| { | |
| "best_metric": 2.2772634029388428, | |
| "best_model_checkpoint": "/tmp/model/checkpoint-1600", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 1600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 3.958333333333333e-06, | |
| "loss": 3.8721, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 8.125000000000001e-06, | |
| "loss": 3.3675, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.2291666666666666e-05, | |
| "loss": 3.5553, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.6458333333333335e-05, | |
| "loss": 3.2187, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 2.0625e-05, | |
| "loss": 3.3576, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 2.479166666666667e-05, | |
| "loss": 3.2111, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 2.8958333333333337e-05, | |
| "loss": 3.1881, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.3125e-05, | |
| "loss": 3.0568, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 3.708333333333334e-05, | |
| "loss": 3.1593, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.125e-05, | |
| "loss": 2.9964, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 4.541666666666667e-05, | |
| "loss": 3.2086, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.958333333333334e-05, | |
| "loss": 3.314, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.958333333333334e-05, | |
| "loss": 3.1285, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.912037037037037e-05, | |
| "loss": 2.8458, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 4.865740740740741e-05, | |
| "loss": 2.6054, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.819444444444445e-05, | |
| "loss": 3.1717, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.773148148148148e-05, | |
| "loss": 3.2007, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.726851851851852e-05, | |
| "loss": 3.0557, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.6805555555555556e-05, | |
| "loss": 2.8199, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.6342592592592595e-05, | |
| "loss": 2.8118, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.590277777777778e-05, | |
| "loss": 2.9147, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.543981481481482e-05, | |
| "loss": 2.6742, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.497685185185185e-05, | |
| "loss": 2.8732, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.4513888888888885e-05, | |
| "loss": 2.6045, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.405092592592593e-05, | |
| "loss": 2.7271, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 4.3587962962962965e-05, | |
| "loss": 2.5203, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 4.3125000000000005e-05, | |
| "loss": 2.7714, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.266203703703704e-05, | |
| "loss": 3.081, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 4.219907407407408e-05, | |
| "loss": 2.6275, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.173611111111112e-05, | |
| "loss": 2.8672, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.127314814814815e-05, | |
| "loss": 2.6542, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.0810185185185184e-05, | |
| "loss": 2.9697, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.0347222222222223e-05, | |
| "loss": 2.8099, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3.988425925925926e-05, | |
| "loss": 2.6577, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 3.94212962962963e-05, | |
| "loss": 2.7843, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 3.8958333333333336e-05, | |
| "loss": 2.7663, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 3.849537037037037e-05, | |
| "loss": 2.6443, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3.803240740740741e-05, | |
| "loss": 2.6549, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 3.756944444444445e-05, | |
| "loss": 2.4063, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.710648148148148e-05, | |
| "loss": 2.5605, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 141.655, | |
| "eval_loss": 2.4260785579681396, | |
| "eval_rouge1": 36.8904, | |
| "eval_rouge2": 14.0444, | |
| "eval_rougeL": 22.5839, | |
| "eval_rougeLsum": 33.8603, | |
| "eval_runtime": 806.3061, | |
| "eval_samples_per_second": 0.248, | |
| "eval_steps_per_second": 0.062, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 3.664351851851852e-05, | |
| "loss": 2.3288, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 3.6180555555555555e-05, | |
| "loss": 2.0535, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 3.5717592592592595e-05, | |
| "loss": 2.0489, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 3.5254629629629635e-05, | |
| "loss": 1.9129, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 3.479166666666667e-05, | |
| "loss": 2.3016, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 3.432870370370371e-05, | |
| "loss": 2.1098, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 3.386574074074074e-05, | |
| "loss": 2.0364, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 3.340277777777778e-05, | |
| "loss": 1.8088, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 3.293981481481482e-05, | |
| "loss": 1.9104, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 3.247685185185185e-05, | |
| "loss": 1.9855, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 3.2013888888888886e-05, | |
| "loss": 2.0865, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 3.1550925925925926e-05, | |
| "loss": 2.127, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 3.1087962962962966e-05, | |
| "loss": 2.1918, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 3.0625000000000006e-05, | |
| "loss": 1.9604, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 3.016203703703704e-05, | |
| "loss": 2.289, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 2.9699074074074075e-05, | |
| "loss": 1.6954, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 2.9236111111111115e-05, | |
| "loss": 2.2102, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 2.8773148148148148e-05, | |
| "loss": 2.287, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 2.8310185185185185e-05, | |
| "loss": 1.9731, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 2.7847222222222224e-05, | |
| "loss": 2.2276, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 2.738425925925926e-05, | |
| "loss": 1.8473, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 2.69212962962963e-05, | |
| "loss": 1.8407, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 2.6458333333333334e-05, | |
| "loss": 1.9851, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 2.599537037037037e-05, | |
| "loss": 2.2261, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 2.553240740740741e-05, | |
| "loss": 1.9708, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 2.5069444444444447e-05, | |
| "loss": 1.9759, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 2.4606481481481483e-05, | |
| "loss": 1.9717, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 2.414351851851852e-05, | |
| "loss": 1.8061, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 2.3680555555555556e-05, | |
| "loss": 1.7061, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 2.3217592592592592e-05, | |
| "loss": 2.0426, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 2.2754629629629632e-05, | |
| "loss": 2.0017, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 2.229166666666667e-05, | |
| "loss": 1.979, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 2.1828703703703705e-05, | |
| "loss": 1.9851, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 2.136574074074074e-05, | |
| "loss": 1.7156, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 2.0902777777777778e-05, | |
| "loss": 2.0078, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 2.0439814814814814e-05, | |
| "loss": 1.8058, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 1.997685185185185e-05, | |
| "loss": 2.3492, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 1.951388888888889e-05, | |
| "loss": 2.169, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 1.9050925925925927e-05, | |
| "loss": 1.9333, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.8587962962962964e-05, | |
| "loss": 2.0535, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 141.54, | |
| "eval_loss": 2.2772634029388428, | |
| "eval_rouge1": 39.2994, | |
| "eval_rouge2": 16.9505, | |
| "eval_rougeL": 24.9144, | |
| "eval_rougeLsum": 36.4135, | |
| "eval_runtime": 807.8602, | |
| "eval_samples_per_second": 0.248, | |
| "eval_steps_per_second": 0.062, | |
| "step": 1600 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 2400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 6931077737742336.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |