| { | |
| "best_metric": 0.6989061236381531, | |
| "best_model_checkpoint": "/mnt/bn/qingyi-bn-lq/llama/saved-cot7b/checkpoint-1600", | |
| "epoch": 2.813805231919103, | |
| "global_step": 1600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 5.9999999999999995e-05, | |
| "loss": 2.3604, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 1.944, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 1.2746, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 0.9818, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0003, | |
| "loss": 0.863, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0002962593516209476, | |
| "loss": 0.8083, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00029251870324189524, | |
| "loss": 0.7923, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00028877805486284284, | |
| "loss": 0.7623, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0002850374064837905, | |
| "loss": 0.7665, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0002812967581047381, | |
| "loss": 0.7554, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_loss": 0.7609586715698242, | |
| "eval_runtime": 32.1749, | |
| "eval_samples_per_second": 62.16, | |
| "eval_steps_per_second": 1.958, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.00027755610972568577, | |
| "loss": 0.7353, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0002738154613466334, | |
| "loss": 0.7354, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00027007481296758103, | |
| "loss": 0.7435, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00026633416458852864, | |
| "loss": 0.7373, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0002625935162094763, | |
| "loss": 0.7332, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0002588528678304239, | |
| "loss": 0.7265, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00025511221945137156, | |
| "loss": 0.7274, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00025137157107231917, | |
| "loss": 0.7279, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.00024763092269326683, | |
| "loss": 0.7231, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00024389027431421443, | |
| "loss": 0.7238, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 0.7300755381584167, | |
| "eval_runtime": 32.1181, | |
| "eval_samples_per_second": 62.27, | |
| "eval_steps_per_second": 1.962, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.00024014962593516207, | |
| "loss": 0.7161, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0002364089775561097, | |
| "loss": 0.7146, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.00023266832917705733, | |
| "loss": 0.7152, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.00022892768079800496, | |
| "loss": 0.7202, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.0002251870324189526, | |
| "loss": 0.7142, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.00022144638403990023, | |
| "loss": 0.7146, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.00021770573566084786, | |
| "loss": 0.7104, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.0002139650872817955, | |
| "loss": 0.7034, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.00021022443890274313, | |
| "loss": 0.7153, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.00020648379052369076, | |
| "loss": 0.7052, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "eval_loss": 0.7185753583908081, | |
| "eval_runtime": 32.4703, | |
| "eval_samples_per_second": 61.595, | |
| "eval_steps_per_second": 1.94, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002027431421446384, | |
| "loss": 0.7061, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.00019900249376558603, | |
| "loss": 0.7096, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.00019526184538653366, | |
| "loss": 0.7065, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00019152119700748126, | |
| "loss": 0.7046, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.0001877805486284289, | |
| "loss": 0.701, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 0.00018403990024937653, | |
| "loss": 0.6922, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 0.00018029925187032416, | |
| "loss": 0.6982, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.0001765586034912718, | |
| "loss": 0.6993, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 0.00017281795511221943, | |
| "loss": 0.6922, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 0.00016907730673316706, | |
| "loss": 0.6989, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_loss": 0.7114558219909668, | |
| "eval_runtime": 32.0158, | |
| "eval_samples_per_second": 62.469, | |
| "eval_steps_per_second": 1.968, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.0001653366583541147, | |
| "loss": 0.6964, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 0.00016159600997506232, | |
| "loss": 0.6969, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.00015785536159600996, | |
| "loss": 0.6982, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 0.0001541147132169576, | |
| "loss": 0.6977, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 0.00015037406483790522, | |
| "loss": 0.7019, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.00014663341645885285, | |
| "loss": 0.6963, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 0.00014289276807980049, | |
| "loss": 0.7006, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00013915211970074812, | |
| "loss": 0.6935, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.00013541147132169575, | |
| "loss": 0.6846, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 0.00013167082294264338, | |
| "loss": 0.701, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "eval_loss": 0.7069133520126343, | |
| "eval_runtime": 31.9943, | |
| "eval_samples_per_second": 62.511, | |
| "eval_steps_per_second": 1.969, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.00012793017456359102, | |
| "loss": 0.6896, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 0.00012418952618453862, | |
| "loss": 0.702, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 0.00012044887780548627, | |
| "loss": 0.6952, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 0.0001167082294264339, | |
| "loss": 0.6902, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 0.00011296758104738153, | |
| "loss": 0.6866, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.00010922693266832918, | |
| "loss": 0.6929, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00010548628428927681, | |
| "loss": 0.6846, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.00010174563591022444, | |
| "loss": 0.6944, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 9.800498753117206e-05, | |
| "loss": 0.6868, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 9.42643391521197e-05, | |
| "loss": 0.6938, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 0.7033773064613342, | |
| "eval_runtime": 32.1557, | |
| "eval_samples_per_second": 62.197, | |
| "eval_steps_per_second": 1.959, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 9.052369077306733e-05, | |
| "loss": 0.6868, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 8.678304239401496e-05, | |
| "loss": 0.6795, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 8.304239401496259e-05, | |
| "loss": 0.6887, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 7.930174563591023e-05, | |
| "loss": 0.6795, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 7.556109725685786e-05, | |
| "loss": 0.6934, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 7.182044887780548e-05, | |
| "loss": 0.6905, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 6.807980049875311e-05, | |
| "loss": 0.685, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 6.433915211970074e-05, | |
| "loss": 0.6887, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 6.0598503740648375e-05, | |
| "loss": 0.6875, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 5.6857855361596e-05, | |
| "loss": 0.6807, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "eval_loss": 0.7009322643280029, | |
| "eval_runtime": 32.0972, | |
| "eval_samples_per_second": 62.311, | |
| "eval_steps_per_second": 1.963, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 5.311720698254363e-05, | |
| "loss": 0.6743, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 4.9376558603491265e-05, | |
| "loss": 0.6872, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 4.56359102244389e-05, | |
| "loss": 0.6776, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 4.1895261845386524e-05, | |
| "loss": 0.6769, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 3.8154613466334156e-05, | |
| "loss": 0.68, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 3.4413965087281796e-05, | |
| "loss": 0.6804, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 3.067331670822942e-05, | |
| "loss": 0.6848, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 2.6932668329177054e-05, | |
| "loss": 0.6825, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 2.3192019950124686e-05, | |
| "loss": 0.6899, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 1.945137157107232e-05, | |
| "loss": 0.6775, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "eval_loss": 0.6989061236381531, | |
| "eval_runtime": 32.1279, | |
| "eval_samples_per_second": 62.251, | |
| "eval_steps_per_second": 1.961, | |
| "step": 1600 | |
| } | |
| ], | |
| "max_steps": 1704, | |
| "num_train_epochs": 3, | |
| "total_flos": 4.159132701791617e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |