| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 32.25806451612903, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.797335147857666, | |
| "eval_rouge1": 0.1301, | |
| "eval_rouge2": 0.0352, | |
| "eval_rougeL": 0.1074, | |
| "eval_rougeLsum": 0.1075, | |
| "eval_runtime": 18.6868, | |
| "eval_samples_per_second": 13.271, | |
| "eval_steps_per_second": 0.856, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.568485975265503, | |
| "eval_rouge1": 0.1455, | |
| "eval_rouge2": 0.051, | |
| "eval_rougeL": 0.1189, | |
| "eval_rougeLsum": 0.1187, | |
| "eval_runtime": 16.801, | |
| "eval_samples_per_second": 14.761, | |
| "eval_steps_per_second": 0.952, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.475400924682617, | |
| "eval_rouge1": 0.1674, | |
| "eval_rouge2": 0.0692, | |
| "eval_rougeL": 0.1397, | |
| "eval_rougeLsum": 0.1397, | |
| "eval_runtime": 17.3933, | |
| "eval_samples_per_second": 14.258, | |
| "eval_steps_per_second": 0.92, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.419504404067993, | |
| "eval_rouge1": 0.1901, | |
| "eval_rouge2": 0.0867, | |
| "eval_rougeL": 0.1586, | |
| "eval_rougeLsum": 0.1587, | |
| "eval_runtime": 16.7871, | |
| "eval_samples_per_second": 14.773, | |
| "eval_steps_per_second": 0.953, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.3755078315734863, | |
| "eval_rouge1": 0.1933, | |
| "eval_rouge2": 0.0907, | |
| "eval_rougeL": 0.1617, | |
| "eval_rougeLsum": 0.1619, | |
| "eval_runtime": 16.6712, | |
| "eval_samples_per_second": 14.876, | |
| "eval_steps_per_second": 0.96, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.3425652980804443, | |
| "eval_rouge1": 0.1946, | |
| "eval_rouge2": 0.0916, | |
| "eval_rougeL": 0.1634, | |
| "eval_rougeLsum": 0.1636, | |
| "eval_runtime": 17.1282, | |
| "eval_samples_per_second": 14.479, | |
| "eval_steps_per_second": 0.934, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.3197405338287354, | |
| "eval_rouge1": 0.1964, | |
| "eval_rouge2": 0.0929, | |
| "eval_rougeL": 0.1646, | |
| "eval_rougeLsum": 0.1648, | |
| "eval_runtime": 16.7039, | |
| "eval_samples_per_second": 14.847, | |
| "eval_steps_per_second": 0.958, | |
| "step": 434 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.2987782955169678, | |
| "eval_rouge1": 0.1968, | |
| "eval_rouge2": 0.0933, | |
| "eval_rougeL": 0.165, | |
| "eval_rougeLsum": 0.1653, | |
| "eval_runtime": 16.6651, | |
| "eval_samples_per_second": 14.881, | |
| "eval_steps_per_second": 0.96, | |
| "step": 496 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 1.6780645161290323e-05, | |
| "loss": 2.7011, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.279834270477295, | |
| "eval_rouge1": 0.1969, | |
| "eval_rouge2": 0.0946, | |
| "eval_rougeL": 0.1662, | |
| "eval_rougeLsum": 0.1665, | |
| "eval_runtime": 17.134, | |
| "eval_samples_per_second": 14.474, | |
| "eval_steps_per_second": 0.934, | |
| "step": 558 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.265596389770508, | |
| "eval_rouge1": 0.1987, | |
| "eval_rouge2": 0.0962, | |
| "eval_rougeL": 0.1672, | |
| "eval_rougeLsum": 0.1673, | |
| "eval_runtime": 17.1955, | |
| "eval_samples_per_second": 14.422, | |
| "eval_steps_per_second": 0.93, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.2547566890716553, | |
| "eval_rouge1": 0.1958, | |
| "eval_rouge2": 0.0965, | |
| "eval_rougeL": 0.1655, | |
| "eval_rougeLsum": 0.1657, | |
| "eval_runtime": 16.9264, | |
| "eval_samples_per_second": 14.652, | |
| "eval_steps_per_second": 0.945, | |
| "step": 682 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.243624210357666, | |
| "eval_rouge1": 0.1965, | |
| "eval_rouge2": 0.096, | |
| "eval_rougeL": 0.1659, | |
| "eval_rougeLsum": 0.166, | |
| "eval_runtime": 16.9191, | |
| "eval_samples_per_second": 14.658, | |
| "eval_steps_per_second": 0.946, | |
| "step": 744 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.2352294921875, | |
| "eval_rouge1": 0.1964, | |
| "eval_rouge2": 0.0971, | |
| "eval_rougeL": 0.1663, | |
| "eval_rougeLsum": 0.1664, | |
| "eval_runtime": 16.9525, | |
| "eval_samples_per_second": 14.629, | |
| "eval_steps_per_second": 0.944, | |
| "step": 806 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.2252049446105957, | |
| "eval_rouge1": 0.197, | |
| "eval_rouge2": 0.097, | |
| "eval_rougeL": 0.1664, | |
| "eval_rougeLsum": 0.1664, | |
| "eval_runtime": 16.904, | |
| "eval_samples_per_second": 14.671, | |
| "eval_steps_per_second": 0.947, | |
| "step": 868 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.2152445316314697, | |
| "eval_rouge1": 0.1954, | |
| "eval_rouge2": 0.0979, | |
| "eval_rougeL": 0.1664, | |
| "eval_rougeLsum": 0.1665, | |
| "eval_runtime": 17.117, | |
| "eval_samples_per_second": 14.489, | |
| "eval_steps_per_second": 0.935, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.207918643951416, | |
| "eval_rouge1": 0.1954, | |
| "eval_rouge2": 0.0988, | |
| "eval_rougeL": 0.1674, | |
| "eval_rougeLsum": 0.1677, | |
| "eval_runtime": 17.2029, | |
| "eval_samples_per_second": 14.416, | |
| "eval_steps_per_second": 0.93, | |
| "step": 992 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "learning_rate": 1.355483870967742e-05, | |
| "loss": 2.3282, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.1986870765686035, | |
| "eval_rouge1": 0.1951, | |
| "eval_rouge2": 0.0995, | |
| "eval_rougeL": 0.1672, | |
| "eval_rougeLsum": 0.1673, | |
| "eval_runtime": 17.1147, | |
| "eval_samples_per_second": 14.49, | |
| "eval_steps_per_second": 0.935, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.1938998699188232, | |
| "eval_rouge1": 0.1974, | |
| "eval_rouge2": 0.1015, | |
| "eval_rougeL": 0.1695, | |
| "eval_rougeLsum": 0.1697, | |
| "eval_runtime": 16.7909, | |
| "eval_samples_per_second": 14.77, | |
| "eval_steps_per_second": 0.953, | |
| "step": 1116 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.18984055519104, | |
| "eval_rouge1": 0.1965, | |
| "eval_rouge2": 0.1014, | |
| "eval_rougeL": 0.1691, | |
| "eval_rougeLsum": 0.1693, | |
| "eval_runtime": 16.6689, | |
| "eval_samples_per_second": 14.878, | |
| "eval_steps_per_second": 0.96, | |
| "step": 1178 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.183218240737915, | |
| "eval_rouge1": 0.1963, | |
| "eval_rouge2": 0.0997, | |
| "eval_rougeL": 0.1683, | |
| "eval_rougeLsum": 0.1685, | |
| "eval_runtime": 17.3129, | |
| "eval_samples_per_second": 14.325, | |
| "eval_steps_per_second": 0.924, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.1765005588531494, | |
| "eval_rouge1": 0.1966, | |
| "eval_rouge2": 0.0991, | |
| "eval_rougeL": 0.1676, | |
| "eval_rougeLsum": 0.1678, | |
| "eval_runtime": 16.8703, | |
| "eval_samples_per_second": 14.7, | |
| "eval_steps_per_second": 0.948, | |
| "step": 1302 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.1725897789001465, | |
| "eval_rouge1": 0.1963, | |
| "eval_rouge2": 0.0989, | |
| "eval_rougeL": 0.1677, | |
| "eval_rougeLsum": 0.1676, | |
| "eval_runtime": 16.7813, | |
| "eval_samples_per_second": 14.778, | |
| "eval_steps_per_second": 0.953, | |
| "step": 1364 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.1676828861236572, | |
| "eval_rouge1": 0.1959, | |
| "eval_rouge2": 0.0988, | |
| "eval_rougeL": 0.168, | |
| "eval_rougeLsum": 0.168, | |
| "eval_runtime": 17.3121, | |
| "eval_samples_per_second": 14.325, | |
| "eval_steps_per_second": 0.924, | |
| "step": 1426 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.1647536754608154, | |
| "eval_rouge1": 0.1967, | |
| "eval_rouge2": 0.0994, | |
| "eval_rougeL": 0.169, | |
| "eval_rougeLsum": 0.1692, | |
| "eval_runtime": 16.9203, | |
| "eval_samples_per_second": 14.657, | |
| "eval_steps_per_second": 0.946, | |
| "step": 1488 | |
| }, | |
| { | |
| "epoch": 24.19, | |
| "learning_rate": 1.0329032258064518e-05, | |
| "loss": 2.2281, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.15854549407959, | |
| "eval_rouge1": 0.1958, | |
| "eval_rouge2": 0.0988, | |
| "eval_rougeL": 0.1685, | |
| "eval_rougeLsum": 0.1687, | |
| "eval_runtime": 17.1171, | |
| "eval_samples_per_second": 14.488, | |
| "eval_steps_per_second": 0.935, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.1557765007019043, | |
| "eval_rouge1": 0.197, | |
| "eval_rouge2": 0.1, | |
| "eval_rougeL": 0.1698, | |
| "eval_rougeLsum": 0.1699, | |
| "eval_runtime": 16.8981, | |
| "eval_samples_per_second": 14.676, | |
| "eval_steps_per_second": 0.947, | |
| "step": 1612 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.1530044078826904, | |
| "eval_rouge1": 0.196, | |
| "eval_rouge2": 0.0994, | |
| "eval_rougeL": 0.1685, | |
| "eval_rougeLsum": 0.1687, | |
| "eval_runtime": 16.9499, | |
| "eval_samples_per_second": 14.631, | |
| "eval_steps_per_second": 0.944, | |
| "step": 1674 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.1497113704681396, | |
| "eval_rouge1": 0.1971, | |
| "eval_rouge2": 0.101, | |
| "eval_rougeL": 0.1697, | |
| "eval_rougeLsum": 0.1699, | |
| "eval_runtime": 16.8885, | |
| "eval_samples_per_second": 14.685, | |
| "eval_steps_per_second": 0.947, | |
| "step": 1736 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.1459004878997803, | |
| "eval_rouge1": 0.1972, | |
| "eval_rouge2": 0.1008, | |
| "eval_rougeL": 0.17, | |
| "eval_rougeLsum": 0.1701, | |
| "eval_runtime": 17.5571, | |
| "eval_samples_per_second": 14.125, | |
| "eval_steps_per_second": 0.911, | |
| "step": 1798 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.142939805984497, | |
| "eval_rouge1": 0.1946, | |
| "eval_rouge2": 0.0989, | |
| "eval_rougeL": 0.1677, | |
| "eval_rougeLsum": 0.1678, | |
| "eval_runtime": 17.4108, | |
| "eval_samples_per_second": 14.244, | |
| "eval_steps_per_second": 0.919, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.14223051071167, | |
| "eval_rouge1": 0.1958, | |
| "eval_rouge2": 0.1, | |
| "eval_rougeL": 0.1691, | |
| "eval_rougeLsum": 0.1692, | |
| "eval_runtime": 17.3596, | |
| "eval_samples_per_second": 14.286, | |
| "eval_steps_per_second": 0.922, | |
| "step": 1922 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_gen_len": 19.0, | |
| "eval_loss": 2.139946699142456, | |
| "eval_rouge1": 0.1952, | |
| "eval_rouge2": 0.0992, | |
| "eval_rougeL": 0.1687, | |
| "eval_rougeLsum": 0.1687, | |
| "eval_runtime": 17.1971, | |
| "eval_samples_per_second": 14.421, | |
| "eval_steps_per_second": 0.93, | |
| "step": 1984 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "learning_rate": 7.103225806451613e-06, | |
| "loss": 2.1696, | |
| "step": 2000 | |
| } | |
| ], | |
| "max_steps": 3100, | |
| "num_train_epochs": 50, | |
| "total_flos": 8635889668325376.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |