| { | |
| "best_global_step": 244, | |
| "best_metric": 0.30717727541923523, | |
| "best_model_checkpoint": "lora_qwen32b_java_abdiff_v2/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": 0.5229843258857727, | |
| "learning_rate": 3.648648648648649e-06, | |
| "loss": 1.149, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08221993833504625, | |
| "grad_norm": 0.5120038390159607, | |
| "learning_rate": 7.702702702702703e-06, | |
| "loss": 1.0995, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12332990750256938, | |
| "grad_norm": 0.4333497881889343, | |
| "learning_rate": 1.1756756756756757e-05, | |
| "loss": 1.0168, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1644398766700925, | |
| "grad_norm": 0.3843521475791931, | |
| "learning_rate": 1.5810810810810808e-05, | |
| "loss": 0.8955, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.20554984583761562, | |
| "grad_norm": 0.44873204827308655, | |
| "learning_rate": 1.9864864864864866e-05, | |
| "loss": 0.7296, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.24665981500513876, | |
| "grad_norm": 0.40769267082214355, | |
| "learning_rate": 2.3918918918918917e-05, | |
| "loss": 0.6065, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.28776978417266186, | |
| "grad_norm": 0.25337493419647217, | |
| "learning_rate": 2.7972972972972975e-05, | |
| "loss": 0.4745, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.328879753340185, | |
| "grad_norm": 0.2596264183521271, | |
| "learning_rate": 2.9772036474164135e-05, | |
| "loss": 0.4406, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.3699897225077081, | |
| "grad_norm": 0.260395348072052, | |
| "learning_rate": 2.9316109422492404e-05, | |
| "loss": 0.3933, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.41109969167523125, | |
| "grad_norm": 0.1907619684934616, | |
| "learning_rate": 2.886018237082067e-05, | |
| "loss": 0.368, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4522096608427544, | |
| "grad_norm": 0.18910694122314453, | |
| "learning_rate": 2.8404255319148935e-05, | |
| "loss": 0.3569, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.4933196300102775, | |
| "grad_norm": 0.21357780694961548, | |
| "learning_rate": 2.7948328267477204e-05, | |
| "loss": 0.3585, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5344295991778006, | |
| "grad_norm": 0.19042538106441498, | |
| "learning_rate": 2.7492401215805473e-05, | |
| "loss": 0.3236, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.5755395683453237, | |
| "grad_norm": 0.20597364008426666, | |
| "learning_rate": 2.7036474164133738e-05, | |
| "loss": 0.3176, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.6166495375128469, | |
| "grad_norm": 0.1616671234369278, | |
| "learning_rate": 2.6580547112462007e-05, | |
| "loss": 0.2951, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.65775950668037, | |
| "grad_norm": 0.19652485847473145, | |
| "learning_rate": 2.6124620060790272e-05, | |
| "loss": 0.3035, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.6988694758478932, | |
| "grad_norm": 0.16938744485378265, | |
| "learning_rate": 2.566869300911854e-05, | |
| "loss": 0.2986, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.7399794450154162, | |
| "grad_norm": 0.1938844472169876, | |
| "learning_rate": 2.521276595744681e-05, | |
| "loss": 0.2844, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.7810894141829393, | |
| "grad_norm": 0.19115138053894043, | |
| "learning_rate": 2.4756838905775076e-05, | |
| "loss": 0.2897, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.8221993833504625, | |
| "grad_norm": 0.20208393037319183, | |
| "learning_rate": 2.4300911854103345e-05, | |
| "loss": 0.2712, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.8633093525179856, | |
| "grad_norm": 0.19500800967216492, | |
| "learning_rate": 2.3844984802431613e-05, | |
| "loss": 0.258, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.9044193216855088, | |
| "grad_norm": 0.21097730100154877, | |
| "learning_rate": 2.338905775075988e-05, | |
| "loss": 0.251, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.9455292908530318, | |
| "grad_norm": 0.25240081548690796, | |
| "learning_rate": 2.2933130699088144e-05, | |
| "loss": 0.2338, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.986639260020555, | |
| "grad_norm": 0.24149411916732788, | |
| "learning_rate": 2.2477203647416413e-05, | |
| "loss": 0.2372, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.30717727541923523, | |
| "eval_runtime": 180.7589, | |
| "eval_samples_per_second": 3.607, | |
| "eval_steps_per_second": 1.804, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 1.024665981500514, | |
| "grad_norm": 0.2045954465866089, | |
| "learning_rate": 2.2021276595744682e-05, | |
| "loss": 0.2294, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.065775950668037, | |
| "grad_norm": 0.22853665053844452, | |
| "learning_rate": 2.156534954407295e-05, | |
| "loss": 0.2142, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.10688591983556, | |
| "grad_norm": 0.26341211795806885, | |
| "learning_rate": 2.1109422492401216e-05, | |
| "loss": 0.2207, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.1479958890030832, | |
| "grad_norm": 0.24737012386322021, | |
| "learning_rate": 2.0653495440729482e-05, | |
| "loss": 0.1934, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.1891058581706064, | |
| "grad_norm": 0.2734522819519043, | |
| "learning_rate": 2.019756838905775e-05, | |
| "loss": 0.1832, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.2302158273381294, | |
| "grad_norm": 0.25597164034843445, | |
| "learning_rate": 1.974164133738602e-05, | |
| "loss": 0.1883, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.2713257965056526, | |
| "grad_norm": 0.3185962438583374, | |
| "learning_rate": 1.928571428571429e-05, | |
| "loss": 0.197, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.3124357656731758, | |
| "grad_norm": 0.30189403891563416, | |
| "learning_rate": 1.8829787234042554e-05, | |
| "loss": 0.1687, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.353545734840699, | |
| "grad_norm": 0.31494829058647156, | |
| "learning_rate": 1.837386018237082e-05, | |
| "loss": 0.1737, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.394655704008222, | |
| "grad_norm": 0.2843112349510193, | |
| "learning_rate": 1.7917933130699088e-05, | |
| "loss": 0.1696, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.435765673175745, | |
| "grad_norm": 0.3340540826320648, | |
| "learning_rate": 1.7462006079027357e-05, | |
| "loss": 0.1491, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.4768756423432683, | |
| "grad_norm": 0.3692454695701599, | |
| "learning_rate": 1.7006079027355622e-05, | |
| "loss": 0.1499, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.5179856115107913, | |
| "grad_norm": 0.3271966874599457, | |
| "learning_rate": 1.655015197568389e-05, | |
| "loss": 0.1398, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.5590955806783144, | |
| "grad_norm": 0.35171371698379517, | |
| "learning_rate": 1.609422492401216e-05, | |
| "loss": 0.1317, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.6002055498458376, | |
| "grad_norm": 0.4514343738555908, | |
| "learning_rate": 1.5638297872340426e-05, | |
| "loss": 0.1405, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.6413155190133608, | |
| "grad_norm": 0.35678255558013916, | |
| "learning_rate": 1.5182370820668691e-05, | |
| "loss": 0.134, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.682425488180884, | |
| "grad_norm": 0.3651377856731415, | |
| "learning_rate": 1.4726443768996962e-05, | |
| "loss": 0.1235, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.723535457348407, | |
| "grad_norm": 0.33257487416267395, | |
| "learning_rate": 1.4270516717325229e-05, | |
| "loss": 0.126, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.7646454265159301, | |
| "grad_norm": 0.36159852147102356, | |
| "learning_rate": 1.3814589665653496e-05, | |
| "loss": 0.1044, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.8057553956834531, | |
| "grad_norm": 0.37774747610092163, | |
| "learning_rate": 1.3358662613981763e-05, | |
| "loss": 0.1251, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.8468653648509763, | |
| "grad_norm": 0.3925345540046692, | |
| "learning_rate": 1.2902735562310032e-05, | |
| "loss": 0.1119, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.8879753340184995, | |
| "grad_norm": 0.3282937705516815, | |
| "learning_rate": 1.2446808510638298e-05, | |
| "loss": 0.087, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.9290853031860227, | |
| "grad_norm": 0.416220486164093, | |
| "learning_rate": 1.1990881458966566e-05, | |
| "loss": 0.1061, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.9701952723535459, | |
| "grad_norm": 0.3819692134857178, | |
| "learning_rate": 1.1534954407294832e-05, | |
| "loss": 0.0961, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.35121023654937744, | |
| "eval_runtime": 185.0611, | |
| "eval_samples_per_second": 3.523, | |
| "eval_steps_per_second": 1.762, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 2.0082219938335046, | |
| "grad_norm": 0.3963669240474701, | |
| "learning_rate": 1.10790273556231e-05, | |
