| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 49.997014925373136, | |
| "global_step": 8350, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.0009880239520958084, | |
| "loss": 5.0465, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0009760479041916168, | |
| "loss": 3.0126, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.0009640718562874252, | |
| "loss": 2.6782, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 0.0009520958083832335, | |
| "loss": 2.4056, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 0.0009401197604790419, | |
| "loss": 2.2999, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 0.0009281437125748503, | |
| "loss": 1.9862, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 0.0009161676646706587, | |
| "loss": 1.8922, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 0.0009041916167664672, | |
| "loss": 1.699, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.39, | |
| "learning_rate": 0.0008922155688622756, | |
| "loss": 1.5454, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 0.0008802395209580839, | |
| "loss": 1.4636, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.59, | |
| "learning_rate": 0.0008682634730538922, | |
| "loss": 1.2505, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.19, | |
| "learning_rate": 0.0008562874251497006, | |
| "loss": 1.1771, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 0.000844311377245509, | |
| "loss": 1.0694, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "learning_rate": 0.0008323353293413174, | |
| "loss": 0.9962, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "learning_rate": 0.0008203592814371258, | |
| "loss": 0.9573, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 9.58, | |
| "learning_rate": 0.0008083832335329342, | |
| "loss": 0.8383, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 10.18, | |
| "learning_rate": 0.0007964071856287425, | |
| "loss": 0.813, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 10.78, | |
| "learning_rate": 0.0007844311377245509, | |
| "loss": 0.7532, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 11.38, | |
| "learning_rate": 0.0007724550898203593, | |
| "loss": 0.6846, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 11.97, | |
| "learning_rate": 0.0007604790419161677, | |
| "loss": 0.6736, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 11.97, | |
| "eval_bleu": 2.5636, | |
| "eval_gen_len": 15.2363, | |
| "eval_loss": 0.29240211844444275, | |
| "eval_runtime": 438.109, | |
| "eval_samples_per_second": 12.228, | |
| "eval_steps_per_second": 3.059, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 12.57, | |
| "learning_rate": 0.0007485029940119761, | |
| "loss": 0.5751, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 13.17, | |
| "learning_rate": 0.0007365269461077845, | |
| "loss": 0.5731, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 13.77, | |
| "learning_rate": 0.0007245508982035929, | |
| "loss": 0.527, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 14.37, | |
| "learning_rate": 0.0007125748502994012, | |
| "loss": 0.5077, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "learning_rate": 0.0007005988023952096, | |
| "loss": 0.4783, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 15.57, | |
| "learning_rate": 0.000688622754491018, | |
| "loss": 0.4261, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 16.17, | |
| "learning_rate": 0.0006766467065868264, | |
| "loss": 0.4283, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 16.76, | |
| "learning_rate": 0.0006646706586826347, | |
| "loss": 0.3955, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 17.36, | |
| "learning_rate": 0.0006526946107784431, | |
| "loss": 0.3785, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 17.96, | |
| "learning_rate": 0.0006407185628742515, | |
| "loss": 0.3739, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 18.56, | |
| "learning_rate": 0.0006287425149700598, | |
| "loss": 0.3335, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 19.16, | |
| "learning_rate": 0.0006167664670658682, | |
| "loss": 0.3438, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 19.76, | |
| "learning_rate": 0.0006047904191616766, | |
| "loss": 0.3164, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 20.36, | |
| "learning_rate": 0.000592814371257485, | |
| "loss": 0.3068, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 20.96, | |
| "learning_rate": 0.0005808383233532934, | |
| "loss": 0.3188, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 21.56, | |
| "learning_rate": 0.0005688622754491018, | |
| "loss": 0.288, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 22.16, | |
| "learning_rate": 0.0005568862275449101, | |
| "loss": 0.2916, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 22.75, | |
| "learning_rate": 0.0005449101796407185, | |
| "loss": 0.2731, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 23.35, | |
| "learning_rate": 0.000532934131736527, | |
| "loss": 0.2655, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 23.95, | |
| "learning_rate": 0.0005209580838323354, | |
| "loss": 0.2621, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 23.95, | |
| "eval_bleu": 3.3704, | |
| "eval_gen_len": 16.0834, | |
| "eval_loss": 0.1354922354221344, | |
| "eval_runtime": 461.5346, | |
| "eval_samples_per_second": 11.607, | |
| "eval_steps_per_second": 2.903, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 24.55, | |
| "learning_rate": 0.0005089820359281438, | |
| "loss": 0.2493, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 25.15, | |
| "learning_rate": 0.0004970059880239521, | |
| "loss": 0.2502, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 25.75, | |
| "learning_rate": 0.00048502994011976046, | |
| "loss": 0.237, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 26.35, | |
| "learning_rate": 0.00047305389221556887, | |
| "loss": 0.2322, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 26.94, | |
| "learning_rate": 0.00046107784431137727, | |
| "loss": 0.2356, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 27.54, | |
| "learning_rate": 0.0004491017964071856, | |
| "loss": 0.2194, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 28.14, | |
| "learning_rate": 0.00043712574850299403, | |
| "loss": 0.2181, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 28.74, | |
| "learning_rate": 0.00042514970059880243, | |
| "loss": 0.2113, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 29.34, | |
| "learning_rate": 0.0004131736526946108, | |
| "loss": 0.2087, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 29.94, | |
| "learning_rate": 0.0004011976047904192, | |
| "loss": 0.214, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 30.54, | |
| "learning_rate": 0.0003892215568862276, | |
| "loss": 0.2028, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 31.14, | |
| "learning_rate": 0.0003772455089820359, | |
| "loss": 0.2036, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 31.73, | |
| "learning_rate": 0.0003652694610778443, | |
| "loss": 0.1953, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 32.33, | |
| "learning_rate": 0.0003532934131736527, | |
| "loss": 0.1926, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 32.93, | |
| "learning_rate": 0.0003413173652694611, | |
| "loss": 0.1974, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 33.53, | |
| "learning_rate": 0.00032934131736526946, | |
| "loss": 0.1866, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 34.13, | |
| "learning_rate": 0.00031736526946107786, | |
| "loss": 0.1839, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 34.73, | |
| "learning_rate": 0.00030538922155688627, | |
| "loss": 0.1816, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 35.33, | |
| "learning_rate": 0.0002934131736526946, | |
| "loss": 0.1837, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 35.93, | |
| "learning_rate": 0.000281437125748503, | |
| "loss": 0.1775, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 35.93, | |
| "eval_bleu": 3.6961, | |
| "eval_gen_len": 16.5854, | |
| "eval_loss": 0.12107560783624649, | |
| "eval_runtime": 477.8329, | |
| "eval_samples_per_second": 11.211, | |
| "eval_steps_per_second": 2.804, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 36.53, | |
| "learning_rate": 0.0002694610778443114, | |
| "loss": 0.1742, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 37.13, | |
| "learning_rate": 0.0002574850299401197, | |
| "loss": 0.176, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 37.72, | |
| "learning_rate": 0.00024550898203592813, | |
| "loss": 0.1704, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 38.32, | |
| "learning_rate": 0.00023353293413173654, | |
| "loss": 0.1734, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 38.92, | |
| "learning_rate": 0.0002215568862275449, | |
| "loss": 0.1662, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 39.52, | |
| "learning_rate": 0.0002095808383233533, | |
| "loss": 0.1653, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 40.12, | |
| "learning_rate": 0.0001976047904191617, | |
| "loss": 0.1684, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 40.72, | |
| "learning_rate": 0.00018562874251497005, | |
| "loss": 0.1584, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 41.32, | |
| "learning_rate": 0.00017365269461077845, | |
| "loss": 0.1622, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 41.91, | |
| "learning_rate": 0.00016167664670658683, | |
| "loss": 0.1625, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 42.51, | |
| "learning_rate": 0.0001497005988023952, | |
| "loss": 0.1558, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 43.11, | |
| "learning_rate": 0.00013772455089820359, | |
| "loss": 0.157, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 43.71, | |
| "learning_rate": 0.00012574850299401196, | |
| "loss": 0.1568, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 44.31, | |
| "learning_rate": 0.00011377245508982036, | |
| "loss": 0.1537, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 44.91, | |
| "learning_rate": 0.00010179640718562875, | |
| "loss": 0.1521, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 45.51, | |
| "learning_rate": 8.982035928143712e-05, | |
| "loss": 0.1508, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 46.11, | |
| "learning_rate": 7.784431137724552e-05, | |
| "loss": 0.149, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 46.7, | |
| "learning_rate": 6.58682634730539e-05, | |
| "loss": 0.1462, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 47.3, | |
| "learning_rate": 5.389221556886228e-05, | |
| "loss": 0.1484, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 47.9, | |
| "learning_rate": 4.191616766467066e-05, | |
| "loss": 0.1478, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 47.9, | |
| "eval_bleu": 3.681, | |
| "eval_gen_len": 16.4697, | |
| "eval_loss": 0.11556760966777802, | |
| "eval_runtime": 415.7833, | |
| "eval_samples_per_second": 12.884, | |
| "eval_steps_per_second": 3.223, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 48.5, | |
| "learning_rate": 2.994011976047904e-05, | |
| "loss": 0.1449, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 49.1, | |
| "learning_rate": 1.7964071856287426e-05, | |
| "loss": 0.1448, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 49.7, | |
| "learning_rate": 5.9880239520958085e-06, | |
| "loss": 0.1425, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 8350, | |
| "total_flos": 6.461676347272704e+16, | |
| "train_loss": 0.584253704162415, | |
| "train_runtime": 11332.4023, | |
| "train_samples_per_second": 23.636, | |
| "train_steps_per_second": 0.737 | |
| } | |
| ], | |
| "max_steps": 8350, | |
| "num_train_epochs": 50, | |
| "total_flos": 6.461676347272704e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |