|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 0.9988554780980127, |
|
"eval_steps": 150, |
|
"global_step": 600, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 5e-06, |
|
"loss": 8.3066, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 1e-05, |
|
"loss": 7.9763, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 1.5e-05, |
|
"loss": 7.2999, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 2e-05, |
|
"loss": 6.3145, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"learning_rate": 2.5e-05, |
|
"loss": 4.8549, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 3e-05, |
|
"loss": 3.7485, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"learning_rate": 3.5e-05, |
|
"loss": 3.0173, |
|
"step": 35 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 4e-05, |
|
"loss": 2.6364, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 4.5e-05, |
|
"loss": 2.4726, |
|
"step": 45 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"learning_rate": 5e-05, |
|
"loss": 2.3895, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"learning_rate": 5.500000000000001e-05, |
|
"loss": 2.2992, |
|
"step": 55 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 6e-05, |
|
"loss": 2.291, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 6.500000000000001e-05, |
|
"loss": 2.2597, |
|
"step": 65 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 7e-05, |
|
"loss": 2.2245, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 7.500000000000001e-05, |
|
"loss": 2.2076, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 0.13, |
|
"learning_rate": 8e-05, |
|
"loss": 2.2132, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 8.5e-05, |
|
"loss": 2.1925, |
|
"step": 85 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 9e-05, |
|
"loss": 2.1868, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"learning_rate": 9.5e-05, |
|
"loss": 2.1916, |
|
"step": 95 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"learning_rate": 0.0001, |
|
"loss": 2.1774, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"learning_rate": 9.900000000000001e-05, |
|
"loss": 2.183, |
|
"step": 105 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"learning_rate": 9.8e-05, |
|
"loss": 2.1712, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"learning_rate": 9.7e-05, |
|
"loss": 2.1713, |
|
"step": 115 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"learning_rate": 9.6e-05, |
|
"loss": 2.1731, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 9.5e-05, |
|
"loss": 2.1565, |
|
"step": 125 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"learning_rate": 9.4e-05, |
|
"loss": 2.1575, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"learning_rate": 9.300000000000001e-05, |
|
"loss": 2.1414, |
|
"step": 135 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"learning_rate": 9.200000000000001e-05, |
|
"loss": 2.1384, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.24, |
|
"learning_rate": 9.1e-05, |
|
"loss": 2.1487, |
|
"step": 145 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"learning_rate": 9e-05, |
|
"loss": 2.1283, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"eval_accuracy": 0.6017748783939872, |
|
"eval_loss": 2.089164972305298, |
|
"eval_runtime": 58.1804, |
|
"eval_samples_per_second": 5.156, |
|
"eval_steps_per_second": 1.289, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"learning_rate": 8.900000000000001e-05, |
|
"loss": 2.1224, |
|
"step": 155 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 8.800000000000001e-05, |
|
"loss": 2.144, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 8.7e-05, |
|
"loss": 2.1293, |
|
"step": 165 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"learning_rate": 8.6e-05, |
|
"loss": 2.1417, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"learning_rate": 8.5e-05, |
|
"loss": 2.1282, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"learning_rate": 8.4e-05, |
|
"loss": 2.1008, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"learning_rate": 8.3e-05, |
|
"loss": 2.1218, |
|
"step": 185 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 8.2e-05, |
|
"loss": 2.1292, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 8.1e-05, |
|
"loss": 2.114, |
|
"step": 195 |
|
}, |
|
{ |
|
"epoch": 0.33, |
|
"learning_rate": 8e-05, |
|
"loss": 2.1135, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"learning_rate": 7.900000000000001e-05, |
|
"loss": 2.118, |
|
"step": 205 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"learning_rate": 7.800000000000001e-05, |
|
"loss": 2.1231, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 0.36, |
|
"learning_rate": 7.7e-05, |
|
"loss": 2.0705, |
|
"step": 215 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"learning_rate": 7.6e-05, |
|
"loss": 2.11, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"learning_rate": 7.500000000000001e-05, |
|
"loss": 2.0866, |
|
"step": 225 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"learning_rate": 7.4e-05, |
|
"loss": 2.1115, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 0.39, |
|
"learning_rate": 7.3e-05, |
|
"loss": 2.1069, |
|
"step": 235 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"learning_rate": 7.2e-05, |
|
"loss": 2.1083, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 0.41, |
|
"learning_rate": 7.1e-05, |
|
"loss": 2.1014, |
|
"step": 245 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"learning_rate": 7e-05, |
|
"loss": 2.1029, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"learning_rate": 6.9e-05, |
|
"loss": 2.0846, |
|
"step": 255 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"learning_rate": 6.800000000000001e-05, |
|
"loss": 2.1059, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.44, |
|
"learning_rate": 6.7e-05, |
|
"loss": 2.0875, |
|
"step": 265 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"learning_rate": 6.6e-05, |
|
"loss": 2.0973, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"learning_rate": 6.500000000000001e-05, |
|
"loss": 2.0932, |
|
"step": 275 |
|
}, |
|
{ |
|
"epoch": 0.47, |
|
"learning_rate": 6.400000000000001e-05, |
|
"loss": 2.0721, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 0.47, |
|
"learning_rate": 6.3e-05, |
|
"loss": 2.0892, |
|
"step": 285 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"learning_rate": 6.2e-05, |
|
"loss": 2.1002, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"learning_rate": 6.1e-05, |
|
"loss": 2.0804, |
|
"step": 295 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"learning_rate": 6e-05, |
|
"loss": 2.0999, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"eval_accuracy": 0.6084190234842455, |
|
"eval_loss": 2.038738965988159, |
|
"eval_runtime": 68.8578, |
|
"eval_samples_per_second": 4.357, |
|
"eval_steps_per_second": 1.089, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"learning_rate": 5.9e-05, |
|
"loss": 2.0905, |
|
"step": 305 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"learning_rate": 5.8e-05, |
|
"loss": 2.046, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"learning_rate": 5.6999999999999996e-05, |
|
"loss": 2.0794, |
|
"step": 315 |
|
}, |
|
{ |
|
"epoch": 0.53, |
|
"learning_rate": 5.6000000000000006e-05, |
|
"loss": 2.0714, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"learning_rate": 5.500000000000001e-05, |
|
"loss": 2.0711, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 0.55, |
|
"learning_rate": 5.4000000000000005e-05, |
|
"loss": 2.0525, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 0.56, |
|
"learning_rate": 5.300000000000001e-05, |
|
"loss": 2.0689, |
|
"step": 335 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"learning_rate": 5.2000000000000004e-05, |
|
"loss": 2.0936, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"learning_rate": 5.1000000000000006e-05, |
|
"loss": 2.062, |
|
"step": 345 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"learning_rate": 5e-05, |
|
"loss": 2.0621, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 0.59, |
|
"learning_rate": 4.9e-05, |
|
"loss": 2.0662, |
|
"step": 355 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"learning_rate": 4.8e-05, |
|
"loss": 2.0779, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 0.61, |
|
"learning_rate": 4.7e-05, |
|
"loss": 2.0773, |
|
"step": 365 |
|
}, |
|
{ |
|
"epoch": 0.62, |
|
"learning_rate": 4.600000000000001e-05, |
|
"loss": 2.0407, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 0.62, |
|
"learning_rate": 4.5e-05, |
|
"loss": 2.0603, |
|
"step": 375 |
|
}, |
|
{ |
|
"epoch": 0.63, |
|
"learning_rate": 4.4000000000000006e-05, |
|
"loss": 2.0556, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"learning_rate": 4.3e-05, |
|
"loss": 2.0742, |
|
"step": 385 |
|
}, |
|
{ |
|
"epoch": 0.65, |
|
"learning_rate": 4.2e-05, |
|
"loss": 2.0444, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 0.66, |
|
"learning_rate": 4.1e-05, |
|
"loss": 2.0409, |
|
"step": 395 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"learning_rate": 4e-05, |
|
"loss": 2.0539, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"learning_rate": 3.9000000000000006e-05, |
|
"loss": 2.0696, |
|
"step": 405 |
|
}, |
|
{ |
|
"epoch": 0.68, |
|
"learning_rate": 3.8e-05, |
|
"loss": 2.0614, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 0.69, |
|
"learning_rate": 3.7e-05, |
|
"loss": 2.0604, |
|
"step": 415 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"learning_rate": 3.6e-05, |
|
"loss": 2.0608, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 0.71, |
|
"learning_rate": 3.5e-05, |
|
"loss": 2.0412, |
|
"step": 425 |
|
}, |
|
{ |
|
"epoch": 0.72, |
|
"learning_rate": 3.4000000000000007e-05, |
|
"loss": 2.0558, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 0.72, |
|
"learning_rate": 3.3e-05, |
|
"loss": 2.0338, |
|
"step": 435 |
|
}, |
|
{ |
|
"epoch": 0.73, |
|
"learning_rate": 3.2000000000000005e-05, |
|
"loss": 2.0322, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 0.74, |
|
"learning_rate": 3.1e-05, |
|
"loss": 2.0511, |
|
"step": 445 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"learning_rate": 3e-05, |
|
"loss": 2.0595, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"eval_accuracy": 0.6142522292024847, |
|
"eval_loss": 1.9970886707305908, |
|
"eval_runtime": 61.0863, |
|
"eval_samples_per_second": 4.911, |
|
"eval_steps_per_second": 1.228, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"learning_rate": 2.9e-05, |
|
"loss": 2.032, |
|
"step": 455 |
|
}, |
|
{ |
|
"epoch": 0.77, |
|
"learning_rate": 2.8000000000000003e-05, |
|
"loss": 2.0421, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 0.77, |
|
"learning_rate": 2.7000000000000002e-05, |
|
"loss": 2.0255, |
|
"step": 465 |
|
}, |
|
{ |
|
"epoch": 0.78, |
|
"learning_rate": 2.6000000000000002e-05, |
|
"loss": 2.0259, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 0.79, |
|
"learning_rate": 2.5e-05, |
|
"loss": 2.0361, |
|
"step": 475 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"learning_rate": 2.4e-05, |
|
"loss": 2.0584, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 0.81, |
|
"learning_rate": 2.3000000000000003e-05, |
|
"loss": 2.0589, |
|
"step": 485 |
|
}, |
|
{ |
|
"epoch": 0.82, |
|
"learning_rate": 2.2000000000000003e-05, |
|
"loss": 2.0555, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 0.82, |
|
"learning_rate": 2.1e-05, |
|
"loss": 2.0413, |
|
"step": 495 |
|
}, |
|
{ |
|
"epoch": 0.83, |
|
"learning_rate": 2e-05, |
|
"loss": 2.0097, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.84, |
|
"learning_rate": 1.9e-05, |
|
"loss": 2.0466, |
|
"step": 505 |
|
}, |
|
{ |
|
"epoch": 0.85, |
|
"learning_rate": 1.8e-05, |
|
"loss": 2.0572, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"learning_rate": 1.7000000000000003e-05, |
|
"loss": 2.0376, |
|
"step": 515 |
|
}, |
|
{ |
|
"epoch": 0.87, |
|
"learning_rate": 1.6000000000000003e-05, |
|
"loss": 2.0202, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 0.87, |
|
"learning_rate": 1.5e-05, |
|
"loss": 2.0643, |
|
"step": 525 |
|
}, |
|
{ |
|
"epoch": 0.88, |
|
"learning_rate": 1.4000000000000001e-05, |
|
"loss": 2.0201, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"learning_rate": 1.3000000000000001e-05, |
|
"loss": 2.0356, |
|
"step": 535 |
|
}, |
|
{ |
|
"epoch": 0.9, |
|
"learning_rate": 1.2e-05, |
|
"loss": 2.0255, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"learning_rate": 1.1000000000000001e-05, |
|
"loss": 2.0374, |
|
"step": 545 |
|
}, |
|
{ |
|
"epoch": 0.92, |
|
"learning_rate": 1e-05, |
|
"loss": 2.0385, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 0.92, |
|
"learning_rate": 9e-06, |
|
"loss": 2.024, |
|
"step": 555 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"learning_rate": 8.000000000000001e-06, |
|
"loss": 2.0144, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 0.94, |
|
"learning_rate": 7.000000000000001e-06, |
|
"loss": 2.04, |
|
"step": 565 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"learning_rate": 6e-06, |
|
"loss": 2.0372, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"learning_rate": 5e-06, |
|
"loss": 2.0122, |
|
"step": 575 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"learning_rate": 4.000000000000001e-06, |
|
"loss": 2.0182, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"learning_rate": 3e-06, |
|
"loss": 2.0323, |
|
"step": 585 |
|
}, |
|
{ |
|
"epoch": 0.98, |
|
"learning_rate": 2.0000000000000003e-06, |
|
"loss": 2.0298, |
|
"step": 590 |
|
}, |
|
{ |
|
"epoch": 0.99, |
|
"learning_rate": 1.0000000000000002e-06, |
|
"loss": 2.0212, |
|
"step": 595 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 0.0, |
|
"loss": 2.0481, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.6152243190218628, |
|
"eval_loss": 1.9892692565917969, |
|
"eval_runtime": 64.3793, |
|
"eval_samples_per_second": 4.66, |
|
"eval_steps_per_second": 1.165, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"step": 600, |
|
"total_flos": 5.41006975991808e+16, |
|
"train_loss": 2.32403786500295, |
|
"train_runtime": 14945.8713, |
|
"train_samples_per_second": 2.572, |
|
"train_steps_per_second": 0.04 |
|
} |
|
], |
|
"logging_steps": 5, |
|
"max_steps": 600, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 1, |
|
"save_steps": 100, |
|
"total_flos": 5.41006975991808e+16, |
|
"train_batch_size": 4, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|