| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 2.0, |
| "eval_steps": 500, |
| "global_step": 2500, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.02, |
| "grad_norm": 1.0315879583358765, |
| "learning_rate": 9.6e-05, |
| "loss": 0.8325, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.04, |
| "grad_norm": 0.20474183559417725, |
| "learning_rate": 0.00019600000000000002, |
| "loss": 0.4556, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.06, |
| "grad_norm": 0.22688248753547668, |
| "learning_rate": 0.000296, |
| "loss": 0.4015, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.08, |
| "grad_norm": 0.19172126054763794, |
| "learning_rate": 0.00039600000000000003, |
| "loss": 0.3645, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.1, |
| "grad_norm": 0.15229463577270508, |
| "learning_rate": 0.000496, |
| "loss": 0.3602, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.12, |
| "grad_norm": 0.20351602137088776, |
| "learning_rate": 0.000596, |
| "loss": 0.3475, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.14, |
| "grad_norm": 0.19812420010566711, |
| "learning_rate": 0.000696, |
| "loss": 0.345, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.16, |
| "grad_norm": 0.18347425758838654, |
| "learning_rate": 0.000796, |
| "loss": 0.3327, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.18, |
| "grad_norm": 0.2595237195491791, |
| "learning_rate": 0.000896, |
| "loss": 0.3302, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.2, |
| "grad_norm": 0.2259773463010788, |
| "learning_rate": 0.000996, |
| "loss": 0.3392, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.22, |
| "grad_norm": 0.2506314516067505, |
| "learning_rate": 0.0009997192908557321, |
| "loss": 0.328, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.24, |
| "grad_norm": 0.22288699448108673, |
| "learning_rate": 0.000998830238119205, |
| "loss": 0.3248, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.26, |
| "grad_norm": 0.29342198371887207, |
| "learning_rate": 0.000997333437576437, |
| "loss": 0.3206, |
| "step": 325 |
| }, |
| { |
| "epoch": 0.28, |
| "grad_norm": 0.23638910055160522, |
| "learning_rate": 0.0009952307128483257, |
| "loss": 0.3119, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.3, |
| "grad_norm": 0.20430760085582733, |
| "learning_rate": 0.0009925246257810518, |
| "loss": 0.3029, |
| "step": 375 |
| }, |
| { |
| "epoch": 0.32, |
| "grad_norm": 0.2517499327659607, |
| "learning_rate": 0.0009892184733248665, |
| "loss": 0.3218, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.34, |
| "grad_norm": 0.2284773737192154, |
| "learning_rate": 0.0009853162835172637, |
| "loss": 0.2969, |
| "step": 425 |
| }, |
| { |
| "epoch": 0.36, |
| "grad_norm": 0.26538360118865967, |
| "learning_rate": 0.0009808228105754376, |
| "loss": 0.3134, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.38, |
| "grad_norm": 0.24021762609481812, |
| "learning_rate": 0.0009757435291040016, |
| "loss": 0.3076, |
| "step": 475 |
| }, |
| { |
| "epoch": 0.4, |
| "grad_norm": 0.27008429169654846, |
| "learning_rate": 0.0009700846274250251, |
| "loss": 0.2964, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.4, |
| "eval_loss": 0.2879575788974762, |
| "eval_runtime": 18.9021, |
| "eval_samples_per_second": 54.121, |
| "eval_steps_per_second": 0.846, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.42, |
| "grad_norm": 0.2501317858695984, |
| "learning_rate": 0.000963853000038517, |
| "loss": 0.2981, |
| "step": 525 |
| }, |
| { |
| "epoch": 0.44, |
| "grad_norm": 0.19082719087600708, |
| "learning_rate": 0.0009570562392225395, |
| "loss": 0.2809, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.46, |
| "grad_norm": 0.24488045275211334, |
| "learning_rate": 0.0009497026257831855, |
| "loss": 0.291, |
| "step": 575 |
| }, |
| { |
| "epoch": 0.48, |
| "grad_norm": 0.21422015130519867, |
| "learning_rate": 0.0009418011189656941, |
| "loss": 0.2851, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.5, |
| "grad_norm": 0.19914306700229645, |
| "learning_rate": 0.0009333613455389882, |
| "loss": 0.2841, |
| "step": 625 |
| }, |
| { |
| "epoch": 0.52, |
| "grad_norm": 0.2367725670337677, |
| "learning_rate": 0.000924393588066941, |
| "loss": 0.2879, |
| "step": 650 |
| }, |
| { |
| "epoch": 0.54, |
| "grad_norm": 0.24288254976272583, |
| "learning_rate": 0.0009149087723806549, |
| "loss": 0.2986, |
| "step": 675 |
| }, |
| { |
| "epoch": 0.56, |
| "grad_norm": 0.22942474484443665, |
| "learning_rate": 0.0009049184542670199, |
| "loss": 0.2842, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.58, |
| "grad_norm": 0.23413586616516113, |
| "learning_rate": 0.0008944348053897671, |
| "loss": 0.2925, |
| "step": 725 |
| }, |
| { |
| "epoch": 0.6, |
| "grad_norm": 0.22139714658260345, |
| "learning_rate": 0.0008834705984601709, |
| "loss": 0.2873, |
| "step": 750 |
| }, |
| { |
| "epoch": 0.62, |
| "grad_norm": 0.2082182914018631, |
| "learning_rate": 0.0008720391916754683, |
| "loss": 0.2712, |
| "step": 775 |
| }, |
| { |
| "epoch": 0.64, |
| "grad_norm": 0.24570266902446747, |
| "learning_rate": 0.0008601545124439535, |
| "loss": 0.2813, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.66, |
| "grad_norm": 0.20145930349826813, |
| "learning_rate": 0.0008478310404165754, |
| "loss": 0.2678, |
| "step": 825 |
| }, |
| { |
| "epoch": 0.68, |
| "grad_norm": 0.1796109676361084, |
| "learning_rate": 0.0008350837898457143, |
| "loss": 0.2732, |
| "step": 850 |
| }, |
| { |
| "epoch": 0.7, |
| "grad_norm": 0.22070881724357605, |
| "learning_rate": 0.0008219282912926269, |
| "loss": 0.2744, |
| "step": 875 |
| }, |
| { |
| "epoch": 0.72, |
| "grad_norm": 0.18907737731933594, |
| "learning_rate": 0.0008083805727058513, |
| "loss": 0.2834, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.74, |
| "grad_norm": 0.1631525754928589, |
| "learning_rate": 0.0007944571398936193, |
| "loss": 0.2615, |
| "step": 925 |
| }, |
| { |
| "epoch": 0.76, |
| "grad_norm": 0.1881309598684311, |
| "learning_rate": 0.0007801749564140723, |
| "loss": 0.2732, |
| "step": 950 |
| }, |
| { |
| "epoch": 0.78, |
| "grad_norm": 0.19089096784591675, |
| "learning_rate": 0.0007655514229077783, |
| "loss": 0.2705, |
| "step": 975 |
| }, |
| { |
| "epoch": 0.8, |
| "grad_norm": 0.17820408940315247, |
| "learning_rate": 0.0007506043558977322, |
| "loss": 0.2716, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_loss": 0.2520262598991394, |
| "eval_runtime": 18.7237, |
| "eval_samples_per_second": 54.637, |
| "eval_steps_per_second": 0.855, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.82, |
| "grad_norm": 0.1798166185617447, |
| "learning_rate": 0.0007353519660826664, |
| "loss": 0.2546, |
| "step": 1025 |
| }, |
| { |
| "epoch": 0.84, |
| "grad_norm": 0.21515925228595734, |
| "learning_rate": 0.00071981283615012, |
| "loss": 0.2647, |
| "step": 1050 |
| }, |
| { |
| "epoch": 0.86, |
| "grad_norm": 0.1794830709695816, |
| "learning_rate": 0.0007040058981362964, |
| "loss": 0.2577, |
| "step": 1075 |
| }, |
| { |
| "epoch": 0.88, |
| "grad_norm": 0.18966132402420044, |
| "learning_rate": 0.0006879504103602934, |
| "loss": 0.2594, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.9, |
| "grad_norm": 0.21163836121559143, |
| "learning_rate": 0.0006716659339608077, |
| "loss": 0.2725, |
| "step": 1125 |
| }, |
| { |
| "epoch": 0.92, |
| "grad_norm": 0.17575646936893463, |
| "learning_rate": 0.0006551723090639006, |
| "loss": 0.2483, |
| "step": 1150 |
| }, |
| { |
| "epoch": 0.94, |
| "grad_norm": 0.17414060235023499, |
| "learning_rate": 0.0006384896306108612, |
| "loss": 0.2555, |
| "step": 1175 |
| }, |
| { |
| "epoch": 0.96, |
| "grad_norm": 0.16540151834487915, |
| "learning_rate": 0.0006216382238756146, |
| "loss": 0.2427, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.98, |
| "grad_norm": 0.16461016237735748, |
| "learning_rate": 0.0006046386197015076, |
| "loss": 0.251, |
| "step": 1225 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 0.16811122000217438, |
| "learning_rate": 0.0005875115294876381, |
| "loss": 0.2451, |
| "step": 1250 |
| }, |
| { |
| "epoch": 1.02, |
| "grad_norm": 0.17021119594573975, |
| "learning_rate": 0.0005702778199552054, |
| "loss": 0.2227, |
| "step": 1275 |
| }, |
| { |
| "epoch": 1.04, |
| "grad_norm": 0.16884687542915344, |
| "learning_rate": 0.000552958487724626, |
| "loss": 0.2164, |
| "step": 1300 |
| }, |
| { |
| "epoch": 1.06, |
| "grad_norm": 0.17195427417755127, |
| "learning_rate": 0.0005355746337343836, |
| "loss": 0.2406, |
| "step": 1325 |
| }, |
| { |
| "epoch": 1.08, |
| "grad_norm": 0.15978871285915375, |
| "learning_rate": 0.0005181474375327879, |
| "loss": 0.2138, |
| "step": 1350 |
| }, |
| { |
| "epoch": 1.1, |
| "grad_norm": 0.1938290148973465, |
| "learning_rate": 0.0005006981314739573, |
| "loss": 0.2322, |
| "step": 1375 |
| }, |
| { |
| "epoch": 1.12, |
| "grad_norm": 0.1653754562139511, |
| "learning_rate": 0.00048324797484946424, |
| "loss": 0.2285, |
| "step": 1400 |
| }, |
| { |
| "epoch": 1.1400000000000001, |
| "grad_norm": 0.16597716510295868, |
| "learning_rate": 0.0004658182279871657, |
| "loss": 0.2195, |
| "step": 1425 |
| }, |
| { |
| "epoch": 1.16, |
| "grad_norm": 0.1771618276834488, |
| "learning_rate": 0.00044843012634876645, |
| "loss": 0.2311, |
| "step": 1450 |
| }, |
| { |
| "epoch": 1.18, |
| "grad_norm": 0.1888245791196823, |
| "learning_rate": 0.000431104854657681, |
| "loss": 0.2224, |
| "step": 1475 |
| }, |
| { |
| "epoch": 1.2, |
| "grad_norm": 0.19004148244857788, |
| "learning_rate": 0.0004138635210887117, |
| "loss": 0.2261, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.2, |
| "eval_loss": 0.23447048664093018, |
| "eval_runtime": 18.7321, |
| "eval_samples_per_second": 54.612, |
| "eval_steps_per_second": 0.854, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.22, |
| "grad_norm": 0.15263979136943817, |
| "learning_rate": 0.0003967271315509884, |
| "loss": 0.2227, |
| "step": 1525 |
| }, |
| { |
| "epoch": 1.24, |
| "grad_norm": 0.16230420768260956, |
| "learning_rate": 0.0003797165640955041, |
| "loss": 0.218, |
| "step": 1550 |
| }, |
| { |
| "epoch": 1.26, |
| "grad_norm": 0.1780393123626709, |
| "learning_rate": 0.0003628525434784268, |
| "loss": 0.2228, |
| "step": 1575 |
| }, |
| { |
| "epoch": 1.28, |
| "grad_norm": 0.1682538390159607, |
| "learning_rate": 0.0003461556159111748, |
| "loss": 0.2274, |
| "step": 1600 |
| }, |
| { |
| "epoch": 1.3, |
| "grad_norm": 0.17813384532928467, |
| "learning_rate": 0.0003296461240280242, |
| "loss": 0.2095, |
| "step": 1625 |
| }, |
| { |
| "epoch": 1.32, |
| "grad_norm": 0.17569252848625183, |
| "learning_rate": 0.00031334418210174266, |
| "loss": 0.2325, |
| "step": 1650 |
| }, |
| { |
| "epoch": 1.34, |
| "grad_norm": 0.14148516952991486, |
| "learning_rate": 0.0002972696515374455, |
| "loss": 0.2152, |
| "step": 1675 |
| }, |
| { |
| "epoch": 1.3599999999999999, |
| "grad_norm": 0.14034417271614075, |
| "learning_rate": 0.00028144211667453366, |
| "loss": 0.2189, |
| "step": 1700 |
| }, |
| { |
| "epoch": 1.38, |
| "grad_norm": 0.16050834953784943, |
| "learning_rate": 0.00026588086092619277, |
| "loss": 0.2252, |
| "step": 1725 |
| }, |
| { |
| "epoch": 1.4, |
| "grad_norm": 0.1896321177482605, |
| "learning_rate": 0.00025060484328552466, |
| "loss": 0.2219, |
| "step": 1750 |
| }, |
| { |
| "epoch": 1.42, |
| "grad_norm": 0.16782040894031525, |
| "learning_rate": 0.00023563267522693415, |
| "loss": 0.2139, |
| "step": 1775 |
| }, |
| { |
| "epoch": 1.44, |
| "grad_norm": 0.1926344931125641, |
| "learning_rate": 0.0002209825980309151, |
| "loss": 0.2211, |
| "step": 1800 |
| }, |
| { |
| "epoch": 1.46, |
| "grad_norm": 0.16422989964485168, |
| "learning_rate": 0.00020667246055985938, |
| "loss": 0.2215, |
| "step": 1825 |
| }, |
| { |
| "epoch": 1.48, |
| "grad_norm": 0.16321636736392975, |
| "learning_rate": 0.00019271969751196778, |
| "loss": 0.2066, |
| "step": 1850 |
| }, |
| { |
| "epoch": 1.5, |
| "grad_norm": 0.15386155247688293, |
| "learning_rate": 0.00017914130817975592, |
| "loss": 0.2179, |
| "step": 1875 |
| }, |
| { |
| "epoch": 1.52, |
| "grad_norm": 0.1478998363018036, |
| "learning_rate": 0.00016595383573903412, |
| "loss": 0.2117, |
| "step": 1900 |
| }, |
| { |
| "epoch": 1.54, |
| "grad_norm": 0.16174481809139252, |
| "learning_rate": 0.0001531733470935976, |
| "loss": 0.2149, |
| "step": 1925 |
| }, |
| { |
| "epoch": 1.56, |
| "grad_norm": 0.13468210399150848, |
| "learning_rate": 0.00014081541330017704, |
| "loss": 0.2037, |
| "step": 1950 |
| }, |
| { |
| "epoch": 1.58, |
| "grad_norm": 0.1549725979566574, |
| "learning_rate": 0.00012889509059750602, |
| "loss": 0.2096, |
| "step": 1975 |
| }, |
| { |
| "epoch": 1.6, |
| "grad_norm": 0.15235114097595215, |
| "learning_rate": 0.00011742690206261292, |
| "loss": 0.215, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_loss": 0.22170303761959076, |
| "eval_runtime": 18.729, |
| "eval_samples_per_second": 54.621, |
| "eval_steps_per_second": 0.854, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.62, |
| "grad_norm": 0.16818706691265106, |
| "learning_rate": 0.0001064248199166884, |
| "loss": 0.2024, |
| "step": 2025 |
| }, |
| { |
| "epoch": 1.6400000000000001, |
| "grad_norm": 0.1615205854177475, |
| "learning_rate": 9.590224850208645e-05, |
| "loss": 0.2116, |
| "step": 2050 |
| }, |
| { |
| "epoch": 1.6600000000000001, |
| "grad_norm": 0.15136006474494934, |
| "learning_rate": 8.587200795119792e-05, |
| "loss": 0.2159, |
| "step": 2075 |
| }, |
| { |
| "epoch": 1.6800000000000002, |
| "grad_norm": 0.17629875242710114, |
| "learning_rate": 7.634631856709389e-05, |
| "loss": 0.2173, |
| "step": 2100 |
| }, |
| { |
| "epoch": 1.7, |
| "grad_norm": 0.17757442593574524, |
| "learning_rate": 6.733678593496901e-05, |
| "loss": 0.2165, |
| "step": 2125 |
| }, |
| { |
| "epoch": 1.72, |
| "grad_norm": 0.17096401751041412, |
| "learning_rate": 5.885438678252342e-05, |
| "loss": 0.2104, |
| "step": 2150 |
| }, |
| { |
| "epoch": 1.74, |
| "grad_norm": 0.1552375704050064, |
| "learning_rate": 5.0909455606510726e-05, |
| "loss": 0.2065, |
| "step": 2175 |
| }, |
| { |
| "epoch": 1.76, |
| "grad_norm": 0.2247905433177948, |
| "learning_rate": 4.3511672081746386e-05, |
| "loss": 0.211, |
| "step": 2200 |
| }, |
| { |
| "epoch": 1.78, |
| "grad_norm": 0.15804603695869446, |
| "learning_rate": 3.667004926791395e-05, |
| "loss": 0.2135, |
| "step": 2225 |
| }, |
| { |
| "epoch": 1.8, |
| "grad_norm": 0.15289010107517242, |
| "learning_rate": 3.0392922628540875e-05, |
| "loss": 0.2079, |
| "step": 2250 |
| }, |
| { |
| "epoch": 1.8199999999999998, |
| "grad_norm": 0.148426815867424, |
| "learning_rate": 2.468793987551998e-05, |
| "loss": 0.2048, |
| "step": 2275 |
| }, |
| { |
| "epoch": 1.8399999999999999, |
| "grad_norm": 0.1801871359348297, |
| "learning_rate": 1.9562051651550784e-05, |
| "loss": 0.1998, |
| "step": 2300 |
| }, |
| { |
| "epoch": 1.8599999999999999, |
| "grad_norm": 0.15381543338298798, |
| "learning_rate": 1.5021503061851349e-05, |
| "loss": 0.2127, |
| "step": 2325 |
| }, |
| { |
| "epoch": 1.88, |
| "grad_norm": 0.1602175533771515, |
| "learning_rate": 1.1071826065460589e-05, |
| "loss": 0.1991, |
| "step": 2350 |
| }, |
| { |
| "epoch": 1.9, |
| "grad_norm": 0.17019180953502655, |
| "learning_rate": 7.717832735397334e-06, |
| "loss": 0.2036, |
| "step": 2375 |
| }, |
| { |
| "epoch": 1.92, |
| "grad_norm": 0.19569824635982513, |
| "learning_rate": 4.963609395891299e-06, |
| "loss": 0.2034, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.94, |
| "grad_norm": 0.15439940989017487, |
| "learning_rate": 2.81251164382601e-06, |
| "loss": 0.2063, |
| "step": 2425 |
| }, |
| { |
| "epoch": 1.96, |
| "grad_norm": 0.1678876131772995, |
| "learning_rate": 1.267160260461253e-06, |
| "loss": 0.1939, |
| "step": 2450 |
| }, |
| { |
| "epoch": 1.98, |
| "grad_norm": 0.14893606305122375, |
| "learning_rate": 3.2943801841439634e-07, |
| "loss": 0.203, |
| "step": 2475 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.13442523777484894, |
| "learning_rate": 4.873877924582715e-10, |
| "loss": 0.2036, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.21816210448741913, |
| "eval_runtime": 18.7265, |
| "eval_samples_per_second": 54.629, |
| "eval_steps_per_second": 0.854, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.0, |
| "step": 2500, |
| "total_flos": 1.62588235137024e+18, |
| "train_loss": 0.26149349727630616, |
| "train_runtime": 2422.0842, |
| "train_samples_per_second": 33.029, |
| "train_steps_per_second": 1.032 |
| } |
| ], |
| "logging_steps": 25, |
| "max_steps": 2500, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 0, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.62588235137024e+18, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
