{ "best_metric": 0.4586097490625901, "best_model_checkpoint": "./p6_severe/checkpoint-1000", "epoch": 476.1904761904762, "eval_steps": 1000, "global_step": 5000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 2.38, "learning_rate": 5.000000000000001e-07, "loss": 1.1712, "step": 25 }, { "epoch": 4.76, "learning_rate": 1.0000000000000002e-06, "loss": 0.9121, "step": 50 }, { "epoch": 7.14, "learning_rate": 1.5e-06, "loss": 0.5593, "step": 75 }, { "epoch": 9.52, "learning_rate": 2.0000000000000003e-06, "loss": 0.2685, "step": 100 }, { "epoch": 11.9, "learning_rate": 2.5e-06, "loss": 0.0974, "step": 125 }, { "epoch": 14.29, "learning_rate": 3e-06, "loss": 0.0248, "step": 150 }, { "epoch": 16.67, "learning_rate": 3.5e-06, "loss": 0.0113, "step": 175 }, { "epoch": 19.05, "learning_rate": 4.000000000000001e-06, "loss": 0.0078, "step": 200 }, { "epoch": 21.43, "learning_rate": 4.5e-06, "loss": 0.0083, "step": 225 }, { "epoch": 23.81, "learning_rate": 5e-06, "loss": 0.0134, "step": 250 }, { "epoch": 26.19, "learning_rate": 5.500000000000001e-06, "loss": 0.0086, "step": 275 }, { "epoch": 28.57, "learning_rate": 6e-06, "loss": 0.0074, "step": 300 }, { "epoch": 30.95, "learning_rate": 6.5000000000000004e-06, "loss": 0.011, "step": 325 }, { "epoch": 33.33, "learning_rate": 7e-06, "loss": 0.0114, "step": 350 }, { "epoch": 35.71, "learning_rate": 7.500000000000001e-06, "loss": 0.0152, "step": 375 }, { "epoch": 38.1, "learning_rate": 8.000000000000001e-06, "loss": 0.0081, "step": 400 }, { "epoch": 40.48, "learning_rate": 8.5e-06, "loss": 0.0025, "step": 425 }, { "epoch": 42.86, "learning_rate": 9e-06, "loss": 0.0065, "step": 450 }, { "epoch": 45.24, "learning_rate": 9.5e-06, "loss": 0.006, "step": 475 }, { "epoch": 47.62, "learning_rate": 1e-05, "loss": 0.0071, "step": 500 }, { "epoch": 50.0, "learning_rate": 9.944444444444445e-06, "loss": 0.0145, "step": 525 }, { "epoch": 52.38, "learning_rate": 9.88888888888889e-06, "loss": 0.0163, "step": 550 }, { "epoch": 54.76, "learning_rate": 9.833333333333333e-06, "loss": 0.0055, "step": 575 }, { "epoch": 57.14, "learning_rate": 9.777777777777779e-06, "loss": 0.0053, "step": 600 }, { "epoch": 59.52, "learning_rate": 9.722222222222223e-06, "loss": 0.0012, "step": 625 }, { "epoch": 61.9, "learning_rate": 9.666666666666667e-06, "loss": 0.0004, "step": 650 }, { "epoch": 64.29, "learning_rate": 9.611111111111112e-06, "loss": 0.0002, "step": 675 }, { "epoch": 66.67, "learning_rate": 9.555555555555556e-06, "loss": 0.0002, "step": 700 }, { "epoch": 69.05, "learning_rate": 9.5e-06, "loss": 0.0001, "step": 725 }, { "epoch": 71.43, "learning_rate": 9.444444444444445e-06, "loss": 0.0001, "step": 750 }, { "epoch": 73.81, "learning_rate": 9.38888888888889e-06, "loss": 0.0001, "step": 775 }, { "epoch": 76.19, "learning_rate": 9.333333333333334e-06, "loss": 0.0001, "step": 800 }, { "epoch": 78.57, "learning_rate": 9.277777777777778e-06, "loss": 0.0001, "step": 825 }, { "epoch": 80.95, "learning_rate": 9.222222222222224e-06, "loss": 0.0001, "step": 850 }, { "epoch": 83.33, "learning_rate": 9.166666666666666e-06, "loss": 0.0001, "step": 875 }, { "epoch": 85.71, "learning_rate": 9.111111111111112e-06, "loss": 0.0001, "step": 900 }, { "epoch": 88.1, "learning_rate": 9.055555555555556e-06, "loss": 0.0001, "step": 925 }, { "epoch": 90.48, "learning_rate": 9e-06, "loss": 0.0001, "step": 950 }, { "epoch": 92.86, "learning_rate": 8.944444444444446e-06, "loss": 0.0001, "step": 975 }, { "epoch": 95.24, "learning_rate": 8.888888888888888e-06, "loss": 0.0001, "step": 1000 }, { "epoch": 95.24, "eval_loss": 1.5191562175750732, "eval_runtime": 133.4977, "eval_samples_per_second": 1.236, "eval_steps_per_second": 0.157, "eval_wer": 0.4586097490625901, "step": 1000 }, { "epoch": 97.62, "learning_rate": 8.833333333333334e-06, "loss": 0.0001, "step": 1025 }, { "epoch": 100.0, "learning_rate": 8.777777777777778e-06, "loss": 0.0001, "step": 1050 }, { "epoch": 102.38, "learning_rate": 8.722222222222224e-06, "loss": 0.0001, "step": 1075 }, { "epoch": 104.76, "learning_rate": 8.666666666666668e-06, "loss": 0.0001, "step": 1100 }, { "epoch": 107.14, "learning_rate": 8.611111111111112e-06, "loss": 0.0001, "step": 1125 }, { "epoch": 109.52, "learning_rate": 8.555555555555556e-06, "loss": 0.0001, "step": 1150 }, { "epoch": 111.9, "learning_rate": 8.5e-06, "loss": 0.0001, "step": 1175 }, { "epoch": 114.29, "learning_rate": 8.444444444444446e-06, "loss": 0.0001, "step": 1200 }, { "epoch": 116.67, "learning_rate": 8.38888888888889e-06, "loss": 0.0, "step": 1225 }, { "epoch": 119.05, "learning_rate": 8.333333333333334e-06, "loss": 0.0, "step": 1250 }, { "epoch": 121.43, "learning_rate": 8.277777777777778e-06, "loss": 0.0, "step": 1275 }, { "epoch": 123.81, "learning_rate": 8.222222222222222e-06, "loss": 0.0, "step": 1300 }, { "epoch": 126.19, "learning_rate": 8.166666666666668e-06, "loss": 0.0, "step": 1325 }, { "epoch": 128.57, "learning_rate": 8.111111111111112e-06, "loss": 0.0, "step": 1350 }, { "epoch": 130.95, "learning_rate": 8.055555555555557e-06, "loss": 0.0, "step": 1375 }, { "epoch": 133.33, "learning_rate": 8.000000000000001e-06, "loss": 0.0, "step": 1400 }, { "epoch": 135.71, "learning_rate": 7.944444444444445e-06, "loss": 0.0, "step": 1425 }, { "epoch": 138.1, "learning_rate": 7.88888888888889e-06, "loss": 0.0, "step": 1450 }, { "epoch": 140.48, "learning_rate": 7.833333333333333e-06, "loss": 0.0, "step": 1475 }, { "epoch": 142.86, "learning_rate": 7.77777777777778e-06, "loss": 0.0, "step": 1500 }, { "epoch": 145.24, "learning_rate": 7.722222222222223e-06, "loss": 0.0, "step": 1525 }, { "epoch": 147.62, "learning_rate": 7.666666666666667e-06, "loss": 0.0, "step": 1550 }, { "epoch": 150.0, "learning_rate": 7.611111111111111e-06, "loss": 0.0, "step": 1575 }, { "epoch": 152.38, "learning_rate": 7.555555555555556e-06, "loss": 0.0, "step": 1600 }, { "epoch": 154.76, "learning_rate": 7.500000000000001e-06, "loss": 0.0, "step": 1625 }, { "epoch": 157.14, "learning_rate": 7.444444444444445e-06, "loss": 0.0, "step": 1650 }, { "epoch": 159.52, "learning_rate": 7.38888888888889e-06, "loss": 0.0, "step": 1675 }, { "epoch": 161.9, "learning_rate": 7.333333333333333e-06, "loss": 0.0, "step": 1700 }, { "epoch": 164.29, "learning_rate": 7.277777777777778e-06, "loss": 0.0, "step": 1725 }, { "epoch": 166.67, "learning_rate": 7.222222222222223e-06, "loss": 0.0, "step": 1750 }, { "epoch": 169.05, "learning_rate": 7.166666666666667e-06, "loss": 0.0, "step": 1775 }, { "epoch": 171.43, "learning_rate": 7.111111111111112e-06, "loss": 0.0, "step": 1800 }, { "epoch": 173.81, "learning_rate": 7.055555555555557e-06, "loss": 0.0, "step": 1825 }, { "epoch": 176.19, "learning_rate": 7e-06, "loss": 0.0, "step": 1850 }, { "epoch": 178.57, "learning_rate": 6.944444444444445e-06, "loss": 0.0, "step": 1875 }, { "epoch": 180.95, "learning_rate": 6.88888888888889e-06, "loss": 0.0, "step": 1900 }, { "epoch": 183.33, "learning_rate": 6.833333333333334e-06, "loss": 0.0, "step": 1925 }, { "epoch": 185.71, "learning_rate": 6.777777777777779e-06, "loss": 0.0, "step": 1950 }, { "epoch": 188.1, "learning_rate": 6.7222222222222235e-06, "loss": 0.0, "step": 1975 }, { "epoch": 190.48, "learning_rate": 6.666666666666667e-06, "loss": 0.0, "step": 2000 }, { "epoch": 190.48, "eval_loss": 1.6149208545684814, "eval_runtime": 133.8069, "eval_samples_per_second": 1.233, "eval_steps_per_second": 0.157, "eval_wer": 0.44793769829824054, "step": 2000 }, { "epoch": 192.86, "learning_rate": 6.6111111111111115e-06, "loss": 0.0, "step": 2025 }, { "epoch": 195.24, "learning_rate": 6.555555555555556e-06, "loss": 0.0, "step": 2050 }, { "epoch": 197.62, "learning_rate": 6.5000000000000004e-06, "loss": 0.0, "step": 2075 }, { "epoch": 200.0, "learning_rate": 6.444444444444445e-06, "loss": 0.0, "step": 2100 }, { "epoch": 202.38, "learning_rate": 6.3888888888888885e-06, "loss": 0.0, "step": 2125 }, { "epoch": 204.76, "learning_rate": 6.333333333333333e-06, "loss": 0.0, "step": 2150 }, { "epoch": 207.14, "learning_rate": 6.277777777777778e-06, "loss": 0.0, "step": 2175 }, { "epoch": 209.52, "learning_rate": 6.222222222222223e-06, "loss": 0.0, "step": 2200 }, { "epoch": 211.9, "learning_rate": 6.166666666666667e-06, "loss": 0.0, "step": 2225 }, { "epoch": 214.29, "learning_rate": 6.111111111111112e-06, "loss": 0.0, "step": 2250 }, { "epoch": 216.67, "learning_rate": 6.055555555555555e-06, "loss": 0.0, "step": 2275 }, { "epoch": 219.05, "learning_rate": 6e-06, "loss": 0.0, "step": 2300 }, { "epoch": 221.43, "learning_rate": 5.944444444444445e-06, "loss": 0.0, "step": 2325 }, { "epoch": 223.81, "learning_rate": 5.88888888888889e-06, "loss": 0.0, "step": 2350 }, { "epoch": 226.19, "learning_rate": 5.833333333333334e-06, "loss": 0.0, "step": 2375 }, { "epoch": 228.57, "learning_rate": 5.777777777777778e-06, "loss": 0.0, "step": 2400 }, { "epoch": 230.95, "learning_rate": 5.722222222222222e-06, "loss": 0.0, "step": 2425 }, { "epoch": 233.33, "learning_rate": 5.666666666666667e-06, "loss": 0.0, "step": 2450 }, { "epoch": 235.71, "learning_rate": 5.611111111111112e-06, "loss": 0.0, "step": 2475 }, { "epoch": 238.1, "learning_rate": 5.555555555555557e-06, "loss": 0.0, "step": 2500 }, { "epoch": 240.48, "learning_rate": 5.500000000000001e-06, "loss": 0.0, "step": 2525 }, { "epoch": 242.86, "learning_rate": 5.444444444444445e-06, "loss": 0.0, "step": 2550 }, { "epoch": 245.24, "learning_rate": 5.388888888888889e-06, "loss": 0.0, "step": 2575 }, { "epoch": 247.62, "learning_rate": 5.333333333333334e-06, "loss": 0.0, "step": 2600 }, { "epoch": 250.0, "learning_rate": 5.2777777777777785e-06, "loss": 0.0, "step": 2625 }, { "epoch": 252.38, "learning_rate": 5.2222222222222226e-06, "loss": 0.0, "step": 2650 }, { "epoch": 254.76, "learning_rate": 5.1666666666666675e-06, "loss": 0.0, "step": 2675 }, { "epoch": 257.14, "learning_rate": 5.1111111111111115e-06, "loss": 0.0, "step": 2700 }, { "epoch": 259.52, "learning_rate": 5.0555555555555555e-06, "loss": 0.0, "step": 2725 }, { "epoch": 261.9, "learning_rate": 5e-06, "loss": 0.0, "step": 2750 }, { "epoch": 264.29, "learning_rate": 4.944444444444445e-06, "loss": 0.0, "step": 2775 }, { "epoch": 266.67, "learning_rate": 4.888888888888889e-06, "loss": 0.0, "step": 2800 }, { "epoch": 269.05, "learning_rate": 4.833333333333333e-06, "loss": 0.0, "step": 2825 }, { "epoch": 271.43, "learning_rate": 4.777777777777778e-06, "loss": 0.0, "step": 2850 }, { "epoch": 273.81, "learning_rate": 4.722222222222222e-06, "loss": 0.0, "step": 2875 }, { "epoch": 276.19, "learning_rate": 4.666666666666667e-06, "loss": 0.0, "step": 2900 }, { "epoch": 278.57, "learning_rate": 4.611111111111112e-06, "loss": 0.0, "step": 2925 }, { "epoch": 280.95, "learning_rate": 4.555555555555556e-06, "loss": 0.0, "step": 2950 }, { "epoch": 283.33, "learning_rate": 4.5e-06, "loss": 0.0, "step": 2975 }, { "epoch": 285.71, "learning_rate": 4.444444444444444e-06, "loss": 0.0, "step": 3000 }, { "epoch": 285.71, "eval_loss": 1.663454532623291, "eval_runtime": 131.6907, "eval_samples_per_second": 1.253, "eval_steps_per_second": 0.159, "eval_wer": 0.4505336025382175, "step": 3000 }, { "epoch": 288.1, "learning_rate": 4.388888888888889e-06, "loss": 0.0, "step": 3025 }, { "epoch": 290.48, "learning_rate": 4.333333333333334e-06, "loss": 0.0, "step": 3050 }, { "epoch": 292.86, "learning_rate": 4.277777777777778e-06, "loss": 0.0, "step": 3075 }, { "epoch": 295.24, "learning_rate": 4.222222222222223e-06, "loss": 0.0, "step": 3100 }, { "epoch": 297.62, "learning_rate": 4.166666666666667e-06, "loss": 0.0, "step": 3125 }, { "epoch": 300.0, "learning_rate": 4.111111111111111e-06, "loss": 0.0, "step": 3150 }, { "epoch": 302.38, "learning_rate": 4.055555555555556e-06, "loss": 0.0, "step": 3175 }, { "epoch": 304.76, "learning_rate": 4.000000000000001e-06, "loss": 0.0, "step": 3200 }, { "epoch": 307.14, "learning_rate": 3.944444444444445e-06, "loss": 0.0, "step": 3225 }, { "epoch": 309.52, "learning_rate": 3.88888888888889e-06, "loss": 0.0, "step": 3250 }, { "epoch": 311.9, "learning_rate": 3.833333333333334e-06, "loss": 0.0, "step": 3275 }, { "epoch": 314.29, "learning_rate": 3.777777777777778e-06, "loss": 0.0, "step": 3300 }, { "epoch": 316.67, "learning_rate": 3.7222222222222225e-06, "loss": 0.0, "step": 3325 }, { "epoch": 319.05, "learning_rate": 3.6666666666666666e-06, "loss": 0.0, "step": 3350 }, { "epoch": 321.43, "learning_rate": 3.6111111111111115e-06, "loss": 0.0, "step": 3375 }, { "epoch": 323.81, "learning_rate": 3.555555555555556e-06, "loss": 0.0, "step": 3400 }, { "epoch": 326.19, "learning_rate": 3.5e-06, "loss": 0.0, "step": 3425 }, { "epoch": 328.57, "learning_rate": 3.444444444444445e-06, "loss": 0.0, "step": 3450 }, { "epoch": 330.95, "learning_rate": 3.3888888888888893e-06, "loss": 0.0, "step": 3475 }, { "epoch": 333.33, "learning_rate": 3.3333333333333333e-06, "loss": 0.0, "step": 3500 }, { "epoch": 335.71, "learning_rate": 3.277777777777778e-06, "loss": 0.0, "step": 3525 }, { "epoch": 338.1, "learning_rate": 3.2222222222222227e-06, "loss": 0.0, "step": 3550 }, { "epoch": 340.48, "learning_rate": 3.1666666666666667e-06, "loss": 0.0, "step": 3575 }, { "epoch": 342.86, "learning_rate": 3.1111111111111116e-06, "loss": 0.0, "step": 3600 }, { "epoch": 345.24, "learning_rate": 3.055555555555556e-06, "loss": 0.0, "step": 3625 }, { "epoch": 347.62, "learning_rate": 3e-06, "loss": 0.0, "step": 3650 }, { "epoch": 350.0, "learning_rate": 2.944444444444445e-06, "loss": 0.0, "step": 3675 }, { "epoch": 352.38, "learning_rate": 2.888888888888889e-06, "loss": 0.0, "step": 3700 }, { "epoch": 354.76, "learning_rate": 2.8333333333333335e-06, "loss": 0.0, "step": 3725 }, { "epoch": 357.14, "learning_rate": 2.7777777777777783e-06, "loss": 0.0, "step": 3750 }, { "epoch": 359.52, "learning_rate": 2.7222222222222224e-06, "loss": 0.0, "step": 3775 }, { "epoch": 361.9, "learning_rate": 2.666666666666667e-06, "loss": 0.0, "step": 3800 }, { "epoch": 364.29, "learning_rate": 2.6111111111111113e-06, "loss": 0.0, "step": 3825 }, { "epoch": 366.67, "learning_rate": 2.5555555555555557e-06, "loss": 0.0, "step": 3850 }, { "epoch": 369.05, "learning_rate": 2.5e-06, "loss": 0.0, "step": 3875 }, { "epoch": 371.43, "learning_rate": 2.4444444444444447e-06, "loss": 0.0, "step": 3900 }, { "epoch": 373.81, "learning_rate": 2.388888888888889e-06, "loss": 0.0, "step": 3925 }, { "epoch": 376.19, "learning_rate": 2.3333333333333336e-06, "loss": 0.0, "step": 3950 }, { "epoch": 378.57, "learning_rate": 2.277777777777778e-06, "loss": 0.0, "step": 3975 }, { "epoch": 380.95, "learning_rate": 2.222222222222222e-06, "loss": 0.0, "step": 4000 }, { "epoch": 380.95, "eval_loss": 1.6946672201156616, "eval_runtime": 129.5666, "eval_samples_per_second": 1.273, "eval_steps_per_second": 0.162, "eval_wer": 0.4485145659071243, "step": 4000 }, { "epoch": 383.33, "learning_rate": 2.166666666666667e-06, "loss": 0.0, "step": 4025 }, { "epoch": 385.71, "learning_rate": 2.1111111111111114e-06, "loss": 0.0, "step": 4050 }, { "epoch": 388.1, "learning_rate": 2.0555555555555555e-06, "loss": 0.0, "step": 4075 }, { "epoch": 390.48, "learning_rate": 2.0000000000000003e-06, "loss": 0.0, "step": 4100 }, { "epoch": 392.86, "learning_rate": 1.944444444444445e-06, "loss": 0.0, "step": 4125 }, { "epoch": 395.24, "learning_rate": 1.888888888888889e-06, "loss": 0.0, "step": 4150 }, { "epoch": 397.62, "learning_rate": 1.8333333333333333e-06, "loss": 0.0, "step": 4175 }, { "epoch": 400.0, "learning_rate": 1.777777777777778e-06, "loss": 0.0, "step": 4200 }, { "epoch": 402.38, "learning_rate": 1.7222222222222224e-06, "loss": 0.0, "step": 4225 }, { "epoch": 404.76, "learning_rate": 1.6666666666666667e-06, "loss": 0.0, "step": 4250 }, { "epoch": 407.14, "learning_rate": 1.6111111111111113e-06, "loss": 0.0, "step": 4275 }, { "epoch": 409.52, "learning_rate": 1.5555555555555558e-06, "loss": 0.0, "step": 4300 }, { "epoch": 411.9, "learning_rate": 1.5e-06, "loss": 0.0, "step": 4325 }, { "epoch": 414.29, "learning_rate": 1.4444444444444445e-06, "loss": 0.0, "step": 4350 }, { "epoch": 416.67, "learning_rate": 1.3888888888888892e-06, "loss": 0.0, "step": 4375 }, { "epoch": 419.05, "learning_rate": 1.3333333333333334e-06, "loss": 0.0, "step": 4400 }, { "epoch": 421.43, "learning_rate": 1.2777777777777779e-06, "loss": 0.0, "step": 4425 }, { "epoch": 423.81, "learning_rate": 1.2222222222222223e-06, "loss": 0.0, "step": 4450 }, { "epoch": 426.19, "learning_rate": 1.1666666666666668e-06, "loss": 0.0, "step": 4475 }, { "epoch": 428.57, "learning_rate": 1.111111111111111e-06, "loss": 0.0, "step": 4500 }, { "epoch": 430.95, "learning_rate": 1.0555555555555557e-06, "loss": 0.0, "step": 4525 }, { "epoch": 433.33, "learning_rate": 1.0000000000000002e-06, "loss": 0.0, "step": 4550 }, { "epoch": 435.71, "learning_rate": 9.444444444444445e-07, "loss": 0.0, "step": 4575 }, { "epoch": 438.1, "learning_rate": 8.88888888888889e-07, "loss": 0.0, "step": 4600 }, { "epoch": 440.48, "learning_rate": 8.333333333333333e-07, "loss": 0.0, "step": 4625 }, { "epoch": 442.86, "learning_rate": 7.777777777777779e-07, "loss": 0.0, "step": 4650 }, { "epoch": 445.24, "learning_rate": 7.222222222222222e-07, "loss": 0.0, "step": 4675 }, { "epoch": 447.62, "learning_rate": 6.666666666666667e-07, "loss": 0.0, "step": 4700 }, { "epoch": 450.0, "learning_rate": 6.111111111111112e-07, "loss": 0.0, "step": 4725 }, { "epoch": 452.38, "learning_rate": 5.555555555555555e-07, "loss": 0.0, "step": 4750 }, { "epoch": 454.76, "learning_rate": 5.000000000000001e-07, "loss": 0.0, "step": 4775 }, { "epoch": 457.14, "learning_rate": 4.444444444444445e-07, "loss": 0.0, "step": 4800 }, { "epoch": 459.52, "learning_rate": 3.8888888888888895e-07, "loss": 0.0, "step": 4825 }, { "epoch": 461.9, "learning_rate": 3.3333333333333335e-07, "loss": 0.0, "step": 4850 }, { "epoch": 464.29, "learning_rate": 2.7777777777777776e-07, "loss": 0.0, "step": 4875 }, { "epoch": 466.67, "learning_rate": 2.2222222222222224e-07, "loss": 0.0, "step": 4900 }, { "epoch": 469.05, "learning_rate": 1.6666666666666668e-07, "loss": 0.0, "step": 4925 }, { "epoch": 471.43, "learning_rate": 1.1111111111111112e-07, "loss": 0.0, "step": 4950 }, { "epoch": 473.81, "learning_rate": 5.555555555555556e-08, "loss": 0.0, "step": 4975 }, { "epoch": 476.19, "learning_rate": 0.0, "loss": 0.0, "step": 5000 }, { "epoch": 476.19, "eval_loss": 1.705546259880066, "eval_runtime": 129.4505, "eval_samples_per_second": 1.275, "eval_steps_per_second": 0.162, "eval_wer": 0.4488029997115662, "step": 5000 }, { "epoch": 476.19, "step": 5000, "total_flos": 2.6209660936126464e+20, "train_loss": 0.01602925512266811, "train_runtime": 26933.9586, "train_samples_per_second": 2.97, "train_steps_per_second": 0.186 } ], "logging_steps": 25, "max_steps": 5000, "num_input_tokens_seen": 0, "num_train_epochs": 500, "save_steps": 1000, "total_flos": 2.6209660936126464e+20, "train_batch_size": 8, "trial_name": null, "trial_params": null }