| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.907120743034056, | |
| "eval_steps": 500, | |
| "global_step": 800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1238390092879257, | |
| "grad_norm": 7.5771355628967285, | |
| "learning_rate": 0.0001999229036240723, | |
| "loss": 1.3532, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2476780185758514, | |
| "grad_norm": 0.3571035861968994, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 1.0315, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3715170278637771, | |
| "grad_norm": 0.3278755843639374, | |
| "learning_rate": 0.00019930684569549264, | |
| "loss": 0.9018, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4953560371517028, | |
| "grad_norm": 0.4095502495765686, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 0.807, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6191950464396285, | |
| "grad_norm": 0.4147421717643738, | |
| "learning_rate": 0.00019807852804032305, | |
| "loss": 0.7204, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.7430340557275542, | |
| "grad_norm": 0.2525322735309601, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 0.6548, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8668730650154799, | |
| "grad_norm": 0.28104496002197266, | |
| "learning_rate": 0.00019624552364536473, | |
| "loss": 0.6687, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.9907120743034056, | |
| "grad_norm": 0.4467855393886566, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.6317, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.1145510835913313, | |
| "grad_norm": 0.2571893036365509, | |
| "learning_rate": 0.00019381913359224842, | |
| "loss": 0.6319, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.238390092879257, | |
| "grad_norm": 0.24546292424201965, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.6262, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.3622291021671826, | |
| "grad_norm": 0.24089373648166656, | |
| "learning_rate": 0.00019081431738250814, | |
| "loss": 0.6309, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.4860681114551084, | |
| "grad_norm": 0.24842403829097748, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 0.6142, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.609907120743034, | |
| "grad_norm": 0.2339727133512497, | |
| "learning_rate": 0.00018724960070727972, | |
| "loss": 0.6131, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.7337461300309598, | |
| "grad_norm": 0.21254688501358032, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 0.5892, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.8575851393188856, | |
| "grad_norm": 0.34352943301200867, | |
| "learning_rate": 0.00018314696123025454, | |
| "loss": 0.6037, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.9814241486068112, | |
| "grad_norm": 0.21427258849143982, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.6051, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.1052631578947367, | |
| "grad_norm": 0.23226872086524963, | |
| "learning_rate": 0.00017853169308807448, | |
| "loss": 0.6109, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.2291021671826625, | |
| "grad_norm": 0.254842072725296, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 0.5912, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 0.2571081519126892, | |
| "learning_rate": 0.00017343225094356855, | |
| "loss": 0.5975, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.476780185758514, | |
| "grad_norm": 0.25343191623687744, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.5786, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.6006191950464395, | |
| "grad_norm": 0.21258015930652618, | |
| "learning_rate": 0.0001678800745532942, | |
| "loss": 0.586, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.7244582043343653, | |
| "grad_norm": 0.25848379731178284, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 0.5714, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.848297213622291, | |
| "grad_norm": 0.26716166734695435, | |
| "learning_rate": 0.00016190939493098344, | |
| "loss": 0.5887, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.972136222910217, | |
| "grad_norm": 0.23578402400016785, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.5902, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.0959752321981426, | |
| "grad_norm": 0.23565009236335754, | |
| "learning_rate": 0.00015555702330196023, | |
| "loss": 0.5792, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.219814241486068, | |
| "grad_norm": 0.2390134632587433, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 0.5676, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.343653250773994, | |
| "grad_norm": 0.25006794929504395, | |
| "learning_rate": 0.00014886212414969553, | |
| "loss": 0.5788, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.4674922600619196, | |
| "grad_norm": 0.2533760666847229, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.5769, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.5913312693498454, | |
| "grad_norm": 0.2808171510696411, | |
| "learning_rate": 0.0001418659737537428, | |
| "loss": 0.5521, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.715170278637771, | |
| "grad_norm": 0.28783777356147766, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.5723, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.8390092879256965, | |
| "grad_norm": 0.29237958788871765, | |
| "learning_rate": 0.0001346117057077493, | |
| "loss": 0.5668, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.9628482972136223, | |
| "grad_norm": 0.2757062315940857, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.579, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 4.086687306501548, | |
| "grad_norm": 0.28595200181007385, | |
| "learning_rate": 0.00012714404498650743, | |
| "loss": 0.5488, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.2105263157894735, | |
| "grad_norm": 0.2707183063030243, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 0.5511, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.3343653250774, | |
| "grad_norm": 0.3559975028038025, | |
| "learning_rate": 0.00011950903220161285, | |
| "loss": 0.5683, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.458204334365325, | |
| "grad_norm": 0.2762058973312378, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 0.5499, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.58204334365325, | |
| "grad_norm": 0.2717606723308563, | |
| "learning_rate": 0.00011175373974578378, | |
| "loss": 0.5612, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 0.27757707238197327, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 0.5602, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.829721362229102, | |
| "grad_norm": 0.3977556526660919, | |
| "learning_rate": 0.00010392598157590688, | |
| "loss": 0.5459, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.953560371517028, | |
| "grad_norm": 0.26867300271987915, | |
| "learning_rate": 0.0001, | |
| "loss": 0.537, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.077399380804954, | |
| "grad_norm": 0.26843276619911194, | |
| "learning_rate": 9.607401842409317e-05, | |