| "loss": 0.0892, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.049331963001028, | |
| "grad_norm": 0.3688429892063141, | |
| "learning_rate": 1.0623100303951368e-05, | |
| "loss": 0.0899, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.090441932168551, | |
| "grad_norm": 0.43983232975006104, | |
| "learning_rate": 1.0167173252279635e-05, | |
| "loss": 0.0683, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.131551901336074, | |
| "grad_norm": 0.34000054001808167, | |
| "learning_rate": 9.711246200607902e-06, | |
| "loss": 0.0781, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.172661870503597, | |
| "grad_norm": 0.42162272334098816, | |
| "learning_rate": 9.255319148936171e-06, | |
| "loss": 0.0825, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.21377183967112, | |
| "grad_norm": 0.30206069350242615, | |
| "learning_rate": 8.799392097264438e-06, | |
| "loss": 0.0718, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.2548818088386433, | |
| "grad_norm": 0.3496485650539398, | |
| "learning_rate": 8.343465045592705e-06, | |
| "loss": 0.0849, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.2959917780061665, | |
| "grad_norm": 0.3885113000869751, | |
| "learning_rate": 7.887537993920974e-06, | |
| "loss": 0.0604, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.3371017471736897, | |
| "grad_norm": 0.4243634343147278, | |
| "learning_rate": 7.43161094224924e-06, | |
| "loss": 0.0669, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.378211716341213, | |
| "grad_norm": 0.3809903562068939, | |
| "learning_rate": 6.975683890577508e-06, | |
| "loss": 0.0637, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.419321685508736, | |
| "grad_norm": 0.42439714074134827, | |
| "learning_rate": 6.519756838905775e-06, | |
| "loss": 0.0707, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.460431654676259, | |
| "grad_norm": 0.3830319344997406, | |
| "learning_rate": 6.063829787234042e-06, | |
| "loss": 0.0604, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.501541623843782, | |
| "grad_norm": 0.3180945813655853, | |
| "learning_rate": 5.607902735562311e-06, | |
| "loss": 0.0639, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.542651593011305, | |
| "grad_norm": 0.35575369000434875, | |
| "learning_rate": 5.151975683890578e-06, | |
| "loss": 0.0538, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.5837615621788284, | |
| "grad_norm": 0.4533900022506714, | |
| "learning_rate": 4.696048632218845e-06, | |
| "loss": 0.0628, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.6248715313463515, | |
| "grad_norm": 0.40408235788345337, | |
| "learning_rate": 4.240121580547112e-06, | |
| "loss": 0.0554, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.6659815005138747, | |
| "grad_norm": 0.3473197817802429, | |
| "learning_rate": 3.7841945288753804e-06, | |
| "loss": 0.0563, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.707091469681398, | |
| "grad_norm": 0.3069269061088562, | |
| "learning_rate": 3.3282674772036475e-06, | |
| "loss": 0.0672, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.7482014388489207, | |
| "grad_norm": 0.300988107919693, | |
| "learning_rate": 2.872340425531915e-06, | |
| "loss": 0.0519, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.789311408016444, | |
| "grad_norm": 0.3877612352371216, | |
| "learning_rate": 2.4164133738601823e-06, | |
| "loss": 0.0606, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.830421377183967, | |
| "grad_norm": 0.35421469807624817, | |
| "learning_rate": 1.96048632218845e-06, | |
| "loss": 0.0647, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.87153134635149, | |
| "grad_norm": 0.37970247864723206, | |
| "learning_rate": 1.5045592705167174e-06, | |
| "loss": 0.054, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.9126413155190134, | |
| "grad_norm": 0.3778038024902344, | |
| "learning_rate": 1.0486322188449848e-06, | |
| "loss": 0.0596, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.9537512846865366, | |
| "grad_norm": 0.5201385617256165, | |
| "learning_rate": 5.927051671732523e-07, | |
| "loss": 0.0509, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.9948612538540598, | |
| "grad_norm": 0.314705491065979, | |
| "learning_rate": 1.3677811550151974e-07, | |
| "loss": 0.0608, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.36909812688827515, | |
| "eval_runtime": 180.72, | |
| "eval_samples_per_second": 3.608, | |
| "eval_steps_per_second": 1.804, | |
| "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": 4.264998691879428e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |