| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9997283346916599, | |
| "global_step": 115, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1341, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.982307470588098e-05, | |
| "loss": 0.1267, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.92935509259118e-05, | |
| "loss": 0.1213, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.841517610611309e-05, | |
| "loss": 0.1211, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 9.719416651541839e-05, | |
| "loss": 0.1172, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.563916325306594e-05, | |
| "loss": 0.1195, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_gen_len": 121.8, | |
| "eval_loss": 2.063453435897827, | |
| "eval_rouge1": 57.8829, | |
| "eval_rouge2": 28.7887, | |
| "eval_rougeL": 44.4256, | |
| "eval_rougeLsum": 54.1299, | |
| "eval_runtime": 3417.7231, | |
| "eval_samples_per_second": 0.293, | |
| "eval_steps_per_second": 0.073, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 9.376117109543769e-05, | |
| "loss": 0.1129, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 0.1161, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 8.90915741234015e-05, | |
| "loss": 0.1184, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 8.633301610170135e-05, | |
| "loss": 0.1038, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 8.33173288976002e-05, | |
| "loss": 0.1064, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 8.006585456492029e-05, | |
| "loss": 0.1107, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 7.660160382576683e-05, | |
| "loss": 0.1084, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_gen_len": 122.893, | |
| "eval_loss": 2.117846965789795, | |
| "eval_rouge1": 58.0416, | |
| "eval_rouge2": 28.6487, | |
| "eval_rougeL": 44.3905, | |
| "eval_rougeLsum": 54.1557, | |
| "eval_runtime": 3584.8893, | |
| "eval_samples_per_second": 0.279, | |
| "eval_steps_per_second": 0.07, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 7.294909322337689e-05, | |
| "loss": 0.1026, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.1059, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 6.518383725548074e-05, | |
| "loss": 0.1078, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 6.112604669781572e-05, | |
| "loss": 0.1053, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 5.698951697677498e-05, | |
| "loss": 0.1027, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 5.2803522361859594e-05, | |
| "loss": 0.0993, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4.859768718620656e-05, | |
| "loss": 0.1019, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_gen_len": 120.524, | |
| "eval_loss": 2.157585382461548, | |
| "eval_rouge1": 57.816, | |
| "eval_rouge2": 28.7069, | |
| "eval_rougeL": 44.4242, | |
| "eval_rougeLsum": 53.9598, | |
| "eval_runtime": 3448.4317, | |
| "eval_samples_per_second": 0.29, | |
| "eval_steps_per_second": 0.072, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.4401776194834613e-05, | |
| "loss": 0.0959, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.1051, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 3.6158224428757535e-05, | |
| "loss": 0.0953, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 3.216892336688435e-05, | |
| "loss": 0.0952, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 2.8305813044122097e-05, | |
| "loss": 0.0957, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 2.459623273767354e-05, | |
| "loss": 0.0975, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_gen_len": 121.793, | |
| "eval_loss": 2.182112693786621, | |
| "eval_rouge1": 57.9597, | |
| "eval_rouge2": 28.8178, | |
| "eval_rougeL": 44.4854, | |
| "eval_rougeLsum": 54.068, | |
| "eval_runtime": 3498.8291, | |
| "eval_samples_per_second": 0.286, | |
| "eval_steps_per_second": 0.071, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 2.1066435191009715e-05, | |
| "loss": 0.0943, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.774140082289563e-05, | |
| "loss": 0.0988, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0954, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 1.1798131208919627e-05, | |
| "loss": 0.0977, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 9.221956552036992e-06, | |
| "loss": 0.0954, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 6.934368588379553e-06, | |
| "loss": 0.0942, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 4.951556604879048e-06, | |
| "loss": 0.0947, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_gen_len": 122.77, | |
| "eval_loss": 2.1846070289611816, | |
| "eval_rouge1": 57.9637, | |
| "eval_rouge2": 28.7446, | |
| "eval_rougeL": 44.3826, | |
| "eval_rougeLsum": 54.0399, | |
| "eval_runtime": 3495.1079, | |
| "eval_samples_per_second": 0.286, | |
| "eval_steps_per_second": 0.072, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 3.2875529852700147e-06, | |
| "loss": 0.0952, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 1.9541339027450256e-06, | |
| "loss": 0.1001, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 0.0953, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 3.143895053378698e-07, | |
| "loss": 0.0963, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 1.9668680847356735e-08, | |
| "loss": 0.0924, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 115, | |
| "total_flos": 1.3570473694593024e+17, | |
| "train_loss": 0.10453087175669877, | |
| "train_runtime": 27358.0249, | |
| "train_samples_per_second": 1.076, | |
| "train_steps_per_second": 0.004 | |
| } | |
| ], | |
| "max_steps": 115, | |
| "num_train_epochs": 1, | |
| "total_flos": 1.3570473694593024e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |