,
"loss": 3.0017,
"step": 70000
},
{
"epoch": 19.63,
"learning_rate": 0.0003988888888888889,
"learning_rate_embeddings": 0.0003988888888888889,
"loss": 3.003,
"step": 70500
},
{
"epoch": 19.77,
"learning_rate": 0.00039777777777777777,
"learning_rate_embeddings": 0.00039777777777777777,
"loss": 2.9999,
"step": 71000
},
{
"epoch": 19.91,
"learning_rate": 0.0003966666666666667,
"learning_rate_embeddings": 0.0003966666666666667,
"loss": 3.0018,
"step": 71500
},
{
"epoch": 20.05,
"learning_rate": 0.00039555555555555557,
"learning_rate_embeddings": 0.00039555555555555557,
"loss": 2.9957,
"step": 72000
},
{
"epoch": 20.19,
"learning_rate": 0.00039444444444444444,
"learning_rate_embeddings": 0.00039444444444444444,
"loss": 2.9824,
"step": 72500
},
{
"epoch": 20.33,
"learning_rate": 0.0003933333333333333,
"learning_rate_embeddings": 0.0003933333333333333,
"loss": 2.9868,
"step": 73000
},
{
"epoch": 20.47,
"learning_rate": 0.00039222222222222225,
"learning_rate_embeddings": 0.00039222222222222225,
"loss": 2.9939,
"step": 73500
},
{
"epoch": 20.61,
"learning_rate": 0.0003911111111111111,
"learning_rate_embeddings": 0.0003911111111111111,
"loss": 3.0016,
"step": 74000
},
{
"epoch": 20.75,
"learning_rate": 0.00039000000000000005,
"learning_rate_embeddings": 0.00039000000000000005,
"loss": 2.9937,
"step": 74500
},
{
"epoch": 20.89,
"learning_rate": 0.0003888888888888889,
"learning_rate_embeddings": 0.0003888888888888889,
"loss": 3.0017,
"step": 75000
},
{
"epoch": 20.89,
"eval_loss": 3.0433623790740967,
"eval_runtime": 372.4329,
"eval_samples_per_second": 495.531,
"eval_steps_per_second": 3.872,
"step": 75000
},
{
"epoch": 21.02,
"learning_rate": 0.0003877777777777778,
"learning_rate_embeddings": 0.0003877777777777778,
"loss": 2.9915,
"step": 75500
},
{
"epoch": 21.16,
"learning_rate": 0.00038666666666666667,
"learning_rate_embeddings": 0.00038666666666666667,
"loss": 2.9781,
"step": 76000
},
{
"epoch": 21.3,
"learning_rate": 0.00038555555555555554,
"learning_rate_embeddings": 0.00038555555555555554,
"loss": 2.9834,
"step": 76500
},
{
"epoch": 21.44,
"learning_rate": 0.0003844444444444444,
"learning_rate_embeddings": 0.0003844444444444444,
"loss": 2.9903,
"step": 77000
},
{
"epoch": 21.58,
"learning_rate": 0.00038333333333333334,
"learning_rate_embeddings": 0.00038333333333333334,
"loss": 2.9895,
"step": 77500
},
{
"epoch": 21.72,
"learning_rate": 0.0003822222222222223,
"learning_rate_embeddings": 0.0003822222222222223,
"loss": 2.9909,
"step": 78000
},
{
"epoch": 21.86,
"learning_rate": 0.00038111111111111115,
"learning_rate_embeddings": 0.00038111111111111115,
"loss": 2.9985,
"step": 78500
},
{
"epoch": 22.0,
"learning_rate": 0.00038,
"learning_rate_embeddings": 0.00038,
"loss": 2.994,
"step": 79000
},
{
"epoch": 22.14,
"learning_rate": 0.0003788888888888889,
"learning_rate_embeddings": 0.0003788888888888889,
"loss": 2.9787,
"step": 79500
},
{
"epoch": 22.28,
"learning_rate": 0.00037777777777777777,
"learning_rate_embeddings": 0.00037777777777777777,
"loss": 2.9791,
"step": 80000
},
{
"epoch": 22.42,
"learning_rate": 0.00037666666666666664,
"learning_rate_embeddings": 0.00037666666666666664,
"loss": 2.9859,
"step": 80500
},
{
"epoch": 22.56,
"learning_rate": 0.0003755555555555555,
"learning_rate_embeddings": 0.0003755555555555555,
"loss": 2.9891,
"step": 81000
},
{
"epoch": 22.7,
"learning_rate": 0.0003744444444444445,
"learning_rate_embeddings": 0.0003744444444444445,
"loss": 2.9873,
"step": 81500
},
{
"epoch": 22.83,
"learning_rate": 0.0003733333333333334,
"learning_rate_embeddings": 0.0003733333333333334,
"loss": 2.9902,
"step": 82000
},
{
"epoch": 22.97,
"learning_rate": 0.00037222222222222225,
"learning_rate_embeddings": 0.00037222222222222225,
"loss": 2.9869,
"step": 82500
},
{
"epoch": 23.11,
"learning_rate": 0.0003711111111111111,
"learning_rate_embeddings": 0.0003711111111111111,
"loss": 2.9785,
"step": 83000
},
{
"epoch": 23.25,
"learning_rate": 0.00037,
"learning_rate_embeddings": 0.00037,
"loss": 2.9751,
"step": 83500
},
{
"epoch": 23.39,
"learning_rate": 0.00036888888888888887,
"learning_rate_embeddings": 0.00036888888888888887,
"loss": 2.9821,
"step": 84000
},
{
"epoch": 23.53,
"learning_rate": 0.00036777777777777774,
"learning_rate_embeddings": 0.00036777777777777774,
"loss": 2.9825,
"step": 84500
},
{
"epoch": 23.67,
"learning_rate": 0.00036666666666666667,
"learning_rate_embeddings": 0.00036666666666666667,
"loss": 2.9846,
"step": 85000
},
{
"epoch": 23.81,
"learning_rate": 0.0003655555555555556,
"learning_rate_embeddings": 0.0003655555555555556,
"loss": 2.9916,
"step": 85500
},
{
"epoch": 23.95,
"learning_rate": 0.00036444444444444447,
"learning_rate_embeddings": 0.00036444444444444447,
"loss": 2.9819,
"step": 86000
},
{
"epoch": 24.09,
"learning_rate": 0.00036333333333333335,
"learning_rate_embeddings": 0.00036333333333333335,
"loss": 2.9715,
"step": 86500
},
{
"epoch": 24.23,
"learning_rate": 0.0003622222222222222,
"learning_rate_embeddings": 0.0003622222222222222,
"loss": 2.981,
"step": 87000
},
{
"epoch": 24.37,
"learning_rate": 0.0003611111111111111,
"learning_rate_embeddings": 0.0003611111111111111,
"loss": 2.9809,
"step": 87500
},
{
"epoch": 24.37,
"eval_loss": 3.0395069122314453,
"eval_runtime": 372.3364,
"eval_samples_per_second": 495.659,
"eval_steps_per_second": 3.873,
"step": 87500
},
{
"epoch": 24.51,
"learning_rate": 0.00035999999999999997,
"learning_rate_embeddings": 0.00035999999999999997,
"loss": 2.9762,
"step": 88000
},
{
"epoch": 24.64,
"learning_rate": 0.0003588888888888889,
"learning_rate_embeddings": 0.0003588888888888889,
"loss": 2.9833,
"step": 88500
},
{
"epoch": 24.78,
"learning_rate": 0.00035777777777777777,
"learning_rate_embeddings": 0.00035777777777777777,
"loss": 2.9825,
"step": 89000
},
{
"epoch": 24.92,
"learning_rate": 0.0003566666666666667,
"learning_rate_embeddings": 0.0003566666666666667,
"loss": 2.9818,
"step": 89500
},
{
"epoch": 25.06,
"learning_rate": 0.00035555555555555557,
"learning_rate_embeddings": 0.00035555555555555557,
"loss": 2.9766,
"step": 90000
},
{
"epoch": 25.2,
"learning_rate": 0.00035444444444444445,
"learning_rate_embeddings": 0.00035444444444444445,
"loss": 2.9756,
"step": 90500
},
{
"epoch": 25.34,
"learning_rate": 0.0003533333333333333,
"learning_rate_embeddings": 0.0003533333333333333,
"loss": 2.9714,
"step": 91000
},
{
"epoch": 25.48,
"learning_rate": 0.00035222222222222225,
"learning_rate_embeddings": 0.00035222222222222225,
"loss": 2.9687,
"step": 91500
},
{
"epoch": 25.62,
"learning_rate": 0.0003511111111111111,
"learning_rate_embeddings": 0.0003511111111111111,
"loss": 2.9801,
"step": 92000
},
{
"epoch": 25.76,
"learning_rate": 0.00035,
"learning_rate_embeddings": 0.00035,
"loss": 2.9767,
"step": 92500
},
{
"epoch": 25.9,
"learning_rate": 0.0003488888888888889,
"learning_rate_embeddings": 0.0003488888888888889,
"loss": 2.9796,
"step": 93000
},
{
"epoch": 26.04,
"learning_rate": 0.0003477777777777778,
"learning_rate_embeddings": 0.0003477777777777778,
"loss": 2.9668,
"step": 93500
},
{
"epoch": 26.18,
"learning_rate": 0.00034666666666666667,
"learning_rate_embeddings": 0.00034666666666666667,
"loss": 2.9654,
"step": 94000
},
{
"epoch": 26.32,
"learning_rate": 0.00034555555555555555,
"learning_rate_embeddings": 0.00034555555555555555,
"loss": 2.9637,
"step": 94500
},
{
"epoch": 26.45,
"learning_rate": 0.0003444444444444445,
"learning_rate_embeddings": 0.0003444444444444445,
"loss": 2.9736,
"step": 95000
},
{
"epoch": 26.59,
"learning_rate": 0.00034333333333333335,
"learning_rate_embeddings": 0.00034333333333333335,
"loss": 2.9742,
"step": 95500
},
{
"epoch": 26.73,
"learning_rate": 0.0003422222222222222,
"learning_rate_embeddings": 0.0003422222222222222,
"loss": 2.9767,
"step": 96000
},
{
"epoch": 26.87,
"learning_rate": 0.0003411111111111111,
"learning_rate_embeddings": 0.0003411111111111111,
"loss": 2.9831,
"step": 96500
},
{
"epoch": 27.01,
"learning_rate": 0.00034,
"learning_rate_embeddings": 0.00034,
"loss": 2.975,
"step": 97000
},
{
"epoch": 27.15,
"learning_rate": 0.0003388888888888889,
"learning_rate_embeddings": 0.0003388888888888889,
"loss": 2.9581,
"step": 97500
},
{
"epoch": 27.29,
"learning_rate": 0.00033777777777777777,
"learning_rate_embeddings": 0.00033777777777777777,
"loss": 2.9659,
"step": 98000
},
{
"epoch": 27.43,
"learning_rate": 0.0003366666666666667,
"learning_rate_embeddings": 0.0003366666666666667,
"loss": 2.9714,
"step": 98500
},
{
"epoch": 27.57,
"learning_rate": 0.0003355555555555556,
"learning_rate_embeddings": 0.0003355555555555556,
"loss": 2.9697,
"step": 99000
},
{
"epoch": 27.71,
"learning_rate": 0.00033444444444444445,
"learning_rate_embeddings": 0.00033444444444444445,
"loss": 2.9753,
"step": 99500
},
{
"epoch": 27.85,
"learning_rate": 0.0003333333333333333,
"learning_rate_embeddings": 0.0003333333333333333,
"loss": 2.9737,
"step": 100000
},
{
"epoch": 27.85,
"eval_loss": 3.029412031173706,
"eval_runtime": 371.951,
"eval_samples_per_second": 496.173,
"eval_steps_per_second": 3.877,
"step": 100000
},
{
"epoch": 27.99,
"learning_rate": 0.0003322222222222222,
"learning_rate_embeddings": 0.0003322222222222222,
"loss": 2.9716,
"step": 100500
},
{
"epoch": 28.13,
"learning_rate": 0.0003311111111111111,
"learning_rate_embeddings": 0.0003311111111111111,
"loss": 2.9588,
"step": 101000
},
{
"epoch": 28.27,
"learning_rate": 0.00033,
"learning_rate_embeddings": 0.00033,
"loss": 2.963,
"step": 101500
},
{
"epoch": 28.4,
"learning_rate": 0.0003288888888888889,
"learning_rate_embeddings": 0.0003288888888888889,
"loss": 2.9687,
"step": 102000
},
{
"epoch": 28.54,
"learning_rate": 0.0003277777777777778,
"learning_rate_embeddings": 0.0003277777777777778,
"loss": 2.9656,
"step": 102500
},
{
"epoch": 28.68,
"learning_rate": 0.0003266666666666667,
"learning_rate_embeddings": 0.0003266666666666667,
"loss": 2.9737,
"step": 103000
},
{
"epoch": 28.82,
"learning_rate": 0.00032555555555555555,
"learning_rate_embeddings": 0.00032555555555555555,
"loss": 2.9756,
"step": 103500
},
{
"epoch": 28.96,
"learning_rate": 0.0003244444444444444,
"learning_rate_embeddings": 0.0003244444444444444,
"loss": 2.9722,
"step": 104000
},
{
"epoch": 29.1,
"learning_rate": 0.0003233333333333333,
"learning_rate_embeddings": 0.0003233333333333333,
"loss": 2.9641,
"step": 104500
},
{
"epoch": 29.24,
"learning_rate": 0.0003222222222222222,
"learning_rate_embeddings": 0.0003222222222222222,
"loss": 2.9624,
"step": 105000
},
{
"epoch": 29.38,
"learning_rate": 0.00032111111111111115,
"learning_rate_embeddings": 0.00032111111111111115,
"loss": 2.9622,
"step": 105500
},
{
"epoch": 29.52,
"learning_rate": 0.00032,
"learning_rate_embeddings": 0.00032,
"loss": 2.9678,
"step": 106000
},
{
"epoch": 29.66,
"learning_rate": 0.0003188888888888889,
"learning_rate_embeddings": 0.0003188888888888889,
"loss": 2.9656,
"step": 106500
},
{
"epoch": 29.8,
"learning_rate": 0.0003177777777777778,
"learning_rate_embeddings": 0.0003177777777777778,
"loss": 2.9703,
"step": 107000
},
{
"epoch": 29.94,
"learning_rate": 0.00031666666666666665,
"learning_rate_embeddings": 0.00031666666666666665,
"loss": 2.9671,
"step": 107500
},
{
"epoch": 30.08,
"learning_rate": 0.0003155555555555555,
"learning_rate_embeddings": 0.0003155555555555555,
"loss": 2.9646,
"step": 108000
},
{
"epoch": 30.21,
"learning_rate": 0.0003144444444444445,
"learning_rate_embeddings": 0.0003144444444444445,
"loss": 2.9542,
"step": 108500
},
{
"epoch": 30.35,
"learning_rate": 0.0003133333333333334,
"learning_rate_embeddings": 0.0003133333333333334,
"loss": 2.9569,
"step": 109000
},
{
"epoch": 30.49,
"learning_rate": 0.00031222222222222225,
"learning_rate_embeddings": 0.00031222222222222225,
"loss": 2.9601,
"step": 109500
},
{
"epoch": 30.63,
"learning_rate": 0.0003111111111111111,
"learning_rate_embeddings": 0.0003111111111111111,
"loss": 2.9666,
"step": 110000
},
{
"epoch": 30.77,
"learning_rate": 0.00031,
"learning_rate_embeddings": 0.00031,
"loss": 2.9671,
"step": 110500
},
{
"epoch": 30.91,
"learning_rate": 0.0003088888888888889,
"learning_rate_embeddings": 0.0003088888888888889,
"loss": 2.9688,
"step": 111000
},
{
"epoch": 31.05,
"learning_rate": 0.00030777777777777775,
"learning_rate_embeddings": 0.00030777777777777775,
"loss": 2.9593,
"step": 111500
},
{
"epoch": 31.19,
"learning_rate": 0.0003066666666666667,
"learning_rate_embeddings": 0.0003066666666666667,
"loss": 2.9461,
"step": 112000
},
{
"epoch": 31.33,
"learning_rate": 0.0003055555555555556,
"learning_rate_embeddings": 0.0003055555555555556,
"loss": 2.9599,
"step": 112500
},
{
"epoch": 31.33,
"eval_loss": 3.028226375579834,
"eval_runtime": 371.9034,
"eval_samples_per_second": 496.236,
"eval_steps_per_second": 3.877,
"step": 112500
},
{
"epoch": 31.47,
"learning_rate": 0.0003044444444444445,
"learning_rate_embeddings": 0.0003044444444444445,
"loss": 2.9567,
"step": 113000
},
{
"epoch": 31.61,
"learning_rate": 0.00030333333333333335,
"learning_rate_embeddings": 0.00030333333333333335,
"loss": 2.9622,
"step": 113500
},
{
"epoch": 31.75,
"learning_rate": 0.0003022222222222222,
"learning_rate_embeddings": 0.0003022222222222222,
"loss": 2.966,
"step": 114000
},
{
"epoch": 31.89,
"learning_rate": 0.0003011111111111111,
"learning_rate_embeddings": 0.0003011111111111111,
"loss": 2.9648,
"step": 114500
},
{
"epoch": 32.02,
"learning_rate": 0.0003,
"learning_rate_embeddings": 0.0003,
"loss": 2.9599,
"step": 115000
},
{
"epoch": 32.16,
"learning_rate": 0.0002988888888888889,
"learning_rate_embeddings": 0.0002988888888888889,
"loss": 2.9498,
"step": 115500
},
{
"epoch": 32.3,
"learning_rate": 0.0002977777777777778,
"learning_rate_embeddings": 0.0002977777777777778,
"loss": 2.9547,
"step": 116000
},
{
"epoch": 32.44,
"learning_rate": 0.0002966666666666667,
"learning_rate_embeddings": 0.0002966666666666667,
"loss": 2.9573,
"step": 116500
},
{
"epoch": 32.58,
"learning_rate": 0.0002955555555555556,
"learning_rate_embeddings": 0.0002955555555555556,
"loss": 2.9509,
"step": 117000
},
{
"epoch": 32.72,
"learning_rate": 0.00029444444444444445,
"learning_rate_embeddings": 0.00029444444444444445,
"loss": 2.9599,
"step": 117500
},
{
"epoch": 32.86,
"learning_rate": 0.0002933333333333333,
"learning_rate_embeddings": 0.0002933333333333333,
"loss": 2.9641,
"step": 118000
},
{
"epoch": 33.0,
"learning_rate": 0.0002922222222222222,
"learning_rate_embeddings": 0.0002922222222222222,
"loss": 2.9676,
"step": 118500
},
{
"epoch": 33.14,
"learning_rate": 0.00029111111111111113,
"learning_rate_embeddings": 0.00029111111111111113,
"loss": 2.9484,
"step": 119000
},
{
"epoch": 33.28,
"learning_rate": 0.00029,
"learning_rate_embeddings": 0.00029,
"loss": 2.9464,
"step": 119500
},
{
"epoch": 33.42,
"learning_rate": 0.0002888888888888889,
"learning_rate_embeddings": 0.0002888888888888889,
"loss": 2.9512,
"step": 120000
},
{
"epoch": 33.56,
"learning_rate": 0.0002877777777777778,
"learning_rate_embeddings": 0.0002877777777777778,
"loss": 2.9515,
"step": 120500
},
{
"epoch": 33.7,
"learning_rate": 0.0002866666666666667,
"learning_rate_embeddings": 0.0002866666666666667,
"loss": 2.9572,
"step": 121000
},
{
"epoch": 33.83,
"learning_rate": 0.00028555555555555555,
"learning_rate_embeddings": 0.00028555555555555555,
"loss": 2.9568,
"step": 121500
},
{
"epoch": 33.97,
"learning_rate": 0.0002844444444444444,
"learning_rate_embeddings": 0.0002844444444444444,
"loss": 2.9584,
"step": 122000
},
{
"epoch": 34.11,
"learning_rate": 0.00028333333333333335,
"learning_rate_embeddings": 0.00028333333333333335,
"loss": 2.9487,
"step": 122500
},
{
"epoch": 34.25,
"learning_rate": 0.00028222222222222223,
"learning_rate_embeddings": 0.00028222222222222223,
"loss": 2.9521,
"step": 123000
},
{
"epoch": 34.39,
"learning_rate": 0.0002811111111111111,
"learning_rate_embeddings": 0.0002811111111111111,
"loss": 2.9571,
"step": 123500
},
{
"epoch": 34.53,
"learning_rate": 0.00028000000000000003,
"learning_rate_embeddings": 0.00028000000000000003,
"loss": 2.9541,
"step": 124000
},
{
"epoch": 34.67,
"learning_rate": 0.0002788888888888889,
"learning_rate_embeddings": 0.0002788888888888889,
"loss": 2.9535,
"step": 124500
},
{
"epoch": 34.81,
"learning_rate": 0.0002777777777777778,
"learning_rate_embeddings": 0.0002777777777777778,
"loss": 2.956,
"step": 125000
},
{
"epoch": 34.81,
"eval_loss": 3.0202476978302,
"eval_runtime": 371.6194,
"eval_samples_per_second": 496.616,
"eval_steps_per_second": 3.88,
"step": 125000
},
{
"epoch": 34.95,
"learning_rate": 0.00027666666666666665,
"learning_rate_embeddings": 0.00027666666666666665,
"loss": 2.9584,
"step": 125500
},
{
"epoch": 35.09,
"learning_rate": 0.0002755555555555556,
"learning_rate_embeddings": 0.0002755555555555556,
"loss": 2.9457,
"step": 126000
},
{
"epoch": 35.23,
"learning_rate": 0.00027444444444444445,
"learning_rate_embeddings": 0.00027444444444444445,
"loss": 2.953,
"step": 126500
},
{
"epoch": 35.37,
"learning_rate": 0.00027333333333333333,
"learning_rate_embeddings": 0.00027333333333333333,
"loss": 2.9454,
"step": 127000
},
{
"epoch": 35.51,
"learning_rate": 0.0002722222222222222,
"learning_rate_embeddings": 0.0002722222222222222,
"loss": 2.951,
"step": 127500
},
{
"epoch": 35.64,
"learning_rate": 0.00027111111111111113,
"learning_rate_embeddings": 0.00027111111111111113,
"loss": 2.9565,
"step": 128000
},
{
"epoch": 35.78,
"learning_rate": 0.00027,
"learning_rate_embeddings": 0.00027,
"loss": 2.9517,
"step": 128500
},
{
"epoch": 35.92,
"learning_rate": 0.00026888888888888893,
"learning_rate_embeddings": 0.00026888888888888893,
"loss": 2.9531,
"step": 129000
},
{
"epoch": 36.06,
"learning_rate": 0.0002677777777777778,
"learning_rate_embeddings": 0.0002677777777777778,
"loss": 2.9478,
"step": 129500
},
{
"epoch": 36.2,
"learning_rate": 0.0002666666666666667,
"learning_rate_embeddings": 0.0002666666666666667,
"loss": 2.941,
"step": 130000
},
{
"epoch": 36.34,
"learning_rate": 0.00026555555555555555,
"learning_rate_embeddings": 0.00026555555555555555,
"loss": 2.9475,
"step": 130500
},
{
"epoch": 36.48,
"learning_rate": 0.00026444444444444443,
"learning_rate_embeddings": 0.00026444444444444443,
"loss": 2.944,
"step": 131000
},
{
"epoch": 36.62,
"learning_rate": 0.0002633333333333333,
"learning_rate_embeddings": 0.0002633333333333333,
"loss": 2.9524,
"step": 131500
},
{
"epoch": 36.76,
"learning_rate": 0.00026222222222222223,
"learning_rate_embeddings": 0.00026222222222222223,
"loss": 2.9598,
"step": 132000
},
{
"epoch": 36.9,
"learning_rate": 0.00026111111111111116,
"learning_rate_embeddings": 0.00026111111111111116,
"loss": 2.9563,
"step": 132500
},
{
"epoch": 37.04,
"learning_rate": 0.00026000000000000003,
"learning_rate_embeddings": 0.00026000000000000003,
"loss": 2.9514,
"step": 133000
},
{
"epoch": 37.18,
"learning_rate": 0.0002588888888888889,
"learning_rate_embeddings": 0.0002588888888888889,
"loss": 2.9466,
"step": 133500
},
{
"epoch": 37.32,
"learning_rate": 0.0002577777777777778,
"learning_rate_embeddings": 0.0002577777777777778,
"loss": 2.9474,
"step": 134000
},
{
"epoch": 37.45,
"learning_rate": 0.00025666666666666665,
"learning_rate_embeddings": 0.00025666666666666665,
"loss": 2.9447,
"step": 134500
},
{
"epoch": 37.59,
"learning_rate": 0.00025555555555555553,
"learning_rate_embeddings": 0.00025555555555555553,
"loss": 2.9452,
"step": 135000
},
{
"epoch": 37.73,
"learning_rate": 0.0002544444444444444,
"learning_rate_embeddings": 0.0002544444444444444,
"loss": 2.9533,
"step": 135500
},
{
"epoch": 37.87,
"learning_rate": 0.0002533333333333334,
"learning_rate_embeddings": 0.0002533333333333334,
"loss": 2.9492,
"step": 136000
},
{
"epoch": 38.01,
"learning_rate": 0.00025222222222222226,
"learning_rate_embeddings": 0.00025222222222222226,
"loss": 2.9486,
"step": 136500
},
{
"epoch": 38.15,
"learning_rate": 0.00025111111111111113,
"learning_rate_embeddings": 0.00025111111111111113,
"loss": 2.94,
"step": 137000
},
{
"epoch": 38.29,
"learning_rate": 0.00025,
"learning_rate_embeddings": 0.00025,
"loss": 2.9434,
"step": 137500
},
{
"epoch": 38.29,
"eval_loss": 3.0181665420532227,
"eval_runtime": 371.7979,
"eval_samples_per_second": 496.377,
"eval_steps_per_second": 3.878,
"step": 137500
},
{
"epoch": 38.43,
"learning_rate": 0.0002488888888888889,
"learning_rate_embeddings": 0.0002488888888888889,
"loss": 2.9442,
"step": 138000
},
{
"epoch": 38.57,
"learning_rate": 0.0002477777777777778,
"learning_rate_embeddings": 0.0002477777777777778,
"loss": 2.9447,
"step": 138500
},
{
"epoch": 38.71,
"learning_rate": 0.0002466666666666667,
"learning_rate_embeddings": 0.0002466666666666667,
"loss": 2.9478,
"step": 139000
},
{
"epoch": 38.85,
"learning_rate": 0.00024555555555555556,
"learning_rate_embeddings": 0.00024555555555555556,
"loss": 2.9467,
"step": 139500
},
{
"epoch": 38.99,
"learning_rate": 0.00024444444444444443,
"learning_rate_embeddings": 0.00024444444444444443,
"loss": 2.9508,
"step": 140000
},
{
"epoch": 39.13,
"learning_rate": 0.00024333333333333336,
"learning_rate_embeddings": 0.00024333333333333336,
"loss": 2.9367,
"step": 140500
},
{
"epoch": 39.26,
"learning_rate": 0.00024222222222222223,
"learning_rate_embeddings": 0.00024222222222222223,
"loss": 2.9416,
"step": 141000
},
{
"epoch": 39.4,
"learning_rate": 0.0002411111111111111,
"learning_rate_embeddings": 0.0002411111111111111,
"loss": 2.9424,
"step": 141500
},
{
"epoch": 39.54,
"learning_rate": 0.00024,
"learning_rate_embeddings": 0.00024,
"loss": 2.9483,
"step": 142000
},
{
"epoch": 39.68,
"learning_rate": 0.0002388888888888889,
"learning_rate_embeddings": 0.0002388888888888889,
"loss": 2.9475,
"step": 142500
},
{
"epoch": 39.82,
"learning_rate": 0.00023777777777777778,
"learning_rate_embeddings": 0.00023777777777777778,
"loss": 2.9446,
"step": 143000
},
{
"epoch": 39.96,
"learning_rate": 0.00023666666666666668,
"learning_rate_embeddings": 0.00023666666666666668,
"loss": 2.9511,
"step": 143500
},
{
"epoch": 40.1,
"learning_rate": 0.00023555555555555556,
"learning_rate_embeddings": 0.00023555555555555556,
"loss": 2.9359,
"step": 144000
},
{
"epoch": 40.24,
"learning_rate": 0.00023444444444444446,
"learning_rate_embeddings": 0.00023444444444444446,
"loss": 2.9381,
"step": 144500
},
{
"epoch": 40.38,
"learning_rate": 0.00023333333333333333,
"learning_rate_embeddings": 0.00023333333333333333,
"loss": 2.9348,
"step": 145000
},
{
"epoch": 40.52,
"learning_rate": 0.00023222222222222223,
"learning_rate_embeddings": 0.00023222222222222223,
"loss": 2.9352,
"step": 145500
},
{
"epoch": 40.66,
"learning_rate": 0.0002311111111111111,
"learning_rate_embeddings": 0.0002311111111111111,
"loss": 2.9368,
"step": 146000
},
{
"epoch": 40.8,
"learning_rate": 0.00023,
"learning_rate_embeddings": 0.00023,
"loss": 2.9408,
"step": 146500
},
{
"epoch": 40.94,
"learning_rate": 0.0002288888888888889,
"learning_rate_embeddings": 0.0002288888888888889,
"loss": 2.9437,
"step": 147000
},
{
"epoch": 41.07,
"learning_rate": 0.00022777777777777778,
"learning_rate_embeddings": 0.00022777777777777778,
"loss": 2.9407,
"step": 147500
},
{
"epoch": 41.21,
"learning_rate": 0.00022666666666666666,
"learning_rate_embeddings": 0.00022666666666666666,
"loss": 2.9401,
"step": 148000
},
{
"epoch": 41.35,
"learning_rate": 0.00022555555555555556,
"learning_rate_embeddings": 0.00022555555555555556,
"loss": 2.9404,
"step": 148500
},
{
"epoch": 41.49,
"learning_rate": 0.00022444444444444446,
"learning_rate_embeddings": 0.00022444444444444446,
"loss": 2.9385,
"step": 149000
},
{
"epoch": 41.63,
"learning_rate": 0.00022333333333333333,
"learning_rate_embeddings": 0.00022333333333333333,
"loss": 2.9441,
"step": 149500
},
{
"epoch": 41.77,
"learning_rate": 0.0002222222222222222,
"learning_rate_embeddings": 0.0002222222222222222,
"loss": 2.9449,
"step": 150000
},
{
"epoch": 41.77,
"eval_loss": 3.011936902999878,
"eval_runtime": 371.5953,
"eval_samples_per_second": 496.648,
"eval_steps_per_second": 3.881,
"step": 150000
},
{
"epoch": 41.91,
"learning_rate": 0.00022111111111111113,
"learning_rate_embeddings": 0.00022111111111111113,
"loss": 2.9414,
"step": 150500
},
{
"epoch": 42.05,
"learning_rate": 0.00022,
"learning_rate_embeddings": 0.00022,
"loss": 2.9422,
"step": 151000
},
{
"epoch": 42.19,
"learning_rate": 0.00021888888888888888,
"learning_rate_embeddings": 0.00021888888888888888,
"loss": 2.9298,
"step": 151500
},
{
"epoch": 42.33,
"learning_rate": 0.00021777777777777776,
"learning_rate_embeddings": 0.00021777777777777776,
"loss": 2.9378,
"step": 152000
},
{
"epoch": 42.47,
"learning_rate": 0.00021666666666666668,
"learning_rate_embeddings": 0.00021666666666666668,
"loss": 2.9386,
"step": 152500
},
{
"epoch": 42.61,
"learning_rate": 0.00021555555555555556,
"learning_rate_embeddings": 0.00021555555555555556,
"loss": 2.9385,
"step": 153000
},
{
"epoch": 42.75,
"learning_rate": 0.00021444444444444443,
"learning_rate_embeddings": 0.00021444444444444443,
"loss": 2.9334,
"step": 153500
},
{
"epoch": 42.88,
"learning_rate": 0.00021333333333333336,
"learning_rate_embeddings": 0.00021333333333333336,
"loss": 2.9427,
"step": 154000
},
{
"epoch": 43.02,
"learning_rate": 0.00021222222222222223,
"learning_rate_embeddings": 0.00021222222222222223,
"loss": 2.9475,
"step": 154500
},
{
"epoch": 43.16,
"learning_rate": 0.0002111111111111111,
"learning_rate_embeddings": 0.0002111111111111111,
"loss": 2.9279,
"step": 155000
},
{
"epoch": 43.3,
"learning_rate": 0.00021,
"learning_rate_embeddings": 0.00021,
"loss": 2.9308,
"step": 155500
},
{
"epoch": 43.44,
"learning_rate": 0.0002088888888888889,
"learning_rate_embeddings": 0.0002088888888888889,
"loss": 2.9331,
"step": 156000
},
{
"epoch": 43.58,
"learning_rate": 0.00020777777777777778,
"learning_rate_embeddings": 0.00020777777777777778,
"loss": 2.9312,
"step": 156500
},
{
"epoch": 43.72,
"learning_rate": 0.00020666666666666666,
"learning_rate_embeddings": 0.00020666666666666666,
"loss": 2.9393,
"step": 157000
},
{
"epoch": 43.86,
"learning_rate": 0.00020555555555555556,
"learning_rate_embeddings": 0.00020555555555555556,
"loss": 2.9401,
"step": 157500
},
{
"epoch": 44.0,
"learning_rate": 0.00020444444444444446,
"learning_rate_embeddings": 0.00020444444444444446,
"loss": 2.9441,
"step": 158000
},
{
"epoch": 44.14,
"learning_rate": 0.00020333333333333333,
"learning_rate_embeddings": 0.00020333333333333333,
"loss": 2.9241,
"step": 158500
},
{
"epoch": 44.28,
"learning_rate": 0.00020222222222222223,
"learning_rate_embeddings": 0.00020222222222222223,
"loss": 2.9266,
"step": 159000
},
{
"epoch": 44.42,
"learning_rate": 0.0002011111111111111,
"learning_rate_embeddings": 0.0002011111111111111,
"loss": 2.9326,
"step": 159500
},
{
"epoch": 44.56,
"learning_rate": 0.0002,
"learning_rate_embeddings": 0.0002,
"loss": 2.9406,
"step": 160000
},
{
"epoch": 44.7,
"learning_rate": 0.00019888888888888888,
"learning_rate_embeddings": 0.00019888888888888888,
"loss": 2.9357,
"step": 160500
},
{
"epoch": 44.83,
"learning_rate": 0.00019777777777777778,
"learning_rate_embeddings": 0.00019777777777777778,
"loss": 2.9355,
"step": 161000
},
{
"epoch": 44.97,
"learning_rate": 0.00019666666666666666,
"learning_rate_embeddings": 0.00019666666666666666,
"loss": 2.945,
"step": 161500
},
{
"epoch": 45.11,
"learning_rate": 0.00019555555555555556,
"learning_rate_embeddings": 0.00019555555555555556,
"loss": 2.9329,
"step": 162000
},
{
"epoch": 45.25,
"learning_rate": 0.00019444444444444446,
"learning_rate_embeddings": 0.00019444444444444446,
"loss": 2.9273,
"step": 162500
},
{
"epoch": 45.25,
"eval_loss": 3.0122811794281006,
"eval_runtime": 372.1511,
"eval_samples_per_second": 495.906,
"eval_steps_per_second": 3.875,
"step": 162500
},
{
"epoch": 45.39,
"learning_rate": 0.00019333333333333333,
"learning_rate_embeddings": 0.00019333333333333333,
"loss": 2.9298,
"step": 163000
},
{
"epoch": 45.53,
"learning_rate": 0.0001922222222222222,
"learning_rate_embeddings": 0.0001922222222222222,
"loss": 2.9366,
"step": 163500
},
{
"epoch": 45.67,
"learning_rate": 0.00019111111111111114,
"learning_rate_embeddings": 0.00019111111111111114,
"loss": 2.9322,
"step": 164000
},
{
"epoch": 45.81,
"learning_rate": 0.00019,
"learning_rate_embeddings": 0.00019,
"loss": 2.9347,
"step": 164500
},
{
"epoch": 45.95,
"learning_rate": 0.00018888888888888888,
"learning_rate_embeddings": 0.00018888888888888888,
"loss": 2.9404,
"step": 165000
},
{
"epoch": 46.09,
"learning_rate": 0.00018777777777777776,
"learning_rate_embeddings": 0.00018777777777777776,
"loss": 2.9306,
"step": 165500
},
{
"epoch": 46.23,
"learning_rate": 0.0001866666666666667,
"learning_rate_embeddings": 0.0001866666666666667,
"loss": 2.9237,
"step": 166000
},
{
"epoch": 46.37,
"learning_rate": 0.00018555555555555556,
"learning_rate_embeddings": 0.00018555555555555556,
"loss": 2.9236,
"step": 166500
},
{
"epoch": 46.51,
"learning_rate": 0.00018444444444444443,
"learning_rate_embeddings": 0.00018444444444444443,
"loss": 2.9265,
"step": 167000
},
{
"epoch": 46.64,
"learning_rate": 0.00018333333333333334,
"learning_rate_embeddings": 0.00018333333333333334,
"loss": 2.9376,
"step": 167500
},
{
"epoch": 46.78,
"learning_rate": 0.00018222222222222224,
"learning_rate_embeddings": 0.00018222222222222224,
"loss": 2.9279,
"step": 168000
},
{
"epoch": 46.92,
"learning_rate": 0.0001811111111111111,
"learning_rate_embeddings": 0.0001811111111111111,
"loss": 2.9361,
"step": 168500
},
{
"epoch": 47.06,
"learning_rate": 0.00017999999999999998,
"learning_rate_embeddings": 0.00017999999999999998,
"loss": 2.935,
"step": 169000
},
{
"epoch": 47.2,
"learning_rate": 0.00017888888888888889,
"learning_rate_embeddings": 0.00017888888888888889,
"loss": 2.922,
"step": 169500
},
{
"epoch": 47.34,
"learning_rate": 0.00017777777777777779,
"learning_rate_embeddings": 0.00017777777777777779,
"loss": 2.9212,
"step": 170000
},
{
"epoch": 47.48,
"learning_rate": 0.00017666666666666666,
"learning_rate_embeddings": 0.00017666666666666666,
"loss": 2.9281,
"step": 170500
},
{
"epoch": 47.62,
"learning_rate": 0.00017555555555555556,
"learning_rate_embeddings": 0.00017555555555555556,
"loss": 2.9334,
"step": 171000
},
{
"epoch": 47.76,
"learning_rate": 0.00017444444444444446,
"learning_rate_embeddings": 0.00017444444444444446,
"loss": 2.9288,
"step": 171500
},
{
"epoch": 47.9,
"learning_rate": 0.00017333333333333334,
"learning_rate_embeddings": 0.00017333333333333334,
"loss": 2.9359,
"step": 172000
},
{
"epoch": 48.04,
"learning_rate": 0.00017222222222222224,
"learning_rate_embeddings": 0.00017222222222222224,
"loss": 2.9363,
"step": 172500
},
{
"epoch": 48.18,
"learning_rate": 0.0001711111111111111,
"learning_rate_embeddings": 0.0001711111111111111,
"loss": 2.9184,
"step": 173000
},
{
"epoch": 48.32,
"learning_rate": 0.00017,
"learning_rate_embeddings": 0.00017,
"loss": 2.93,
"step": 173500
},
{
"epoch": 48.45,
"learning_rate": 0.00016888888888888889,
"learning_rate_embeddings": 0.00016888888888888889,
"loss": 2.9244,
"step": 174000
},
{
"epoch": 48.59,
"learning_rate": 0.0001677777777777778,
"learning_rate_embeddings": 0.0001677777777777778,
"loss": 2.9349,
"step": 174500
},
{
"epoch": 48.73,
"learning_rate": 0.00016666666666666666,
"learning_rate_embeddings": 0.00016666666666666666,
"loss": 2.9239,
"step": 175000
},
{
"epoch": 48.73,
"eval_loss": 3.005641460418701,
"eval_runtime": 371.5929,
"eval_samples_per_second": 496.651,
"eval_steps_per_second": 3.881,
"step": 175000
},
{
"epoch": 48.87,
"learning_rate": 0.00016555555555555556,
"learning_rate_embeddings": 0.00016555555555555556,
"loss": 2.9306,
"step": 175500
},
{
"epoch": 49.01,
"learning_rate": 0.00016444444444444446,
"learning_rate_embeddings": 0.00016444444444444446,
"loss": 2.9388,
"step": 176000
},
{
"epoch": 49.15,
"learning_rate": 0.00016333333333333334,
"learning_rate_embeddings": 0.00016333333333333334,
"loss": 2.9193,
"step": 176500
},
{
"epoch": 49.29,
"learning_rate": 0.0001622222222222222,
"learning_rate_embeddings": 0.0001622222222222222,
"loss": 2.9203,
"step": 177000
},
{
"epoch": 49.43,
"learning_rate": 0.0001611111111111111,
"learning_rate_embeddings": 0.0001611111111111111,
"loss": 2.9198,
"step": 177500
},
{
"epoch": 49.57,
"learning_rate": 0.00016,
"learning_rate_embeddings": 0.00016,
"loss": 2.9254,
"step": 178000
},
{
"epoch": 49.71,
"learning_rate": 0.0001588888888888889,
"learning_rate_embeddings": 0.0001588888888888889,
"loss": 2.9323,
"step": 178500
},
{
"epoch": 49.85,
"learning_rate": 0.00015777777777777776,
"learning_rate_embeddings": 0.00015777777777777776,
"loss": 2.9322,
"step": 179000
},
{
"epoch": 49.99,
"learning_rate": 0.0001566666666666667,
"learning_rate_embeddings": 0.0001566666666666667,
"loss": 2.9324,
"step": 179500
},
{
"epoch": 50.13,
"learning_rate": 0.00015555555555555556,
"learning_rate_embeddings": 0.00015555555555555556,
"loss": 2.9181,
"step": 180000
},
{
"epoch": 50.26,
"learning_rate": 0.00015444444444444444,
"learning_rate_embeddings": 0.00015444444444444444,
"loss": 2.9278,
"step": 180500
},
{
"epoch": 50.4,
"learning_rate": 0.00015333333333333334,
"learning_rate_embeddings": 0.00015333333333333334,
"loss": 2.9298,
"step": 181000
},
{
"epoch": 50.54,
"learning_rate": 0.00015222222222222224,
"learning_rate_embeddings": 0.00015222222222222224,
"loss": 2.9277,
"step": 181500
},
{
"epoch": 50.68,
"learning_rate": 0.0001511111111111111,
"learning_rate_embeddings": 0.0001511111111111111,
"loss": 2.9312,
"step": 182000
},
{
"epoch": 50.82,
"learning_rate": 0.00015,
"learning_rate_embeddings": 0.00015,
"loss": 2.9307,
"step": 182500
},
{
"epoch": 50.96,
"learning_rate": 0.0001488888888888889,
"learning_rate_embeddings": 0.0001488888888888889,
"loss": 2.9271,
"step": 183000
},
{
"epoch": 51.1,
"learning_rate": 0.0001477777777777778,
"learning_rate_embeddings": 0.0001477777777777778,
"loss": 2.9173,
"step": 183500
},
{
"epoch": 51.24,
"learning_rate": 0.00014666666666666666,
"learning_rate_embeddings": 0.00014666666666666666,
"loss": 2.9242,
"step": 184000
},
{
"epoch": 51.38,
"learning_rate": 0.00014555555555555556,
"learning_rate_embeddings": 0.00014555555555555556,
"loss": 2.931,
"step": 184500
},
{
"epoch": 51.52,
"learning_rate": 0.00014444444444444444,
"learning_rate_embeddings": 0.00014444444444444444,
"loss": 2.9222,
"step": 185000
},
{
"epoch": 51.66,
"learning_rate": 0.00014333333333333334,
"learning_rate_embeddings": 0.00014333333333333334,
"loss": 2.9242,
"step": 185500
},
{
"epoch": 51.8,
"learning_rate": 0.0001422222222222222,
"learning_rate_embeddings": 0.0001422222222222222,
"loss": 2.9271,
"step": 186000
},
{
"epoch": 51.94,
"learning_rate": 0.00014111111111111111,
"learning_rate_embeddings": 0.00014111111111111111,
"loss": 2.9257,
"step": 186500
},
{
"epoch": 52.07,
"learning_rate": 0.00014000000000000001,
"learning_rate_embeddings": 0.00014000000000000001,
"loss": 2.925,
"step": 187000
},
{
"epoch": 52.21,
"learning_rate": 0.0001388888888888889,
"learning_rate_embeddings": 0.0001388888888888889,
"loss": 2.9166,
"step": 187500
},
{
"epoch": 52.21,
"eval_loss": 3.0037143230438232,
"eval_runtime": 371.8941,
"eval_samples_per_second": 496.249,
"eval_steps_per_second": 3.877,
"step": 187500
},
{
"epoch": 52.35,
"learning_rate": 0.0001377777777777778,
"learning_rate_embeddings": 0.0001377777777777778,
"loss": 2.9189,
"step": 188000
},
{
"epoch": 52.49,
"learning_rate": 0.00013666666666666666,
"learning_rate_embeddings": 0.00013666666666666666,
"loss": 2.9241,
"step": 188500
},
{
"epoch": 52.63,
"learning_rate": 0.00013555555555555556,
"learning_rate_embeddings": 0.00013555555555555556,
"loss": 2.9269,
"step": 189000
},
{
"epoch": 52.77,
"learning_rate": 0.00013444444444444447,
"learning_rate_embeddings": 0.00013444444444444447,
"loss": 2.9249,
"step": 189500
},
{
"epoch": 52.91,
"learning_rate": 0.00013333333333333334,
"learning_rate_embeddings": 0.00013333333333333334,
"loss": 2.9311,
"step": 190000
},
{
"epoch": 53.05,
"learning_rate": 0.00013222222222222221,
"learning_rate_embeddings": 0.00013222222222222221,
"loss": 2.9271,
"step": 190500
},
{
"epoch": 53.19,
"learning_rate": 0.00013111111111111111,
"learning_rate_embeddings": 0.00013111111111111111,
"loss": 2.9154,
"step": 191000
},
{
"epoch": 53.33,
"learning_rate": 0.00013000000000000002,
"learning_rate_embeddings": 0.00013000000000000002,
"loss": 2.9217,
"step": 191500
},
{
"epoch": 53.47,
"learning_rate": 0.0001288888888888889,
"learning_rate_embeddings": 0.0001288888888888889,
"loss": 2.9171,
"step": 192000
},
{
"epoch": 53.61,
"learning_rate": 0.00012777777777777776,
"learning_rate_embeddings": 0.00012777777777777776,
"loss": 2.9267,
"step": 192500
},
{
"epoch": 53.75,
"learning_rate": 0.0001266666666666667,
"learning_rate_embeddings": 0.0001266666666666667,
"loss": 2.9191,
"step": 193000
},
{
"epoch": 53.88,
"learning_rate": 0.00012555555555555557,
"learning_rate_embeddings": 0.00012555555555555557,
"loss": 2.9272,
"step": 193500
},
{
"epoch": 54.02,
"learning_rate": 0.00012444444444444444,
"learning_rate_embeddings": 0.00012444444444444444,
"loss": 2.9261,
"step": 194000
},
{
"epoch": 54.16,
"learning_rate": 0.00012333333333333334,
"learning_rate_embeddings": 0.00012333333333333334,
"loss": 2.9099,
"step": 194500
},
{
"epoch": 54.3,
"learning_rate": 0.00012222222222222221,
"learning_rate_embeddings": 0.00012222222222222221,
"loss": 2.9197,
"step": 195000
},
{
"epoch": 54.44,
"learning_rate": 0.00012111111111111112,
"learning_rate_embeddings": 0.00012111111111111112,
"loss": 2.9187,
"step": 195500
},
{
"epoch": 54.58,
"learning_rate": 0.00012,
"learning_rate_embeddings": 0.00012,
"loss": 2.9207,
"step": 196000
},
{
"epoch": 54.72,
"learning_rate": 0.00011888888888888889,
"learning_rate_embeddings": 0.00011888888888888889,
"loss": 2.927,
"step": 196500
},
{
"epoch": 54.86,
"learning_rate": 0.00011777777777777778,
"learning_rate_embeddings": 0.00011777777777777778,
"loss": 2.9222,
"step": 197000
},
{
"epoch": 55.0,
"learning_rate": 0.00011666666666666667,
"learning_rate_embeddings": 0.00011666666666666667,
"loss": 2.925,
"step": 197500
},
{
"epoch": 55.14,
"learning_rate": 0.00011555555555555555,
"learning_rate_embeddings": 0.00011555555555555555,
"loss": 2.9116,
"step": 198000
},
{
"epoch": 55.28,
"learning_rate": 0.00011444444444444445,
"learning_rate_embeddings": 0.00011444444444444445,
"loss": 2.9173,
"step": 198500
},
{
"epoch": 55.42,
"learning_rate": 0.00011333333333333333,
"learning_rate_embeddings": 0.00011333333333333333,
"loss": 2.9176,
"step": 199000
},
{
"epoch": 55.56,
"learning_rate": 0.00011222222222222223,
"learning_rate_embeddings": 0.00011222222222222223,
"loss": 2.9137,
"step": 199500
},
{
"epoch": 55.69,
"learning_rate": 0.0001111111111111111,
"learning_rate_embeddings": 0.0001111111111111111,
"loss": 2.9225,
"step": 200000
},
{
"epoch": 55.69,
"eval_loss": 3.000563621520996,
"eval_runtime": 371.9938,
"eval_samples_per_second": 496.116,
"eval_steps_per_second": 3.876,
"step": 200000
},
{
"epoch": 55.83,
"learning_rate": 0.00011,
"learning_rate_embeddings": 0.00011,
"loss": 2.9211,
"step": 200500
},
{
"epoch": 55.97,
"learning_rate": 0.00010888888888888888,
"learning_rate_embeddings": 0.00010888888888888888,
"loss": 2.9202,
"step": 201000
},
{
"epoch": 56.11,
"learning_rate": 0.00010777777777777778,
"learning_rate_embeddings": 0.00010777777777777778,
"loss": 2.916,
"step": 201500
},
{
"epoch": 56.25,
"learning_rate": 0.00010666666666666668,
"learning_rate_embeddings": 0.00010666666666666668,
"loss": 2.9217,
"step": 202000
},
{
"epoch": 56.39,
"learning_rate": 0.00010555555555555555,
"learning_rate_embeddings": 0.00010555555555555555,
"loss": 2.9111,
"step": 202500
},
{
"epoch": 56.53,
"learning_rate": 0.00010444444444444445,
"learning_rate_embeddings": 0.00010444444444444445,
"loss": 2.916,
"step": 203000
},
{
"epoch": 56.67,
"learning_rate": 0.00010333333333333333,
"learning_rate_embeddings": 0.00010333333333333333,
"loss": 2.9217,
"step": 203500
},
{
"epoch": 56.81,
"learning_rate": 0.00010222222222222223,
"learning_rate_embeddings": 0.00010222222222222223,
"loss": 2.9174,
"step": 204000
},
{
"epoch": 56.95,
"learning_rate": 0.00010111111111111112,
"learning_rate_embeddings": 0.00010111111111111112,
"loss": 2.9179,
"step": 204500
},
{
"epoch": 57.09,
"learning_rate": 0.0001,
"learning_rate_embeddings": 0.0001,
"loss": 2.9108,
"step": 205000
},
{
"epoch": 57.23,
"learning_rate": 9.888888888888889e-05,
"learning_rate_embeddings": 9.888888888888889e-05,
"loss": 2.9113,
"step": 205500
},
{
"epoch": 57.37,
"learning_rate": 9.777777777777778e-05,
"learning_rate_embeddings": 9.777777777777778e-05,
"loss": 2.9194,
"step": 206000
},
{
"epoch": 57.5,
"learning_rate": 9.666666666666667e-05,
"learning_rate_embeddings": 9.666666666666667e-05,
"loss": 2.9101,
"step": 206500
},
{
"epoch": 57.64,
"learning_rate": 9.555555555555557e-05,
"learning_rate_embeddings": 9.555555555555557e-05,
"loss": 2.9138,
"step": 207000
},
{
"epoch": 57.78,
"learning_rate": 9.444444444444444e-05,
"learning_rate_embeddings": 9.444444444444444e-05,
"loss": 2.9161,
"step": 207500
},
{
"epoch": 57.92,
"learning_rate": 9.333333333333334e-05,
"learning_rate_embeddings": 9.333333333333334e-05,
"loss": 2.9162,
"step": 208000
},
{
"epoch": 58.06,
"learning_rate": 9.222222222222222e-05,
"learning_rate_embeddings": 9.222222222222222e-05,
"loss": 2.9065,
"step": 208500
},
{
"epoch": 58.2,
"learning_rate": 9.111111111111112e-05,
"learning_rate_embeddings": 9.111111111111112e-05,
"loss": 2.9155,
"step": 209000
},
{
"epoch": 58.34,
"learning_rate": 8.999999999999999e-05,
"learning_rate_embeddings": 8.999999999999999e-05,
"loss": 2.9155,
"step": 209500
},
{
"epoch": 58.48,
"learning_rate": 8.888888888888889e-05,
"learning_rate_embeddings": 8.888888888888889e-05,
"loss": 2.9149,
"step": 210000
},
{
"epoch": 58.62,
"learning_rate": 8.777777777777778e-05,
"learning_rate_embeddings": 8.777777777777778e-05,
"loss": 2.913,
"step": 210500
},
{
"epoch": 58.76,
"learning_rate": 8.666666666666667e-05,
"learning_rate_embeddings": 8.666666666666667e-05,
"loss": 2.9152,
"step": 211000
},
{
"epoch": 58.9,
"learning_rate": 8.555555555555556e-05,
"learning_rate_embeddings": 8.555555555555556e-05,
"loss": 2.9119,
"step": 211500
},
{
"epoch": 59.04,
"learning_rate": 8.444444444444444e-05,
"learning_rate_embeddings": 8.444444444444444e-05,
"loss": 2.9173,
"step": 212000
},
{
"epoch": 59.18,
"learning_rate": 8.333333333333333e-05,
"learning_rate_embeddings": 8.333333333333333e-05,
"loss": 2.9085,
"step": 212500
},
{
"epoch": 59.18,
"eval_loss": 2.9969003200531006,
"eval_runtime": 371.9315,
"eval_samples_per_second": 496.199,
"eval_steps_per_second": 3.877,
"step": 212500
},
{
"epoch": 59.31,
"learning_rate": 8.222222222222223e-05,
"learning_rate_embeddings": 8.222222222222223e-05,
"loss": 2.9039,
"step": 213000
},
{
"epoch": 59.45,
"learning_rate": 8.11111111111111e-05,
"learning_rate_embeddings": 8.11111111111111e-05,
"loss": 2.9117,
"step": 213500
},
{
"epoch": 59.59,
"learning_rate": 8e-05,
"learning_rate_embeddings": 8e-05,
"loss": 2.9157,
"step": 214000
},
{
"epoch": 59.73,
"learning_rate": 7.888888888888888e-05,
"learning_rate_embeddings": 7.888888888888888e-05,
"loss": 2.9101,
"step": 214500
},
{
"epoch": 59.87,
"learning_rate": 7.777777777777778e-05,
"learning_rate_embeddings": 7.777777777777778e-05,
"loss": 2.9161,
"step": 215000
},
{
"epoch": 60.01,
"learning_rate": 7.666666666666667e-05,
"learning_rate_embeddings": 7.666666666666667e-05,
"loss": 2.9101,
"step": 215500
},
{
"epoch": 60.15,
"learning_rate": 7.555555555555556e-05,
"learning_rate_embeddings": 7.555555555555556e-05,
"loss": 2.9074,
"step": 216000
},
{
"epoch": 60.29,
"learning_rate": 7.444444444444444e-05,
"learning_rate_embeddings": 7.444444444444444e-05,
"loss": 2.9081,
"step": 216500
},
{
"epoch": 60.43,
"learning_rate": 7.333333333333333e-05,
"learning_rate_embeddings": 7.333333333333333e-05,
"loss": 2.9121,
"step": 217000
},
{
"epoch": 60.57,
"learning_rate": 7.222222222222222e-05,
"learning_rate_embeddings": 7.222222222222222e-05,
"loss": 2.907,
"step": 217500
},
{
"epoch": 60.71,
"learning_rate": 7.11111111111111e-05,
"learning_rate_embeddings": 7.11111111111111e-05,
"loss": 2.9098,
"step": 218000
},
{
"epoch": 60.85,
"learning_rate": 7.000000000000001e-05,
"learning_rate_embeddings": 7.000000000000001e-05,
"loss": 2.9116,
"step": 218500
},
{
"epoch": 60.99,
"learning_rate": 6.88888888888889e-05,
"learning_rate_embeddings": 6.88888888888889e-05,
"loss": 2.9078,
"step": 219000
},
{
"epoch": 61.13,
"learning_rate": 6.777777777777778e-05,
"learning_rate_embeddings": 6.777777777777778e-05,
"loss": 2.9076,
"step": 219500
},
{
"epoch": 61.26,
"learning_rate": 6.666666666666667e-05,
"learning_rate_embeddings": 6.666666666666667e-05,
"loss": 2.9083,
"step": 220000
},
{
"epoch": 61.4,
"learning_rate": 6.555555555555556e-05,
"learning_rate_embeddings": 6.555555555555556e-05,
"loss": 2.909,
"step": 220500
},
{
"epoch": 61.54,
"learning_rate": 6.444444444444444e-05,
"learning_rate_embeddings": 6.444444444444444e-05,
"loss": 2.9109,
"step": 221000
},
{
"epoch": 61.68,
"learning_rate": 6.333333333333335e-05,
"learning_rate_embeddings": 6.333333333333335e-05,
"loss": 2.9115,
"step": 221500
},
{
"epoch": 61.82,
"learning_rate": 6.222222222222222e-05,
"learning_rate_embeddings": 6.222222222222222e-05,
"loss": 2.9083,
"step": 222000
},
{
"epoch": 61.96,
"learning_rate": 6.111111111111111e-05,
"learning_rate_embeddings": 6.111111111111111e-05,
"loss": 2.9076,
"step": 222500
},
{
"epoch": 62.1,
"learning_rate": 6e-05,
"learning_rate_embeddings": 6e-05,
"loss": 2.9018,
"step": 223000
},
{
"epoch": 62.24,
"learning_rate": 5.888888888888889e-05,
"learning_rate_embeddings": 5.888888888888889e-05,
"loss": 2.9016,
"step": 223500
},
{
"epoch": 62.38,
"learning_rate": 5.7777777777777776e-05,
"learning_rate_embeddings": 5.7777777777777776e-05,
"loss": 2.9047,
"step": 224000
},
{
"epoch": 62.52,
"learning_rate": 5.6666666666666664e-05,
"learning_rate_embeddings": 5.6666666666666664e-05,
"loss": 2.9047,
"step": 224500
},
{
"epoch": 62.66,
"learning_rate": 5.555555555555555e-05,
"learning_rate_embeddings": 5.555555555555555e-05,
"loss": 2.9084,
"step": 225000
},
{
"epoch": 62.66,
"eval_loss": 2.994610548019409,
"eval_runtime": 371.8603,
"eval_samples_per_second": 496.294,
"eval_steps_per_second": 3.878,
"step": 225000
},
{
"epoch": 62.8,
"learning_rate": 5.444444444444444e-05,
"learning_rate_embeddings": 5.444444444444444e-05,
"loss": 2.9048,
"step": 225500
},
{
"epoch": 62.94,
"learning_rate": 5.333333333333334e-05,
"learning_rate_embeddings": 5.333333333333334e-05,
"loss": 2.9025,
"step": 226000
},
{
"epoch": 63.07,
"learning_rate": 5.222222222222223e-05,
"learning_rate_embeddings": 5.222222222222223e-05,
"loss": 2.9048,
"step": 226500
},
{
"epoch": 63.21,
"learning_rate": 5.1111111111111115e-05,
"learning_rate_embeddings": 5.1111111111111115e-05,
"loss": 2.906,
"step": 227000
},
{
"epoch": 63.35,
"learning_rate": 5e-05,
"learning_rate_embeddings": 5e-05,
"loss": 2.9042,
"step": 227500
},
{
"epoch": 63.49,
"learning_rate": 4.888888888888889e-05,
"learning_rate_embeddings": 4.888888888888889e-05,
"loss": 2.9018,
"step": 228000
},
{
"epoch": 63.63,
"learning_rate": 4.7777777777777784e-05,
"learning_rate_embeddings": 4.7777777777777784e-05,
"loss": 2.9103,
"step": 228500
},
{
"epoch": 63.77,
"learning_rate": 4.666666666666667e-05,
"learning_rate_embeddings": 4.666666666666667e-05,
"loss": 2.9079,
"step": 229000
},
{
"epoch": 63.91,
"learning_rate": 4.555555555555556e-05,
"learning_rate_embeddings": 4.555555555555556e-05,
"loss": 2.9017,
"step": 229500
},
{
"epoch": 64.05,
"learning_rate": 4.4444444444444447e-05,
"learning_rate_embeddings": 4.4444444444444447e-05,
"loss": 2.9066,
"step": 230000
},
{
"epoch": 64.19,
"learning_rate": 4.3333333333333334e-05,
"learning_rate_embeddings": 4.3333333333333334e-05,
"loss": 2.9032,
"step": 230500
},
{
"epoch": 64.33,
"learning_rate": 4.222222222222222e-05,
"learning_rate_embeddings": 4.222222222222222e-05,
"loss": 2.8985,
"step": 231000
},
{
"epoch": 64.47,
"learning_rate": 4.1111111111111116e-05,
"learning_rate_embeddings": 4.1111111111111116e-05,
"loss": 2.8984,
"step": 231500
},
{
"epoch": 64.61,
"learning_rate": 4e-05,
"learning_rate_embeddings": 4e-05,
"loss": 2.8997,
"step": 232000
},
{
"epoch": 64.75,
"learning_rate": 3.888888888888889e-05,
"learning_rate_embeddings": 3.888888888888889e-05,
"loss": 2.9084,
"step": 232500
},
{
"epoch": 64.88,
"learning_rate": 3.777777777777778e-05,
"learning_rate_embeddings": 3.777777777777778e-05,
"loss": 2.9057,
"step": 233000
},
{
"epoch": 65.02,
"learning_rate": 3.6666666666666666e-05,
"learning_rate_embeddings": 3.6666666666666666e-05,
"loss": 2.9019,
"step": 233500
},
{
"epoch": 65.16,
"learning_rate": 3.555555555555555e-05,
"learning_rate_embeddings": 3.555555555555555e-05,
"loss": 2.9003,
"step": 234000
},
{
"epoch": 65.3,
"learning_rate": 3.444444444444445e-05,
"learning_rate_embeddings": 3.444444444444445e-05,
"loss": 2.8984,
"step": 234500
},
{
"epoch": 65.44,
"learning_rate": 3.3333333333333335e-05,
"learning_rate_embeddings": 3.3333333333333335e-05,
"loss": 2.8975,
"step": 235000
},
{
"epoch": 65.58,
"learning_rate": 3.222222222222222e-05,
"learning_rate_embeddings": 3.222222222222222e-05,
"loss": 2.9005,
"step": 235500
},
{
"epoch": 65.72,
"learning_rate": 3.111111111111111e-05,
"learning_rate_embeddings": 3.111111111111111e-05,
"loss": 2.901,
"step": 236000
},
{
"epoch": 65.86,
"learning_rate": 3e-05,
"learning_rate_embeddings": 3e-05,
"loss": 2.9041,
"step": 236500
},
{
"epoch": 66.0,
"learning_rate": 2.8888888888888888e-05,
"learning_rate_embeddings": 2.8888888888888888e-05,
"loss": 2.8981,
"step": 237000
},
{
"epoch": 66.14,
"learning_rate": 2.7777777777777776e-05,
"learning_rate_embeddings": 2.7777777777777776e-05,
"loss": 2.9033,
"step": 237500
},
{
"epoch": 66.14,
"eval_loss": 2.990591287612915,
"eval_runtime": 371.7717,
"eval_samples_per_second": 496.412,
"eval_steps_per_second": 3.879,
"step": 237500
},
{
"epoch": 66.28,
"learning_rate": 2.666666666666667e-05,
"learning_rate_embeddings": 2.666666666666667e-05,
"loss": 2.898,
"step": 238000
},
{
"epoch": 66.42,
"learning_rate": 2.5555555555555557e-05,
"learning_rate_embeddings": 2.5555555555555557e-05,
"loss": 2.9022,
"step": 238500
},
{
"epoch": 66.56,
"learning_rate": 2.4444444444444445e-05,
"learning_rate_embeddings": 2.4444444444444445e-05,
"loss": 2.8963,
"step": 239000
},
{
"epoch": 66.69,
"learning_rate": 2.3333333333333336e-05,
"learning_rate_embeddings": 2.3333333333333336e-05,
"loss": 2.9019,
"step": 239500
},
{
"epoch": 66.83,
"learning_rate": 2.2222222222222223e-05,
"learning_rate_embeddings": 2.2222222222222223e-05,
"loss": 2.8966,
"step": 240000
},
{
"epoch": 66.97,
"learning_rate": 2.111111111111111e-05,
"learning_rate_embeddings": 2.111111111111111e-05,
"loss": 2.8934,
"step": 240500
},
{
"epoch": 67.11,
"learning_rate": 2e-05,
"learning_rate_embeddings": 2e-05,
"loss": 2.8972,
"step": 241000
},
{
"epoch": 67.25,
"learning_rate": 1.888888888888889e-05,
"learning_rate_embeddings": 1.888888888888889e-05,
"loss": 2.8937,
"step": 241500
},
{
"epoch": 67.39,
"learning_rate": 1.7777777777777777e-05,
"learning_rate_embeddings": 1.7777777777777777e-05,
"loss": 2.8924,
"step": 242000
},
{
"epoch": 67.53,
"learning_rate": 1.6666666666666667e-05,
"learning_rate_embeddings": 1.6666666666666667e-05,
"loss": 2.8954,
"step": 242500
},
{
"epoch": 67.67,
"learning_rate": 1.5555555555555555e-05,
"learning_rate_embeddings": 1.5555555555555555e-05,
"loss": 2.8983,
"step": 243000
},
{
"epoch": 67.81,
"learning_rate": 1.4444444444444444e-05,
"learning_rate_embeddings": 1.4444444444444444e-05,
"loss": 2.9004,
"step": 243500
},
{
"epoch": 67.95,
"learning_rate": 1.3333333333333335e-05,
"learning_rate_embeddings": 1.3333333333333335e-05,
"loss": 2.9029,
"step": 244000
},
{
"epoch": 68.09,
"learning_rate": 1.2222222222222222e-05,
"learning_rate_embeddings": 1.2222222222222222e-05,
"loss": 2.8969,
"step": 244500
},
{
"epoch": 68.23,
"learning_rate": 1.1111111111111112e-05,
"learning_rate_embeddings": 1.1111111111111112e-05,
"loss": 2.8966,
"step": 245000
},
{
"epoch": 68.37,
"learning_rate": 1e-05,
"learning_rate_embeddings": 1e-05,
"loss": 2.895,
"step": 245500
},
{
"epoch": 68.5,
"learning_rate": 8.888888888888888e-06,
"learning_rate_embeddings": 8.888888888888888e-06,
"loss": 2.8982,
"step": 246000
},
{
"epoch": 68.64,
"learning_rate": 7.777777777777777e-06,
"learning_rate_embeddings": 7.777777777777777e-06,
"loss": 2.8934,
"step": 246500
},
{
"epoch": 68.78,
"learning_rate": 6.6666666666666675e-06,
"learning_rate_embeddings": 6.6666666666666675e-06,
"loss": 2.8999,
"step": 247000
},
{
"epoch": 68.92,
"learning_rate": 5.555555555555556e-06,
"learning_rate_embeddings": 5.555555555555556e-06,
"loss": 2.9024,
"step": 247500
},
{
"epoch": 69.06,
"learning_rate": 4.444444444444444e-06,
"learning_rate_embeddings": 4.444444444444444e-06,
"loss": 2.8982,
"step": 248000
},
{
"epoch": 69.2,
"learning_rate": 3.3333333333333337e-06,
"learning_rate_embeddings": 3.3333333333333337e-06,
"loss": 2.8987,
"step": 248500
},
{
"epoch": 69.34,
"learning_rate": 2.222222222222222e-06,
"learning_rate_embeddings": 2.222222222222222e-06,
"loss": 2.8941,
"step": 249000
},
{
"epoch": 69.48,
"learning_rate": 1.111111111111111e-06,
"learning_rate_embeddings": 1.111111111111111e-06,
"loss": 2.8917,
"step": 249500
},
{
"epoch": 69.62,
"learning_rate": 0.0,
"learning_rate_embeddings": 0.0,
"loss": 2.8921,
"step": 250000
},
{
"epoch": 69.62,
"eval_loss": 2.9811859130859375,
"eval_runtime": 371.5712,
"eval_samples_per_second": 496.68,
"eval_steps_per_second": 3.881,
"step": 250000
}
],
"max_steps": 250000,
"num_train_epochs": 70,
"total_flos": 4.180960978403328e+18,
"trial_name": null,
"trial_params": null
}