NonsenseUpdateDiffStringBart / trainer_state.json
hyesunyun's picture
Update from yun.hy
9f570f1
raw
history blame
11.7 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"global_step": 50000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 4.9500000000000004e-05,
"loss": 2.2718,
"step": 500
},
{
"epoch": 0.02,
"learning_rate": 4.9e-05,
"loss": 1.2406,
"step": 1000
},
{
"epoch": 0.03,
"learning_rate": 4.85e-05,
"loss": 1.1883,
"step": 1500
},
{
"epoch": 0.04,
"learning_rate": 4.8e-05,
"loss": 1.1338,
"step": 2000
},
{
"epoch": 0.05,
"learning_rate": 4.75e-05,
"loss": 1.1181,
"step": 2500
},
{
"epoch": 0.06,
"learning_rate": 4.7e-05,
"loss": 1.0929,
"step": 3000
},
{
"epoch": 0.07,
"learning_rate": 4.6500000000000005e-05,
"loss": 1.0919,
"step": 3500
},
{
"epoch": 0.08,
"learning_rate": 4.600000000000001e-05,
"loss": 1.0702,
"step": 4000
},
{
"epoch": 0.09,
"learning_rate": 4.55e-05,
"loss": 1.0663,
"step": 4500
},
{
"epoch": 0.1,
"learning_rate": 4.5e-05,
"loss": 1.0498,
"step": 5000
},
{
"epoch": 0.11,
"learning_rate": 4.4500000000000004e-05,
"loss": 1.0621,
"step": 5500
},
{
"epoch": 0.12,
"learning_rate": 4.4000000000000006e-05,
"loss": 1.0597,
"step": 6000
},
{
"epoch": 0.13,
"learning_rate": 4.35e-05,
"loss": 1.0455,
"step": 6500
},
{
"epoch": 0.14,
"learning_rate": 4.3e-05,
"loss": 1.0512,
"step": 7000
},
{
"epoch": 0.15,
"learning_rate": 4.25e-05,
"loss": 1.0517,
"step": 7500
},
{
"epoch": 0.16,
"learning_rate": 4.2e-05,
"loss": 1.0448,
"step": 8000
},
{
"epoch": 0.17,
"learning_rate": 4.15e-05,
"loss": 1.0406,
"step": 8500
},
{
"epoch": 0.18,
"learning_rate": 4.1e-05,
"loss": 1.0402,
"step": 9000
},
{
"epoch": 0.19,
"learning_rate": 4.05e-05,
"loss": 1.0482,
"step": 9500
},
{
"epoch": 0.2,
"learning_rate": 4e-05,
"loss": 1.0263,
"step": 10000
},
{
"epoch": 0.21,
"learning_rate": 3.9500000000000005e-05,
"loss": 1.0524,
"step": 10500
},
{
"epoch": 0.22,
"learning_rate": 3.9000000000000006e-05,
"loss": 1.0331,
"step": 11000
},
{
"epoch": 0.23,
"learning_rate": 3.85e-05,
"loss": 1.0416,
"step": 11500
},
{
"epoch": 0.24,
"learning_rate": 3.8e-05,
"loss": 1.0463,
"step": 12000
},
{
"epoch": 0.25,
"learning_rate": 3.7500000000000003e-05,
"loss": 1.0303,
"step": 12500
},
{
"epoch": 0.26,
"learning_rate": 3.7e-05,
"loss": 1.0245,
"step": 13000
},
{
"epoch": 0.27,
"learning_rate": 3.65e-05,
"loss": 1.0283,
"step": 13500
},
{
"epoch": 0.28,
"learning_rate": 3.6e-05,
"loss": 1.0186,
"step": 14000
},
{
"epoch": 0.29,
"learning_rate": 3.55e-05,
"loss": 1.0323,
"step": 14500
},
{
"epoch": 0.3,
"learning_rate": 3.5e-05,
"loss": 1.0303,
"step": 15000
},
{
"epoch": 0.31,
"learning_rate": 3.45e-05,
"loss": 1.0275,
"step": 15500
},
{
"epoch": 0.32,
"learning_rate": 3.4000000000000007e-05,
"loss": 1.0308,
"step": 16000
},
{
"epoch": 0.33,
"learning_rate": 3.35e-05,
"loss": 1.0481,
"step": 16500
},
{
"epoch": 0.34,
"learning_rate": 3.3e-05,
"loss": 1.0371,
"step": 17000
},
{
"epoch": 0.35,
"learning_rate": 3.2500000000000004e-05,
"loss": 1.0364,
"step": 17500
},
{
"epoch": 0.36,
"learning_rate": 3.2000000000000005e-05,
"loss": 1.0359,
"step": 18000
},
{
"epoch": 0.37,
"learning_rate": 3.15e-05,
"loss": 1.0244,
"step": 18500
},
{
"epoch": 0.38,
"learning_rate": 3.1e-05,
"loss": 1.0228,
"step": 19000
},
{
"epoch": 0.39,
"learning_rate": 3.05e-05,
"loss": 1.027,
"step": 19500
},
{
"epoch": 0.4,
"learning_rate": 3e-05,
"loss": 1.0282,
"step": 20000
},
{
"epoch": 0.41,
"learning_rate": 2.95e-05,
"loss": 1.024,
"step": 20500
},
{
"epoch": 0.42,
"learning_rate": 2.9e-05,
"loss": 1.0353,
"step": 21000
},
{
"epoch": 0.43,
"learning_rate": 2.8499999999999998e-05,
"loss": 1.0281,
"step": 21500
},
{
"epoch": 0.44,
"learning_rate": 2.8000000000000003e-05,
"loss": 1.0316,
"step": 22000
},
{
"epoch": 0.45,
"learning_rate": 2.7500000000000004e-05,
"loss": 1.0191,
"step": 22500
},
{
"epoch": 0.46,
"learning_rate": 2.7000000000000002e-05,
"loss": 1.0192,
"step": 23000
},
{
"epoch": 0.47,
"learning_rate": 2.6500000000000004e-05,
"loss": 1.0246,
"step": 23500
},
{
"epoch": 0.48,
"learning_rate": 2.6000000000000002e-05,
"loss": 1.0215,
"step": 24000
},
{
"epoch": 0.49,
"learning_rate": 2.5500000000000003e-05,
"loss": 1.0231,
"step": 24500
},
{
"epoch": 0.5,
"learning_rate": 2.5e-05,
"loss": 1.0218,
"step": 25000
},
{
"epoch": 0.51,
"learning_rate": 2.45e-05,
"loss": 1.0224,
"step": 25500
},
{
"epoch": 0.52,
"learning_rate": 2.4e-05,
"loss": 1.0224,
"step": 26000
},
{
"epoch": 0.53,
"learning_rate": 2.35e-05,
"loss": 1.0256,
"step": 26500
},
{
"epoch": 0.54,
"learning_rate": 2.3000000000000003e-05,
"loss": 1.0225,
"step": 27000
},
{
"epoch": 0.55,
"learning_rate": 2.25e-05,
"loss": 1.029,
"step": 27500
},
{
"epoch": 0.56,
"learning_rate": 2.2000000000000003e-05,
"loss": 1.0223,
"step": 28000
},
{
"epoch": 0.57,
"learning_rate": 2.15e-05,
"loss": 1.0285,
"step": 28500
},
{
"epoch": 0.58,
"learning_rate": 2.1e-05,
"loss": 1.0369,
"step": 29000
},
{
"epoch": 0.59,
"learning_rate": 2.05e-05,
"loss": 1.0221,
"step": 29500
},
{
"epoch": 0.6,
"learning_rate": 2e-05,
"loss": 1.0242,
"step": 30000
},
{
"epoch": 0.61,
"learning_rate": 1.9500000000000003e-05,
"loss": 1.0257,
"step": 30500
},
{
"epoch": 0.62,
"learning_rate": 1.9e-05,
"loss": 1.0194,
"step": 31000
},
{
"epoch": 0.63,
"learning_rate": 1.85e-05,
"loss": 1.0115,
"step": 31500
},
{
"epoch": 0.64,
"learning_rate": 1.8e-05,
"loss": 1.0236,
"step": 32000
},
{
"epoch": 0.65,
"learning_rate": 1.75e-05,
"loss": 1.0298,
"step": 32500
},
{
"epoch": 0.66,
"learning_rate": 1.7000000000000003e-05,
"loss": 1.0247,
"step": 33000
},
{
"epoch": 0.67,
"learning_rate": 1.65e-05,
"loss": 1.03,
"step": 33500
},
{
"epoch": 0.68,
"learning_rate": 1.6000000000000003e-05,
"loss": 1.0107,
"step": 34000
},
{
"epoch": 0.69,
"learning_rate": 1.55e-05,
"loss": 1.0152,
"step": 34500
},
{
"epoch": 0.7,
"learning_rate": 1.5e-05,
"loss": 1.0288,
"step": 35000
},
{
"epoch": 0.71,
"learning_rate": 1.45e-05,
"loss": 1.0216,
"step": 35500
},
{
"epoch": 0.72,
"learning_rate": 1.4000000000000001e-05,
"loss": 1.025,
"step": 36000
},
{
"epoch": 0.73,
"learning_rate": 1.3500000000000001e-05,
"loss": 1.0156,
"step": 36500
},
{
"epoch": 0.74,
"learning_rate": 1.3000000000000001e-05,
"loss": 1.0054,
"step": 37000
},
{
"epoch": 0.75,
"learning_rate": 1.25e-05,
"loss": 1.0218,
"step": 37500
},
{
"epoch": 0.76,
"learning_rate": 1.2e-05,
"loss": 1.0102,
"step": 38000
},
{
"epoch": 0.77,
"learning_rate": 1.1500000000000002e-05,
"loss": 1.0142,
"step": 38500
},
{
"epoch": 0.78,
"learning_rate": 1.1000000000000001e-05,
"loss": 1.0269,
"step": 39000
},
{
"epoch": 0.79,
"learning_rate": 1.05e-05,
"loss": 1.0121,
"step": 39500
},
{
"epoch": 0.8,
"learning_rate": 1e-05,
"loss": 1.0192,
"step": 40000
},
{
"epoch": 0.81,
"learning_rate": 9.5e-06,
"loss": 1.0176,
"step": 40500
},
{
"epoch": 0.82,
"learning_rate": 9e-06,
"loss": 1.0241,
"step": 41000
},
{
"epoch": 0.83,
"learning_rate": 8.500000000000002e-06,
"loss": 1.018,
"step": 41500
},
{
"epoch": 0.84,
"learning_rate": 8.000000000000001e-06,
"loss": 1.0177,
"step": 42000
},
{
"epoch": 0.85,
"learning_rate": 7.5e-06,
"loss": 1.0031,
"step": 42500
},
{
"epoch": 0.86,
"learning_rate": 7.000000000000001e-06,
"loss": 1.0051,
"step": 43000
},
{
"epoch": 0.87,
"learning_rate": 6.5000000000000004e-06,
"loss": 1.0113,
"step": 43500
},
{
"epoch": 0.88,
"learning_rate": 6e-06,
"loss": 1.0138,
"step": 44000
},
{
"epoch": 0.89,
"learning_rate": 5.500000000000001e-06,
"loss": 1.0105,
"step": 44500
},
{
"epoch": 0.9,
"learning_rate": 5e-06,
"loss": 1.0208,
"step": 45000
},
{
"epoch": 0.91,
"learning_rate": 4.5e-06,
"loss": 1.014,
"step": 45500
},
{
"epoch": 0.92,
"learning_rate": 4.000000000000001e-06,
"loss": 1.0215,
"step": 46000
},
{
"epoch": 0.93,
"learning_rate": 3.5000000000000004e-06,
"loss": 1.0187,
"step": 46500
},
{
"epoch": 0.94,
"learning_rate": 3e-06,
"loss": 1.0159,
"step": 47000
},
{
"epoch": 0.95,
"learning_rate": 2.5e-06,
"loss": 1.0084,
"step": 47500
},
{
"epoch": 0.96,
"learning_rate": 2.0000000000000003e-06,
"loss": 1.0141,
"step": 48000
},
{
"epoch": 0.97,
"learning_rate": 1.5e-06,
"loss": 1.0134,
"step": 48500
},
{
"epoch": 0.98,
"learning_rate": 1.0000000000000002e-06,
"loss": 1.0096,
"step": 49000
},
{
"epoch": 0.99,
"learning_rate": 5.000000000000001e-07,
"loss": 1.0005,
"step": 49500
},
{
"epoch": 1.0,
"learning_rate": 0.0,
"loss": 1.0132,
"step": 50000
}
],
"max_steps": 50000,
"num_train_epochs": 1,
"total_flos": 2.167104602112e+17,
"trial_name": null,
"trial_params": null
}