| "loss": 0.5601, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 5.201238390092879, | |
| "grad_norm": 0.30268290638923645, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 0.5317, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 5.325077399380805, | |
| "grad_norm": 0.3163929581642151, | |
| "learning_rate": 8.824626025421626e-05, | |
| "loss": 0.5343, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 5.4489164086687305, | |
| "grad_norm": 0.2883571982383728, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 0.5255, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 5.572755417956657, | |
| "grad_norm": 0.3254496157169342, | |
| "learning_rate": 8.049096779838719e-05, | |
| "loss": 0.5281, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.696594427244582, | |
| "grad_norm": 0.2983749508857727, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 0.5515, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.820433436532507, | |
| "grad_norm": 0.2880017161369324, | |
| "learning_rate": 7.285595501349258e-05, | |
| "loss": 0.5575, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 5.944272445820434, | |
| "grad_norm": 0.43873119354248047, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.5367, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 6.068111455108359, | |
| "grad_norm": 0.33720219135284424, | |
| "learning_rate": 6.538829429225069e-05, | |
| "loss": 0.5509, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 6.191950464396285, | |
| "grad_norm": 0.3185509741306305, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.5176, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.315789473684211, | |
| "grad_norm": 0.3240034878253937, | |
| "learning_rate": 5.8134026246257225e-05, | |
| "loss": 0.5306, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 6.439628482972136, | |
| "grad_norm": 0.33068713545799255, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 0.517, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 6.563467492260062, | |
| "grad_norm": 0.33544909954071045, | |
| "learning_rate": 5.113787585030454e-05, | |
| "loss": 0.5288, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 6.687306501547988, | |
| "grad_norm": 0.3468843400478363, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 0.5222, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 6.811145510835914, | |
| "grad_norm": 0.33482104539871216, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 0.5227, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 6.934984520123839, | |
| "grad_norm": 0.3804668188095093, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.5251, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 7.0588235294117645, | |
| "grad_norm": 0.3186335861682892, | |
| "learning_rate": 3.8090605069016595e-05, | |
| "loss": 0.5264, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 7.182662538699691, | |
| "grad_norm": 0.34853196144104004, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 0.5281, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 7.306501547987616, | |
| "grad_norm": 0.36451995372772217, | |
| "learning_rate": 3.211992544670582e-05, | |
| "loss": 0.5074, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 7.430340557275541, | |
| "grad_norm": 0.3326849639415741, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.5249, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.554179566563468, | |
| "grad_norm": 0.3516250550746918, | |
| "learning_rate": 2.6567749056431467e-05, | |
| "loss": 0.5112, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 7.678018575851393, | |
| "grad_norm": 0.3434501588344574, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 0.5176, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 7.801857585139319, | |
| "grad_norm": 0.3573139011859894, | |
| "learning_rate": 2.146830691192553e-05, | |
| "loss": 0.5251, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 7.925696594427245, | |
| "grad_norm": 0.3552079200744629, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.5066, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 8.04953560371517, | |
| "grad_norm": 0.3211444020271301, | |
| "learning_rate": 1.6853038769745467e-05, | |
| "loss": 0.5255, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 8.173374613003096, | |
| "grad_norm": 0.3437272310256958, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 0.5152, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 8.297213622291022, | |
| "grad_norm": 0.39420753717422485, | |
| "learning_rate": 1.2750399292720283e-05, | |
| "loss": 0.5147, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 8.421052631578947, | |
| "grad_norm": 0.33711323142051697, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 0.5053, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 8.544891640866872, | |
| "grad_norm": 0.41042107343673706, | |
| "learning_rate": 9.185682617491863e-06, | |
| "loss": 0.4921, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 8.6687306501548, | |
| "grad_norm": 0.3305673897266388, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.5182, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 8.792569659442725, | |
| "grad_norm": 0.42124322056770325, | |
| "learning_rate": 6.180866407751595e-06, | |
| "loss": 0.5229, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 8.91640866873065, | |
| "grad_norm": 0.3476842939853668, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.517, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 9.040247678018575, | |
| "grad_norm": 0.3766673803329468, | |
| "learning_rate": 3.7544763546352834e-06, | |
| "loss": 0.4968, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 9.1640866873065, | |
| "grad_norm": 0.34576883912086487, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 0.5143, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 9.287925696594428, | |
| "grad_norm": 0.3424369692802429, | |
| "learning_rate": 1.921471959676957e-06, | |
| "loss": 0.5028, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 9.411764705882353, | |
| "grad_norm": 0.3952752947807312, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 0.4951, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 9.535603715170279, | |
| "grad_norm": 0.35501015186309814, | |
| "learning_rate": 6.931543045073708e-07, | |
| "loss": 0.5218, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 9.659442724458204, | |
| "grad_norm": 0.3750287592411041, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 0.5143, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 9.78328173374613, | |
| "grad_norm": 0.336580365896225, | |
| "learning_rate": 7.709637592770991e-08, | |
| "loss": 0.5106, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 9.907120743034056, | |
| "grad_norm": 0.35525137186050415, | |
| "learning_rate": 0.0, | |
| "loss": 0.514, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 9.907120743034056, | |
| "step": 800, | |
| "total_flos": 3.90379106992128e+16, | |
| "train_loss": 0.5782659471035003, | |
| "train_runtime": 2175.2449, | |
| "train_samples_per_second": 1.485, | |
| "train_steps_per_second": 0.368 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 3.90379106992128e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |