{
  "best_global_step": 383,
  "best_metric": 0.3293873369693756,
  "best_model_checkpoint": "./lora_qwen32b_cpp_abdiff_v1/checkpoint-383",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1149,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02611818478615736,
      "grad_norm": 0.6525485515594482,
      "learning_rate": 2.347826086956522e-06,
      "loss": 1.5822,
      "step": 10
    },
    {
      "epoch": 0.05223636957231472,
      "grad_norm": 0.832341730594635,
      "learning_rate": 4.956521739130435e-06,
      "loss": 1.5516,
      "step": 20
    },
    {
      "epoch": 0.07835455435847209,
      "grad_norm": 0.9468288421630859,
      "learning_rate": 7.5652173913043475e-06,
      "loss": 1.5544,
      "step": 30
    },
    {
      "epoch": 0.10447273914462944,
      "grad_norm": 0.6425743103027344,
      "learning_rate": 1.017391304347826e-05,
      "loss": 1.3415,
      "step": 40
    },
    {
      "epoch": 0.1305909239307868,
      "grad_norm": 0.6995680332183838,
      "learning_rate": 1.2782608695652173e-05,
      "loss": 1.2298,
      "step": 50
    },
    {
      "epoch": 0.15670910871694418,
      "grad_norm": 0.559788703918457,
      "learning_rate": 1.5391304347826088e-05,
      "loss": 1.0121,
      "step": 60
    },
    {
      "epoch": 0.18282729350310153,
      "grad_norm": 0.565027117729187,
      "learning_rate": 1.8e-05,
      "loss": 0.7633,
      "step": 70
    },
    {
      "epoch": 0.20894547828925888,
      "grad_norm": 0.2943669557571411,
      "learning_rate": 2.0608695652173913e-05,
      "loss": 0.6203,
      "step": 80
    },
    {
      "epoch": 0.23506366307541626,
      "grad_norm": 0.2539921998977661,
      "learning_rate": 2.3217391304347826e-05,
      "loss": 0.5356,
      "step": 90
    },
    {
      "epoch": 0.2611818478615736,
      "grad_norm": 0.32755884528160095,
      "learning_rate": 2.582608695652174e-05,
      "loss": 0.5063,
      "step": 100
    },
    {
      "epoch": 0.28730003264773096,
      "grad_norm": 0.2826889753341675,
      "learning_rate": 2.8434782608695652e-05,
      "loss": 0.4491,
      "step": 110
    },
    {
      "epoch": 0.31341821743388837,
      "grad_norm": 0.26905569434165955,
      "learning_rate": 2.988394584139265e-05,
      "loss": 0.3921,
      "step": 120
    },
    {
      "epoch": 0.3395364022200457,
      "grad_norm": 0.23731912672519684,
      "learning_rate": 2.9593810444874276e-05,
      "loss": 0.3865,
      "step": 130
    },
    {
      "epoch": 0.36565458700620307,
      "grad_norm": 0.34480321407318115,
      "learning_rate": 2.93036750483559e-05,
      "loss": 0.37,
      "step": 140
    },
    {
      "epoch": 0.3917727717923604,
      "grad_norm": 0.33365756273269653,
      "learning_rate": 2.9013539651837528e-05,
      "loss": 0.3918,
      "step": 150
    },
    {
      "epoch": 0.41789095657851777,
      "grad_norm": 0.24923010170459747,
      "learning_rate": 2.872340425531915e-05,
      "loss": 0.3652,
      "step": 160
    },
    {
      "epoch": 0.4440091413646752,
      "grad_norm": 0.2503843605518341,
      "learning_rate": 2.8433268858800773e-05,
      "loss": 0.3506,
      "step": 170
    },
    {
      "epoch": 0.4701273261508325,
      "grad_norm": 0.2814203202724457,
      "learning_rate": 2.81431334622824e-05,
      "loss": 0.3173,
      "step": 180
    },
    {
      "epoch": 0.4962455109369899,
      "grad_norm": 0.22126232087612152,
      "learning_rate": 2.785299806576402e-05,
      "loss": 0.3291,
      "step": 190
    },
    {
      "epoch": 0.5223636957231472,
      "grad_norm": 0.3136242926120758,
      "learning_rate": 2.7562862669245647e-05,
      "loss": 0.3456,
      "step": 200
    },
    {
      "epoch": 0.5484818805093046,
      "grad_norm": 0.26128554344177246,
      "learning_rate": 2.7272727272727273e-05,
      "loss": 0.2983,
      "step": 210
    },
    {
      "epoch": 0.5746000652954619,
      "grad_norm": 0.21475030481815338,
      "learning_rate": 2.69825918762089e-05,
      "loss": 0.3252,
      "step": 220
    },
    {
      "epoch": 0.6007182500816193,
      "grad_norm": 0.20808374881744385,
      "learning_rate": 2.669245647969052e-05,
      "loss": 0.3057,
      "step": 230
    },
    {
      "epoch": 0.6268364348677767,
      "grad_norm": 0.3940744698047638,
      "learning_rate": 2.6402321083172148e-05,
      "loss": 0.3157,
      "step": 240
    },
    {
      "epoch": 0.6529546196539341,
      "grad_norm": 0.28545427322387695,
      "learning_rate": 2.6112185686653773e-05,
      "loss": 0.3109,
      "step": 250
    },
    {
      "epoch": 0.6790728044400914,
      "grad_norm": 0.2335038036108017,
      "learning_rate": 2.5822050290135396e-05,
      "loss": 0.3023,
      "step": 260
    },
    {
      "epoch": 0.7051909892262488,
      "grad_norm": 0.23341821134090424,
      "learning_rate": 2.5531914893617022e-05,
      "loss": 0.2742,
      "step": 270
    },
    {
      "epoch": 0.7313091740124061,
      "grad_norm": 0.2653687000274658,
      "learning_rate": 2.5241779497098648e-05,
      "loss": 0.2947,
      "step": 280
    },
    {
      "epoch": 0.7574273587985635,
      "grad_norm": 0.21947993338108063,
      "learning_rate": 2.495164410058027e-05,
      "loss": 0.2753,
      "step": 290
    },
    {
      "epoch": 0.7835455435847208,
      "grad_norm": 0.2334894835948944,
      "learning_rate": 2.4661508704061896e-05,
      "loss": 0.2558,
      "step": 300
    },
    {
      "epoch": 0.8096637283708782,
      "grad_norm": 0.2747754752635956,
      "learning_rate": 2.4371373307543522e-05,
      "loss": 0.2852,
      "step": 310
    },
    {
      "epoch": 0.8357819131570355,
      "grad_norm": 0.2383272498846054,
      "learning_rate": 2.408123791102515e-05,
      "loss": 0.3035,
      "step": 320
    },
    {
      "epoch": 0.861900097943193,
      "grad_norm": 0.2756940722465515,
      "learning_rate": 2.379110251450677e-05,
      "loss": 0.2706,
      "step": 330
    },
    {
      "epoch": 0.8880182827293504,
      "grad_norm": 0.4882807731628418,
      "learning_rate": 2.3500967117988397e-05,
      "loss": 0.2538,
      "step": 340
    },
    {
      "epoch": 0.9141364675155077,
      "grad_norm": 0.2678774893283844,
      "learning_rate": 2.321083172147002e-05,
      "loss": 0.2715,
      "step": 350
    },
    {
      "epoch": 0.940254652301665,
      "grad_norm": 0.2795256972312927,
      "learning_rate": 2.2920696324951642e-05,
      "loss": 0.2563,
      "step": 360
    },
    {
      "epoch": 0.9663728370878224,
      "grad_norm": 0.29648348689079285,
      "learning_rate": 2.2630560928433268e-05,
      "loss": 0.2649,
      "step": 370
    },
    {
      "epoch": 0.9924910218739798,
      "grad_norm": 0.2755793333053589,
      "learning_rate": 2.2340425531914894e-05,
      "loss": 0.2495,
      "step": 380
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.3293873369693756,
      "eval_runtime": 167.8172,
      "eval_samples_per_second": 5.47,
      "eval_steps_per_second": 2.735,
      "step": 383
    },
    {
      "epoch": 1.01828272935031,
      "grad_norm": 0.3155263364315033,
      "learning_rate": 2.2050290135396516e-05,
      "loss": 0.2493,
      "step": 390
    },
    {
      "epoch": 1.0444009141364674,
      "grad_norm": 0.34601202607154846,
      "learning_rate": 2.1760154738878142e-05,
      "loss": 0.222,
      "step": 400
    },
    {
      "epoch": 1.070519098922625,
      "grad_norm": 0.2998008728027344,
      "learning_rate": 2.1470019342359768e-05,
      "loss": 0.2354,
      "step": 410
    },
    {
      "epoch": 1.0966372837087823,
      "grad_norm": 0.34074831008911133,
      "learning_rate": 2.1179883945841394e-05,
      "loss": 0.2235,
      "step": 420
    },
    {
      "epoch": 1.1227554684949397,
      "grad_norm": 0.2818804681301117,
      "learning_rate": 2.0889748549323017e-05,
      "loss": 0.2417,
      "step": 430
    },
    {
      "epoch": 1.148873653281097,
      "grad_norm": 0.30673709511756897,
      "learning_rate": 2.0599613152804643e-05,
      "loss": 0.2261,
      "step": 440
    },
    {
      "epoch": 1.1749918380672544,
      "grad_norm": 0.24605165421962738,
      "learning_rate": 2.030947775628627e-05,
      "loss": 0.2326,
      "step": 450
    },
    {
      "epoch": 1.2011100228534117,
      "grad_norm": 0.35388073325157166,
      "learning_rate": 2.001934235976789e-05,
      "loss": 0.2239,
      "step": 460
    },
    {
      "epoch": 1.227228207639569,
      "grad_norm": 0.3778790533542633,
      "learning_rate": 1.9729206963249517e-05,
      "loss": 0.2098,
      "step": 470
    },
    {
      "epoch": 1.2533463924257264,
      "grad_norm": 0.3274332880973816,
      "learning_rate": 1.9439071566731143e-05,
      "loss": 0.2024,
      "step": 480
    },
    {
      "epoch": 1.2794645772118838,
      "grad_norm": 0.2900117337703705,
      "learning_rate": 1.914893617021277e-05,
      "loss": 0.1877,
      "step": 490
    },
    {
      "epoch": 1.3055827619980411,
      "grad_norm": 0.3676786720752716,
      "learning_rate": 1.885880077369439e-05,
      "loss": 0.2242,
      "step": 500
    },
    {
      "epoch": 1.3317009467841985,
      "grad_norm": 0.29458677768707275,
      "learning_rate": 1.8568665377176018e-05,
      "loss": 0.2063,
      "step": 510
    },
    {
      "epoch": 1.3578191315703558,
      "grad_norm": 0.3401682376861572,
      "learning_rate": 1.8278529980657643e-05,
      "loss": 0.1911,
      "step": 520
    },
    {
      "epoch": 1.3839373163565132,
      "grad_norm": 0.3856654167175293,
      "learning_rate": 1.7988394584139263e-05,
      "loss": 0.1922,
      "step": 530
    },
    {
      "epoch": 1.4100555011426705,
      "grad_norm": 0.34166622161865234,
      "learning_rate": 1.769825918762089e-05,
      "loss": 0.2023,
      "step": 540
    },
    {
      "epoch": 1.4361736859288279,
      "grad_norm": 0.29851317405700684,
      "learning_rate": 1.7408123791102515e-05,
      "loss": 0.2233,
      "step": 550
    },
    {
      "epoch": 1.4622918707149852,
      "grad_norm": 0.3802024722099304,
      "learning_rate": 1.7117988394584137e-05,
      "loss": 0.1796,
      "step": 560
    },
    {
      "epoch": 1.4884100555011428,
      "grad_norm": 0.40166980028152466,
      "learning_rate": 1.6827852998065763e-05,
      "loss": 0.1746,
      "step": 570
    },
    {
      "epoch": 1.5145282402873002,
      "grad_norm": 0.3295914828777313,
      "learning_rate": 1.653771760154739e-05,
      "loss": 0.1692,
      "step": 580
    },
    {
      "epoch": 1.5406464250734575,
      "grad_norm": 0.3520311117172241,
      "learning_rate": 1.6247582205029015e-05,
      "loss": 0.1608,
      "step": 590
    },
    {
      "epoch": 1.5667646098596149,
      "grad_norm": 0.3754492402076721,
      "learning_rate": 1.5957446808510637e-05,
      "loss": 0.1615,
      "step": 600
    },
    {
      "epoch": 1.5928827946457722,
      "grad_norm": 0.34858810901641846,
      "learning_rate": 1.5667311411992263e-05,
      "loss": 0.1944,
      "step": 610
    },
    {
      "epoch": 1.6190009794319296,
      "grad_norm": 0.3938177525997162,
      "learning_rate": 1.537717601547389e-05,
      "loss": 0.1703,
      "step": 620
    },
    {
      "epoch": 1.645119164218087,
      "grad_norm": 0.32162296772003174,
      "learning_rate": 1.5087040618955514e-05,
      "loss": 0.1943,
      "step": 630
    },
    {
      "epoch": 1.6712373490042443,
      "grad_norm": 0.41002345085144043,
      "learning_rate": 1.4796905222437138e-05,
      "loss": 0.1653,
      "step": 640
    },
    {
      "epoch": 1.6973555337904016,
      "grad_norm": 0.48307564854621887,
      "learning_rate": 1.4506769825918764e-05,
      "loss": 0.1863,
      "step": 650
    },
    {
      "epoch": 1.723473718576559,
      "grad_norm": 0.2567991614341736,
      "learning_rate": 1.4216634429400386e-05,
      "loss": 0.1696,
      "step": 660
    },
    {
      "epoch": 1.7495919033627163,
      "grad_norm": 0.4855695068836212,
      "learning_rate": 1.392649903288201e-05,
      "loss": 0.1676,
      "step": 670
    },
    {
      "epoch": 1.7757100881488737,
      "grad_norm": 0.39257341623306274,
      "learning_rate": 1.3636363636363637e-05,
      "loss": 0.1717,
      "step": 680
    },
    {
      "epoch": 1.801828272935031,
      "grad_norm": 0.4006061553955078,
      "learning_rate": 1.334622823984526e-05,
      "loss": 0.1694,
      "step": 690
    },
    {
      "epoch": 1.8279464577211884,
      "grad_norm": 0.4542832672595978,
      "learning_rate": 1.3056092843326887e-05,
      "loss": 0.1656,
      "step": 700
    },
    {
      "epoch": 1.8540646425073457,
      "grad_norm": 0.3904496431350708,
      "learning_rate": 1.2765957446808511e-05,
      "loss": 0.1695,
      "step": 710
    },
    {
      "epoch": 1.880182827293503,
      "grad_norm": 0.40930867195129395,
      "learning_rate": 1.2475822050290135e-05,
      "loss": 0.1677,
      "step": 720
    },
    {
      "epoch": 1.9063010120796604,
      "grad_norm": 0.3501437306404114,
      "learning_rate": 1.2185686653771761e-05,
      "loss": 0.1607,
      "step": 730
    },
    {
      "epoch": 1.9324191968658178,
      "grad_norm": 0.433753103017807,
      "learning_rate": 1.1895551257253385e-05,
      "loss": 0.1691,
      "step": 740
    },
    {
      "epoch": 1.958537381651975,
      "grad_norm": 0.45074334740638733,
      "learning_rate": 1.160541586073501e-05,
      "loss": 0.1588,
      "step": 750
    },
    {
      "epoch": 1.9846555664381325,
      "grad_norm": 0.3673350214958191,
      "learning_rate": 1.1315280464216634e-05,
      "loss": 0.1426,
      "step": 760
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.3427020311355591,
      "eval_runtime": 168.9488,
      "eval_samples_per_second": 5.434,
      "eval_steps_per_second": 2.717,
      "step": 766
    },
    {
      "epoch": 2.0104472739144628,
      "grad_norm": 0.31791359186172485,
      "learning_rate": 1.1025145067698258e-05,
      "loss": 0.1728,
      "step": 770
    },
    {
      "epoch": 2.03656545870062,
      "grad_norm": 0.4077305495738983,
      "learning_rate": 1.0735009671179884e-05,
      "loss": 0.1528,
      "step": 780
    },
    {
      "epoch": 2.0626836434867775,
      "grad_norm": 0.42285236716270447,
      "learning_rate": 1.0444874274661508e-05,
      "loss": 0.1353,
      "step": 790
    },
    {
      "epoch": 2.088801828272935,
      "grad_norm": 0.48228877782821655,
      "learning_rate": 1.0154738878143134e-05,
      "loss": 0.1316,
      "step": 800
    },
    {
      "epoch": 2.1149200130590926,
      "grad_norm": 0.44379907846450806,
      "learning_rate": 9.864603481624759e-06,
      "loss": 0.1262,
      "step": 810
    },
    {
      "epoch": 2.14103819784525,
      "grad_norm": 0.37220650911331177,
      "learning_rate": 9.574468085106385e-06,
      "loss": 0.1296,
      "step": 820
    },
    {
      "epoch": 2.1671563826314073,
      "grad_norm": 0.4380229711532593,
      "learning_rate": 9.284332688588009e-06,
      "loss": 0.1596,
      "step": 830
    },
    {
      "epoch": 2.1932745674175647,
      "grad_norm": 0.36796846985816956,
      "learning_rate": 8.994197292069631e-06,
      "loss": 0.1347,
      "step": 840
    },
    {
      "epoch": 2.219392752203722,
      "grad_norm": 0.39259183406829834,
      "learning_rate": 8.704061895551257e-06,
      "loss": 0.1536,
      "step": 850
    },
    {
      "epoch": 2.2455109369898794,
      "grad_norm": 0.49359890818595886,
      "learning_rate": 8.413926499032882e-06,
      "loss": 0.1342,
      "step": 860
    },
    {
      "epoch": 2.2716291217760367,
      "grad_norm": 0.3219316899776459,
      "learning_rate": 8.123791102514507e-06,
      "loss": 0.1432,
      "step": 870
    },
    {
      "epoch": 2.297747306562194,
      "grad_norm": 0.3993335962295532,
      "learning_rate": 7.833655705996132e-06,
      "loss": 0.1338,
      "step": 880
    },
    {
      "epoch": 2.3238654913483514,
      "grad_norm": 0.4508567750453949,
      "learning_rate": 7.543520309477757e-06,
      "loss": 0.1518,
      "step": 890
    },
    {
      "epoch": 2.3499836761345088,
      "grad_norm": 0.3847745358943939,
      "learning_rate": 7.253384912959382e-06,
      "loss": 0.1451,
      "step": 900
    },
    {
      "epoch": 2.376101860920666,
      "grad_norm": 0.4204421937465668,
      "learning_rate": 6.963249516441005e-06,
      "loss": 0.1304,
      "step": 910
    },
    {
      "epoch": 2.4022200457068235,
      "grad_norm": 0.4769323766231537,
      "learning_rate": 6.67311411992263e-06,
      "loss": 0.1336,
      "step": 920
    },
    {
      "epoch": 2.428338230492981,
      "grad_norm": 0.42628422379493713,
      "learning_rate": 6.3829787234042555e-06,
      "loss": 0.1401,
      "step": 930
    },
    {
      "epoch": 2.454456415279138,
      "grad_norm": 0.6018841862678528,
      "learning_rate": 6.092843326885881e-06,
      "loss": 0.1257,
      "step": 940
    },
    {
      "epoch": 2.4805746000652955,
      "grad_norm": 0.46257638931274414,
      "learning_rate": 5.802707930367505e-06,
      "loss": 0.1351,
      "step": 950
    },
    {
      "epoch": 2.506692784851453,
      "grad_norm": 0.42667606472969055,
      "learning_rate": 5.512572533849129e-06,
      "loss": 0.1386,
      "step": 960
    },
    {
      "epoch": 2.53281096963761,
      "grad_norm": 0.4261831045150757,
      "learning_rate": 5.222437137330754e-06,
      "loss": 0.1339,
      "step": 970
    },
    {
      "epoch": 2.5589291544237676,
      "grad_norm": 0.41808250546455383,
      "learning_rate": 4.932301740812379e-06,
      "loss": 0.117,
      "step": 980
    },
    {
      "epoch": 2.585047339209925,
      "grad_norm": 0.4173148572444916,
      "learning_rate": 4.642166344294004e-06,
      "loss": 0.1226,
      "step": 990
    },
    {
      "epoch": 2.6111655239960823,
      "grad_norm": 0.5112546682357788,
      "learning_rate": 4.352030947775629e-06,
      "loss": 0.138,
      "step": 1000
    },
    {
      "epoch": 2.6372837087822396,
      "grad_norm": 0.42845726013183594,
      "learning_rate": 4.061895551257254e-06,
      "loss": 0.1236,
      "step": 1010
    },
    {
      "epoch": 2.663401893568397,
      "grad_norm": 0.45304423570632935,
      "learning_rate": 3.7717601547388784e-06,
      "loss": 0.1356,
      "step": 1020
    },
    {
      "epoch": 2.6895200783545543,
      "grad_norm": 0.4409390985965729,
      "learning_rate": 3.4816247582205027e-06,
      "loss": 0.1272,
      "step": 1030
    },
    {
      "epoch": 2.7156382631407117,
      "grad_norm": 0.43914973735809326,
      "learning_rate": 3.1914893617021277e-06,
      "loss": 0.1173,
      "step": 1040
    },
    {
      "epoch": 2.741756447926869,
      "grad_norm": 0.4517093002796173,
      "learning_rate": 2.9013539651837524e-06,
      "loss": 0.1259,
      "step": 1050
    },
    {
      "epoch": 2.7678746327130264,
      "grad_norm": 0.42893633246421814,
      "learning_rate": 2.611218568665377e-06,
      "loss": 0.121,
      "step": 1060
    },
    {
      "epoch": 2.7939928174991837,
      "grad_norm": 0.33101484179496765,
      "learning_rate": 2.321083172147002e-06,
      "loss": 0.1484,
      "step": 1070
    },
    {
      "epoch": 2.820111002285341,
      "grad_norm": 0.46963414549827576,
      "learning_rate": 2.030947775628627e-06,
      "loss": 0.1286,
      "step": 1080
    },
    {
      "epoch": 2.8462291870714984,
      "grad_norm": 0.48841312527656555,
      "learning_rate": 1.7408123791102513e-06,
      "loss": 0.1364,
      "step": 1090
    },
    {
      "epoch": 2.8723473718576558,
      "grad_norm": 0.3498128354549408,
      "learning_rate": 1.4506769825918762e-06,
      "loss": 0.1176,
      "step": 1100
    },
    {
      "epoch": 2.898465556643813,
      "grad_norm": 0.49255019426345825,
      "learning_rate": 1.160541586073501e-06,
      "loss": 0.1283,
      "step": 1110
    },
    {
      "epoch": 2.9245837414299705,
      "grad_norm": 0.45757627487182617,
      "learning_rate": 8.704061895551257e-07,
      "loss": 0.131,
      "step": 1120
    },
    {
      "epoch": 2.950701926216128,
      "grad_norm": 0.4321511387825012,
      "learning_rate": 5.802707930367505e-07,
      "loss": 0.142,
      "step": 1130
    },
    {
      "epoch": 2.9768201110022856,
      "grad_norm": 0.48838484287261963,
      "learning_rate": 2.901353965183753e-07,
      "loss": 0.1229,
      "step": 1140
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.35567396879196167,
      "eval_runtime": 167.2748,
      "eval_samples_per_second": 5.488,
      "eval_steps_per_second": 2.744,
      "step": 1149
    }
  ],
  "logging_steps": 10,
  "max_steps": 1149,
  "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": 5.001355808563446e+18,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}