| { | |
| "best_global_step": 244, | |
| "best_metric": 0.25023409724235535, | |
| "best_model_checkpoint": "./lora_qwen32b_java_abdiff_v1/checkpoint-244", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 732, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.041109969167523124, | |
| "grad_norm": 0.8196802735328674, | |
| "learning_rate": 3.648648648648649e-06, | |
| "loss": 0.9434, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08221993833504625, | |
| "grad_norm": 0.44579723477363586, | |
| "learning_rate": 7.702702702702703e-06, | |
| "loss": 0.9215, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12332990750256938, | |
| "grad_norm": 0.41646608710289, | |
| "learning_rate": 1.1756756756756757e-05, | |
| "loss": 0.8404, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1644398766700925, | |
| "grad_norm": 0.39053428173065186, | |
| "learning_rate": 1.5810810810810808e-05, | |
| "loss": 0.7482, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.20554984583761562, | |
| "grad_norm": 0.2627861797809601, | |
| "learning_rate": 1.9864864864864866e-05, | |
| "loss": 0.6263, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.24665981500513876, | |
| "grad_norm": 0.2621787488460541, | |
| "learning_rate": 2.3918918918918917e-05, | |
| "loss": 0.5306, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.28776978417266186, | |
| "grad_norm": 0.19896556437015533, | |
| "learning_rate": 2.7972972972972975e-05, | |
| "loss": 0.4205, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.328879753340185, | |
| "grad_norm": 0.20418603718280792, | |
| "learning_rate": 2.9772036474164135e-05, | |
| "loss": 0.3725, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.3699897225077081, | |
| "grad_norm": 0.16360916197299957, | |
| "learning_rate": 2.9316109422492404e-05, | |
| "loss": 0.3384, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.41109969167523125, | |
| "grad_norm": 0.13916964828968048, | |
| "learning_rate": 2.886018237082067e-05, | |
| "loss": 0.3088, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4522096608427544, | |
| "grad_norm": 0.15655504167079926, | |
| "learning_rate": 2.8404255319148935e-05, | |
| "loss": 0.2883, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.4933196300102775, | |
| "grad_norm": 0.1462835967540741, | |
| "learning_rate": 2.7948328267477204e-05, | |
| "loss": 0.2884, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5344295991778006, | |
| "grad_norm": 0.1710139513015747, | |
| "learning_rate": 2.7492401215805473e-05, | |
| "loss": 0.2752, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.5755395683453237, | |
| "grad_norm": 0.17515398561954498, | |
| "learning_rate": 2.7036474164133738e-05, | |
| "loss": 0.2525, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.6166495375128469, | |
| "grad_norm": 0.16058574616909027, | |
| "learning_rate": 2.6580547112462007e-05, | |
| "loss": 0.2512, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.65775950668037, | |
| "grad_norm": 0.208830326795578, | |
| "learning_rate": 2.6124620060790272e-05, | |
| "loss": 0.2483, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.6988694758478932, | |
| "grad_norm": 0.15300804376602173, | |
| "learning_rate": 2.566869300911854e-05, | |
| "loss": 0.2448, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.7399794450154162, | |
| "grad_norm": 0.21872150897979736, | |
| "learning_rate": 2.521276595744681e-05, | |
| "loss": 0.2269, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.7810894141829393, | |
| "grad_norm": 0.14277832210063934, | |
| "learning_rate": 2.4756838905775076e-05, | |
| "loss": 0.2364, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.8221993833504625, | |
| "grad_norm": 0.1629635989665985, | |
| "learning_rate": 2.4300911854103345e-05, | |
| "loss": 0.2248, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.8633093525179856, | |
| "grad_norm": 0.170692577958107, | |
| "learning_rate": 2.3844984802431613e-05, | |
| "loss": 0.2077, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.9044193216855088, | |
| "grad_norm": 0.168124720454216, | |
| "learning_rate": 2.338905775075988e-05, | |
| "loss": 0.215, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.9455292908530318, | |
| "grad_norm": 0.1866914927959442, | |
| "learning_rate": 2.2933130699088144e-05, | |
| "loss": 0.1893, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.986639260020555, | |
| "grad_norm": 0.17521585524082184, | |
| "learning_rate": 2.2477203647416413e-05, | |
| "loss": 0.1922, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.25023409724235535, | |
| "eval_runtime": 335.9934, | |
| "eval_samples_per_second": 1.941, | |
| "eval_steps_per_second": 0.97, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 1.024665981500514, | |
| "grad_norm": 0.15969552099704742, | |
| "learning_rate": 2.2021276595744682e-05, | |
| "loss": 0.1919, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.065775950668037, | |
| "grad_norm": 0.162578746676445, | |
| "learning_rate": 2.156534954407295e-05, | |
| "loss": 0.1762, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.10688591983556, | |
| "grad_norm": 0.2019297331571579, | |
| "learning_rate": 2.1109422492401216e-05, | |
| "loss": 0.1768, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.1479958890030832, | |
| "grad_norm": 0.15636153519153595, | |
| "learning_rate": 2.0653495440729482e-05, | |
| "loss": 0.1577, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.1891058581706064, | |
| "grad_norm": 0.21196545660495758, | |
| "learning_rate": 2.019756838905775e-05, | |
| "loss": 0.1594, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.2302158273381294, | |
| "grad_norm": 0.17270150780677795, | |
| "learning_rate": 1.974164133738602e-05, | |
| "loss": 0.1615, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.2713257965056526, | |
| "grad_norm": 0.2115948349237442, | |
| "learning_rate": 1.928571428571429e-05, | |
| "loss": 0.1602, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.3124357656731758, | |
| "grad_norm": 0.20007292926311493, | |
| "learning_rate": 1.8829787234042554e-05, | |
| "loss": 0.1508, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.353545734840699, | |
| "grad_norm": 0.8320184350013733, | |
| "learning_rate": 1.837386018237082e-05, | |
| "loss": 0.1424, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.394655704008222, | |
| "grad_norm": 0.2084386944770813, | |
| "learning_rate": 1.7917933130699088e-05, | |
| "loss": 0.1384, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.435765673175745, | |
| "grad_norm": 0.2273188680410385, | |
| "learning_rate": 1.7462006079027357e-05, | |
| "loss": 0.127, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.4768756423432683, | |
| "grad_norm": 0.2669423818588257, | |
| "learning_rate": 1.7006079027355622e-05, | |
| "loss": 0.1265, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.5179856115107913, | |
| "grad_norm": 0.2080197036266327, | |
| "learning_rate": 1.655015197568389e-05, | |
| "loss": 0.124, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.5590955806783144, | |
| "grad_norm": 0.2571563422679901, | |
| "learning_rate": 1.609422492401216e-05, | |
| "loss": 0.1242, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.6002055498458376, | |
| "grad_norm": 0.2905978858470917, | |
| "learning_rate": 1.5638297872340426e-05, | |
| "loss": 0.1248, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.6413155190133608, | |
| "grad_norm": 0.3066595792770386, | |
| "learning_rate": 1.5182370820668691e-05, | |
| "loss": 0.1141, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.682425488180884, | |
| "grad_norm": 0.22845837473869324, | |
| "learning_rate": 1.4726443768996962e-05, | |
| "loss": 0.1122, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.723535457348407, | |
| "grad_norm": 0.2635312080383301, | |
| "learning_rate": 1.4270516717325229e-05, | |
| "loss": 0.1129, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.7646454265159301, | |
| "grad_norm": 0.3327837884426117, | |
| "learning_rate": 1.3814589665653496e-05, | |
| "loss": 0.0988, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.8057553956834531, | |
| "grad_norm": 0.30640456080436707, | |
| "learning_rate": 1.3358662613981763e-05, | |
| "loss": 0.1093, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.8468653648509763, | |
| "grad_norm": 0.28711435198783875, | |
| "learning_rate": 1.2902735562310032e-05, | |
| "loss": 0.1012, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.8879753340184995, | |
| "grad_norm": 0.29652899503707886, | |
| "learning_rate": 1.2446808510638298e-05, | |
| "loss": 0.0846, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.9290853031860227, | |
| "grad_norm": 0.29864785075187683, | |
| "learning_rate": 1.1990881458966566e-05, | |
| "loss": 0.0953, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.9701952723535459, | |
| "grad_norm": 0.2623487412929535, | |
| "learning_rate": 1.1534954407294832e-05, | |
| "loss": 0.0884, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.2823970913887024, | |
| "eval_runtime": 336.941, | |
| "eval_samples_per_second": 1.935, | |
| "eval_steps_per_second": 0.968, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 2.0082219938335046, | |
| "grad_norm": 0.2646639943122864, | |
| "learning_rate": 1.10790273556231e-05, | |
| "loss": 0.0835, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.049331963001028, | |
| "grad_norm": 0.2684798240661621, | |
| "learning_rate": 1.0623100303951368e-05, | |
| "loss": 0.0838, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.090441932168551, | |
| "grad_norm": 0.26659277081489563, | |
| "learning_rate": 1.0167173252279635e-05, | |
| "loss": 0.0628, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.131551901336074, | |
| "grad_norm": 0.26738440990448, | |
| "learning_rate": 9.711246200607902e-06, | |
| "loss": 0.0765, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.172661870503597, | |
| "grad_norm": 0.2757914960384369, | |
| "learning_rate": 9.255319148936171e-06, | |
| "loss": 0.0787, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.21377183967112, | |
| "grad_norm": 0.28704580664634705, | |
| "learning_rate": 8.799392097264438e-06, | |
| "loss": 0.069, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.2548818088386433, | |
| "grad_norm": 0.30440813302993774, | |
| "learning_rate": 8.343465045592705e-06, | |
| "loss": 0.0777, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.2959917780061665, | |
| "grad_norm": 0.30448320508003235, | |
| "learning_rate": 7.887537993920974e-06, | |
| "loss": 0.0603, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.3371017471736897, | |
| "grad_norm": 0.31719279289245605, | |
| "learning_rate": 7.43161094224924e-06, | |
| "loss": 0.0644, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.378211716341213, | |
| "grad_norm": 0.26711922883987427, | |
| "learning_rate": 6.975683890577508e-06, | |
| "loss": 0.0611, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.419321685508736, | |
| "grad_norm": 0.22036238014698029, | |
| "learning_rate": 6.519756838905775e-06, | |
| "loss": 0.066, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.460431654676259, | |
| "grad_norm": 0.2766261696815491, | |
| "learning_rate": 6.063829787234042e-06, | |
| "loss": 0.0572, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.501541623843782, | |
| "grad_norm": 0.27715882658958435, | |
| "learning_rate": 5.607902735562311e-06, | |
| "loss": 0.0634, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.542651593011305, | |
| "grad_norm": 0.26951295137405396, | |
| "learning_rate": 5.151975683890578e-06, | |
| "loss": 0.0527, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.5837615621788284, | |
| "grad_norm": 0.3271823227405548, | |
| "learning_rate": 4.696048632218845e-06, | |
| "loss": 0.0627, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.6248715313463515, | |
| "grad_norm": 0.347106009721756, | |
| "learning_rate": 4.240121580547112e-06, | |
| "loss": 0.0553, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.6659815005138747, | |
| "grad_norm": 0.21918581426143646, | |
| "learning_rate": 3.7841945288753804e-06, | |
| "loss": 0.0553, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.707091469681398, | |
| "grad_norm": 0.20088477432727814, | |
| "learning_rate": 3.3282674772036475e-06, | |
| "loss": 0.0658, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.7482014388489207, | |
| "grad_norm": 0.21134094893932343, | |
| "learning_rate": 2.872340425531915e-06, | |
| "loss": 0.0494, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.789311408016444, | |
| "grad_norm": 0.315378874540329, | |
| "learning_rate": 2.4164133738601823e-06, | |
| "loss": 0.0598, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.830421377183967, | |
| "grad_norm": 0.2581155300140381, | |
| "learning_rate": 1.96048632218845e-06, | |
| "loss": 0.059, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.87153134635149, | |
| "grad_norm": 0.27886152267456055, | |
| "learning_rate": 1.5045592705167174e-06, | |
| "loss": 0.0518, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.9126413155190134, | |
| "grad_norm": 0.2675166726112366, | |
| "learning_rate": 1.0486322188449848e-06, | |
| "loss": 0.0536, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.9537512846865366, | |
| "grad_norm": 0.24460436403751373, | |
| "learning_rate": 5.927051671732523e-07, | |
| "loss": 0.0464, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.9948612538540598, | |
| "grad_norm": 0.23885396122932434, | |
| "learning_rate": 1.3677811550151974e-07, | |
| "loss": 0.0589, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.2979394197463989, | |
| "eval_runtime": 335.8215, | |
| "eval_samples_per_second": 1.942, | |
| "eval_steps_per_second": 0.971, | |
| "step": 732 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 732, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.082850061982306e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |