Chrisyichuan's picture
Upload multi-image 3x LoRA (LLM-judge 0.9)
9b1bdbc verified
Raw History Blame Contribute Delete
40.9 kB
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.0,
"eval_steps": 800,
"global_step": 6502,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.009227929867733005,
"grad_norm": 50.66204833984375,
"learning_rate": 1.479591836734694e-06,
"loss": 2.4591,
"step": 30
},
{
"epoch": 0.01845585973546601,
"grad_norm": 96.23739624023438,
"learning_rate": 3.0102040816326534e-06,
"loss": 0.8107,
"step": 60
},
{
"epoch": 0.027683789603199015,
"grad_norm": 668.5885009765625,
"learning_rate": 4.540816326530613e-06,
"loss": 0.3919,
"step": 90
},
{
"epoch": 0.03691171947093202,
"grad_norm": 9.798393249511719,
"learning_rate": 6.071428571428571e-06,
"loss": 0.3028,
"step": 120
},
{
"epoch": 0.046139649338665024,
"grad_norm": 10.637200355529785,
"learning_rate": 7.602040816326531e-06,
"loss": 0.309,
"step": 150
},
{
"epoch": 0.05536757920639803,
"grad_norm": 247.37608337402344,
"learning_rate": 9.13265306122449e-06,
"loss": 0.2733,
"step": 180
},
{
"epoch": 0.06459550907413103,
"grad_norm": 32.60674285888672,
"learning_rate": 9.999895138272554e-06,
"loss": 0.2555,
"step": 210
},
{
"epoch": 0.07382343894186404,
"grad_norm": 21.95768928527832,
"learning_rate": 9.998852765698074e-06,
"loss": 0.194,
"step": 240
},
{
"epoch": 0.08305136880959704,
"grad_norm": 11.552107810974121,
"learning_rate": 9.99669379579836e-06,
"loss": 0.224,
"step": 270
},
{
"epoch": 0.09227929867733005,
"grad_norm": 16.594078063964844,
"learning_rate": 9.993418710824068e-06,
"loss": 0.1934,
"step": 300
},
{
"epoch": 0.10150722854506306,
"grad_norm": 11.209543228149414,
"learning_rate": 9.989028242333275e-06,
"loss": 0.2084,
"step": 330
},
{
"epoch": 0.11073515841279606,
"grad_norm": 23.406234741210938,
"learning_rate": 9.983523371028076e-06,
"loss": 0.1876,
"step": 360
},
{
"epoch": 0.11996308828052907,
"grad_norm": 9.357805252075195,
"learning_rate": 9.976905326535517e-06,
"loss": 0.191,
"step": 390
},
{
"epoch": 0.12919101814826206,
"grad_norm": 21.04151725769043,
"learning_rate": 9.969175587132942e-06,
"loss": 0.1781,
"step": 420
},
{
"epoch": 0.13841894801599508,
"grad_norm": 13.652193069458008,
"learning_rate": 9.960335879417782e-06,
"loss": 0.1768,
"step": 450
},
{
"epoch": 0.14764687788372807,
"grad_norm": 7.896795272827148,
"learning_rate": 9.950388177921886e-06,
"loss": 0.1792,
"step": 480
},
{
"epoch": 0.1568748077514611,
"grad_norm": 7.785383701324463,
"learning_rate": 9.939334704670464e-06,
"loss": 0.1792,
"step": 510
},
{
"epoch": 0.16610273761919409,
"grad_norm": 10.756692886352539,
"learning_rate": 9.92717792868576e-06,
"loss": 0.1794,
"step": 540
},
{
"epoch": 0.1753306674869271,
"grad_norm": 7.182068347930908,
"learning_rate": 9.913920565435543e-06,
"loss": 0.1719,
"step": 570
},
{
"epoch": 0.1845585973546601,
"grad_norm": 130.00680541992188,
"learning_rate": 9.89956557622654e-06,
"loss": 0.162,
"step": 600
},
{
"epoch": 0.19378652722239312,
"grad_norm": 13.810831069946289,
"learning_rate": 9.884116167542981e-06,
"loss": 0.1667,
"step": 630
},
{
"epoch": 0.2030144570901261,
"grad_norm": 13.16609001159668,
"learning_rate": 9.867575790330356e-06,
"loss": 0.161,
"step": 660
},
{
"epoch": 0.21224238695785913,
"grad_norm": 25.982439041137695,
"learning_rate": 9.849948139224581e-06,
"loss": 0.1831,
"step": 690
},
{
"epoch": 0.22147031682559212,
"grad_norm": 9.271159172058105,
"learning_rate": 9.831237151726716e-06,
"loss": 0.1719,
"step": 720
},
{
"epoch": 0.23069824669332514,
"grad_norm": 4.983822822570801,
"learning_rate": 9.81144700732345e-06,
"loss": 0.16,
"step": 750
},
{
"epoch": 0.23992617656105814,
"grad_norm": 6.652640342712402,
"learning_rate": 9.790582126553526e-06,
"loss": 0.1726,
"step": 780
},
{
"epoch": 0.24607812980621346,
"eval_accuracy": 0.9531888982703618,
"eval_exact_match": 0.7418741355463347,
"eval_loss": 0.16981439292430878,
"eval_runtime": 124.3815,
"eval_samples_per_second": 46.462,
"eval_steps_per_second": 5.813,
"step": 800
},
{
"epoch": 0.24915410642879113,
"grad_norm": 22.226713180541992,
"learning_rate": 9.76864717002031e-06,
"loss": 0.1694,
"step": 810
},
{
"epoch": 0.2583820362965241,
"grad_norm": 15.475312232971191,
"learning_rate": 9.745647037350772e-06,
"loss": 0.1644,
"step": 840
},
{
"epoch": 0.26760996616425714,
"grad_norm": 7.272610187530518,
"learning_rate": 9.721586866101026e-06,
"loss": 0.1636,
"step": 870
},
{
"epoch": 0.27683789603199016,
"grad_norm": 17.45174217224121,
"learning_rate": 9.696472030608775e-06,
"loss": 0.1469,
"step": 900
},
{
"epoch": 0.2860658258997232,
"grad_norm": 22.935935974121094,
"learning_rate": 9.670308140792822e-06,
"loss": 0.1626,
"step": 930
},
{
"epoch": 0.29529375576745615,
"grad_norm": 23.110702514648438,
"learning_rate": 9.643101040899984e-06,
"loss": 0.1485,
"step": 960
},
{
"epoch": 0.30452168563518917,
"grad_norm": 9.469276428222656,
"learning_rate": 9.614856808199671e-06,
"loss": 0.1512,
"step": 990
},
{
"epoch": 0.3137496155029222,
"grad_norm": 14.428640365600586,
"learning_rate": 9.585581751626383e-06,
"loss": 0.155,
"step": 1020
},
{
"epoch": 0.3229775453706552,
"grad_norm": 22.722126007080078,
"learning_rate": 9.555282410370487e-06,
"loss": 0.1777,
"step": 1050
},
{
"epoch": 0.33220547523838817,
"grad_norm": 8.687199592590332,
"learning_rate": 9.523965552417563e-06,
"loss": 0.1639,
"step": 1080
},
{
"epoch": 0.3414334051061212,
"grad_norm": 29.839000701904297,
"learning_rate": 9.491638173036613e-06,
"loss": 0.1619,
"step": 1110
},
{
"epoch": 0.3506613349738542,
"grad_norm": 9.729668617248535,
"learning_rate": 9.458307493217542e-06,
"loss": 0.1452,
"step": 1140
},
{
"epoch": 0.35988926484158723,
"grad_norm": 25.106536865234375,
"learning_rate": 9.423980958058192e-06,
"loss": 0.168,
"step": 1170
},
{
"epoch": 0.3691171947093202,
"grad_norm": 9.98951244354248,
"learning_rate": 9.388666235101325e-06,
"loss": 0.1508,
"step": 1200
},
{
"epoch": 0.3783451245770532,
"grad_norm": 13.271267890930176,
"learning_rate": 9.352371212621918e-06,
"loss": 0.1673,
"step": 1230
},
{
"epoch": 0.38757305444478624,
"grad_norm": 9.981292724609375,
"learning_rate": 9.315103997865161e-06,
"loss": 0.1527,
"step": 1260
},
{
"epoch": 0.3968009843125192,
"grad_norm": 12.183391571044922,
"learning_rate": 9.27687291523552e-06,
"loss": 0.1625,
"step": 1290
},
{
"epoch": 0.4060289141802522,
"grad_norm": 18.46415138244629,
"learning_rate": 9.237686504437329e-06,
"loss": 0.1572,
"step": 1320
},
{
"epoch": 0.41525684404798524,
"grad_norm": 7.23091459274292,
"learning_rate": 9.197553518567255e-06,
"loss": 0.1795,
"step": 1350
},
{
"epoch": 0.42448477391571826,
"grad_norm": 8.87973403930664,
"learning_rate": 9.156482922159113e-06,
"loss": 0.181,
"step": 1380
},
{
"epoch": 0.4337127037834512,
"grad_norm": 2.988166570663452,
"learning_rate": 9.114483889181465e-06,
"loss": 0.1696,
"step": 1410
},
{
"epoch": 0.44294063365118425,
"grad_norm": 11.367295265197754,
"learning_rate": 9.071565800988406e-06,
"loss": 0.1591,
"step": 1440
},
{
"epoch": 0.45216856351891727,
"grad_norm": 13.207951545715332,
"learning_rate": 9.027738244224062e-06,
"loss": 0.1405,
"step": 1470
},
{
"epoch": 0.4613964933866503,
"grad_norm": 7.953868865966797,
"learning_rate": 8.983011008681199e-06,
"loss": 0.1588,
"step": 1500
},
{
"epoch": 0.47062442325438325,
"grad_norm": 15.074382781982422,
"learning_rate": 8.93739408511448e-06,
"loss": 0.1467,
"step": 1530
},
{
"epoch": 0.47985235312211627,
"grad_norm": 5.083987712860107,
"learning_rate": 8.890897663008815e-06,
"loss": 0.1631,
"step": 1560
},
{
"epoch": 0.4890802829898493,
"grad_norm": 10.089990615844727,
"learning_rate": 8.84353212830334e-06,
"loss": 0.1526,
"step": 1590
},
{
"epoch": 0.4921562596124269,
"eval_accuracy": 0.957732243119853,
"eval_exact_match": 0.7619294605809128,
"eval_loss": 0.150468111038208,
"eval_runtime": 123.479,
"eval_samples_per_second": 46.801,
"eval_steps_per_second": 5.855,
"step": 1600
},
{
"epoch": 0.49830821285758226,
"grad_norm": 10.789036750793457,
"learning_rate": 8.795308061071491e-06,
"loss": 0.1504,
"step": 1620
},
{
"epoch": 0.5075361427253153,
"grad_norm": 40.19456100463867,
"learning_rate": 8.746236233157728e-06,
"loss": 0.1575,
"step": 1650
},
{
"epoch": 0.5167640725930482,
"grad_norm": 11.352813720703125,
"learning_rate": 8.696327605771427e-06,
"loss": 0.1635,
"step": 1680
},
{
"epoch": 0.5259920024607813,
"grad_norm": 4.414275646209717,
"learning_rate": 8.64559332703846e-06,
"loss": 0.1598,
"step": 1710
},
{
"epoch": 0.5352199323285143,
"grad_norm": 7.247404098510742,
"learning_rate": 8.594044729511029e-06,
"loss": 0.1343,
"step": 1740
},
{
"epoch": 0.5444478621962473,
"grad_norm": 16.993389129638672,
"learning_rate": 8.541693327636314e-06,
"loss": 0.1588,
"step": 1770
},
{
"epoch": 0.5536757920639803,
"grad_norm": 6.359425067901611,
"learning_rate": 8.488550815184473e-06,
"loss": 0.1448,
"step": 1800
},
{
"epoch": 0.5629037219317133,
"grad_norm": 18.355993270874023,
"learning_rate": 8.434629062636604e-06,
"loss": 0.1437,
"step": 1830
},
{
"epoch": 0.5721316517994464,
"grad_norm": 8.389277458190918,
"learning_rate": 8.379940114533222e-06,
"loss": 0.1403,
"step": 1860
},
{
"epoch": 0.5813595816671794,
"grad_norm": 27.69739532470703,
"learning_rate": 8.324496186783864e-06,
"loss": 0.1588,
"step": 1890
},
{
"epoch": 0.5905875115349123,
"grad_norm": 2.400268077850342,
"learning_rate": 8.26830966393841e-06,
"loss": 0.1377,
"step": 1920
},
{
"epoch": 0.5998154414026453,
"grad_norm": 5.174127578735352,
"learning_rate": 8.211393096420742e-06,
"loss": 0.134,
"step": 1950
},
{
"epoch": 0.6090433712703783,
"grad_norm": 4.358870983123779,
"learning_rate": 8.15375919772534e-06,
"loss": 0.1407,
"step": 1980
},
{
"epoch": 0.6182713011381114,
"grad_norm": 13.652819633483887,
"learning_rate": 8.095420841577466e-06,
"loss": 0.1232,
"step": 2010
},
{
"epoch": 0.6274992310058444,
"grad_norm": 16.534194946289062,
"learning_rate": 8.036391059057545e-06,
"loss": 0.1278,
"step": 2040
},
{
"epoch": 0.6367271608735774,
"grad_norm": 21.05171775817871,
"learning_rate": 7.976683035690409e-06,
"loss": 0.1357,
"step": 2070
},
{
"epoch": 0.6459550907413104,
"grad_norm": 5.839425086975098,
"learning_rate": 7.916310108500038e-06,
"loss": 0.1321,
"step": 2100
},
{
"epoch": 0.6551830206090433,
"grad_norm": 4.549000263214111,
"learning_rate": 7.855285763030445e-06,
"loss": 0.1303,
"step": 2130
},
{
"epoch": 0.6644109504767763,
"grad_norm": 18.829463958740234,
"learning_rate": 7.793623630333422e-06,
"loss": 0.1415,
"step": 2160
},
{
"epoch": 0.6736388803445094,
"grad_norm": 51.823917388916016,
"learning_rate": 7.731337483923744e-06,
"loss": 0.1405,
"step": 2190
},
{
"epoch": 0.6828668102122424,
"grad_norm": 17.93177032470703,
"learning_rate": 7.668441236702578e-06,
"loss": 0.1375,
"step": 2220
},
{
"epoch": 0.6920947400799754,
"grad_norm": 9.073467254638672,
"learning_rate": 7.6049489378497475e-06,
"loss": 0.1529,
"step": 2250
},
{
"epoch": 0.7013226699477084,
"grad_norm": 6.653486251831055,
"learning_rate": 7.5408747696855575e-06,
"loss": 0.1245,
"step": 2280
},
{
"epoch": 0.7105505998154414,
"grad_norm": 78.53523254394531,
"learning_rate": 7.476233044502874e-06,
"loss": 0.1297,
"step": 2310
},
{
"epoch": 0.7197785296831745,
"grad_norm": 29.850419998168945,
"learning_rate": 7.4110382013701846e-06,
"loss": 0.1408,
"step": 2340
},
{
"epoch": 0.7290064595509074,
"grad_norm": 8.267908096313477,
"learning_rate": 7.345304802906328e-06,
"loss": 0.1323,
"step": 2370
},
{
"epoch": 0.7382343894186404,
"grad_norm": 15.625893592834473,
"learning_rate": 7.279047532027635e-06,
"loss": 0.1301,
"step": 2400
},
{
"epoch": 0.7382343894186404,
"eval_accuracy": 0.9621128756732333,
"eval_exact_match": 0.7802558782849239,
"eval_loss": 0.13456065952777863,
"eval_runtime": 122.4449,
"eval_samples_per_second": 47.197,
"eval_steps_per_second": 5.905,
"step": 2400
},
{
"epoch": 0.7474623192863734,
"grad_norm": 22.09284210205078,
"learning_rate": 7.212281188668191e-06,
"loss": 0.115,
"step": 2430
},
{
"epoch": 0.7566902491541064,
"grad_norm": 22.269826889038086,
"learning_rate": 7.145020686473968e-06,
"loss": 0.1264,
"step": 2460
},
{
"epoch": 0.7659181790218395,
"grad_norm": 12.40841007232666,
"learning_rate": 7.077281049471549e-06,
"loss": 0.1347,
"step": 2490
},
{
"epoch": 0.7751461088895725,
"grad_norm": 10.491952896118164,
"learning_rate": 7.009077408712207e-06,
"loss": 0.1347,
"step": 2520
},
{
"epoch": 0.7843740387573055,
"grad_norm": 18.861055374145508,
"learning_rate": 6.940424998892063e-06,
"loss": 0.1515,
"step": 2550
},
{
"epoch": 0.7936019686250384,
"grad_norm": 8.669054985046387,
"learning_rate": 6.871339154949104e-06,
"loss": 0.1348,
"step": 2580
},
{
"epoch": 0.8028298984927714,
"grad_norm": 8.927881240844727,
"learning_rate": 6.801835308637809e-06,
"loss": 0.1238,
"step": 2610
},
{
"epoch": 0.8120578283605044,
"grad_norm": 26.579429626464844,
"learning_rate": 6.731928985082148e-06,
"loss": 0.1394,
"step": 2640
},
{
"epoch": 0.8212857582282375,
"grad_norm": 19.812923431396484,
"learning_rate": 6.661635799307718e-06,
"loss": 0.1231,
"step": 2670
},
{
"epoch": 0.8305136880959705,
"grad_norm": 12.649471282958984,
"learning_rate": 6.590971452753807e-06,
"loss": 0.1385,
"step": 2700
},
{
"epoch": 0.8397416179637035,
"grad_norm": 12.52941608428955,
"learning_rate": 6.519951729766157e-06,
"loss": 0.1339,
"step": 2730
},
{
"epoch": 0.8489695478314365,
"grad_norm": 2.309412956237793,
"learning_rate": 6.448592494071195e-06,
"loss": 0.1297,
"step": 2760
},
{
"epoch": 0.8581974776991694,
"grad_norm": 11.02414321899414,
"learning_rate": 6.376909685232544e-06,
"loss": 0.1417,
"step": 2790
},
{
"epoch": 0.8674254075669025,
"grad_norm": 10.928828239440918,
"learning_rate": 6.304919315090594e-06,
"loss": 0.1233,
"step": 2820
},
{
"epoch": 0.8766533374346355,
"grad_norm": 7.780458927154541,
"learning_rate": 6.232637464185924e-06,
"loss": 0.1287,
"step": 2850
},
{
"epoch": 0.8858812673023685,
"grad_norm": 21.29149627685547,
"learning_rate": 6.1600802781673775e-06,
"loss": 0.1198,
"step": 2880
},
{
"epoch": 0.8951091971701015,
"grad_norm": 18.80558967590332,
"learning_rate": 6.087263964185601e-06,
"loss": 0.1166,
"step": 2910
},
{
"epoch": 0.9043371270378345,
"grad_norm": 8.064098358154297,
"learning_rate": 6.014204787272842e-06,
"loss": 0.1385,
"step": 2940
},
{
"epoch": 0.9135650569055676,
"grad_norm": 13.307632446289062,
"learning_rate": 5.940919066709812e-06,
"loss": 0.1447,
"step": 2970
},
{
"epoch": 0.9227929867733006,
"grad_norm": 9.524603843688965,
"learning_rate": 5.867423172380441e-06,
"loss": 0.1448,
"step": 3000
},
{
"epoch": 0.9320209166410335,
"grad_norm": 13.8273286819458,
"learning_rate": 5.793733521115325e-06,
"loss": 0.1254,
"step": 3030
},
{
"epoch": 0.9412488465087665,
"grad_norm": 27.4748477935791,
"learning_rate": 5.719866573024684e-06,
"loss": 0.1293,
"step": 3060
},
{
"epoch": 0.9504767763764995,
"grad_norm": 10.63260269165039,
"learning_rate": 5.645838827821658e-06,
"loss": 0.1116,
"step": 3090
},
{
"epoch": 0.9597047062442325,
"grad_norm": 10.844072341918945,
"learning_rate": 5.571666821136751e-06,
"loss": 0.1393,
"step": 3120
},
{
"epoch": 0.9689326361119656,
"grad_norm": 13.513023376464844,
"learning_rate": 5.4973671208242616e-06,
"loss": 0.1247,
"step": 3150
},
{
"epoch": 0.9781605659796986,
"grad_norm": 4.520773410797119,
"learning_rate": 5.422956323261503e-06,
"loss": 0.1181,
"step": 3180
},
{
"epoch": 0.9843125192248539,
"eval_accuracy": 0.9619281811791872,
"eval_exact_match": 0.7844052558782849,
"eval_loss": 0.1342000812292099,
"eval_runtime": 122.3515,
"eval_samples_per_second": 47.233,
"eval_steps_per_second": 5.909,
"step": 3200
},
{
"epoch": 0.9873884958474316,
"grad_norm": 23.68109130859375,
"learning_rate": 5.348451049641662e-06,
"loss": 0.1047,
"step": 3210
},
{
"epoch": 0.9966164257151645,
"grad_norm": 19.535175323486328,
"learning_rate": 5.273867942261107e-06,
"loss": 0.1158,
"step": 3240
},
{
"epoch": 1.0058443555828975,
"grad_norm": 35.44476318359375,
"learning_rate": 5.199223660801993e-06,
"loss": 0.1354,
"step": 3270
},
{
"epoch": 1.0150722854506307,
"grad_norm": 50.02422332763672,
"learning_rate": 5.12453487861097e-06,
"loss": 0.0888,
"step": 3300
},
{
"epoch": 1.0243002153183636,
"grad_norm": 24.165525436401367,
"learning_rate": 5.049818278974851e-06,
"loss": 0.1287,
"step": 3330
},
{
"epoch": 1.0335281451860965,
"grad_norm": 17.49215316772461,
"learning_rate": 4.975090551394051e-06,
"loss": 0.0954,
"step": 3360
},
{
"epoch": 1.0427560750538296,
"grad_norm": 7.993144989013672,
"learning_rate": 4.90036838785464e-06,
"loss": 0.0996,
"step": 3390
},
{
"epoch": 1.0519840049215625,
"grad_norm": 43.465579986572266,
"learning_rate": 4.825668479099844e-06,
"loss": 0.0828,
"step": 3420
},
{
"epoch": 1.0612119347892957,
"grad_norm": 9.943622589111328,
"learning_rate": 4.751007510901823e-06,
"loss": 0.1083,
"step": 3450
},
{
"epoch": 1.0704398646570286,
"grad_norm": 18.9721736907959,
"learning_rate": 4.676402160334561e-06,
"loss": 0.0947,
"step": 3480
},
{
"epoch": 1.0796677945247617,
"grad_norm": 7.654590129852295,
"learning_rate": 4.601869092048686e-06,
"loss": 0.0977,
"step": 3510
},
{
"epoch": 1.0888957243924946,
"grad_norm": 11.893392562866211,
"learning_rate": 4.527424954549082e-06,
"loss": 0.0891,
"step": 3540
},
{
"epoch": 1.0981236542602275,
"grad_norm": 8.733033180236816,
"learning_rate": 4.453086376476103e-06,
"loss": 0.0802,
"step": 3570
},
{
"epoch": 1.1073515841279606,
"grad_norm": 8.004829406738281,
"learning_rate": 4.378869962891213e-06,
"loss": 0.0867,
"step": 3600
},
{
"epoch": 1.1165795139956936,
"grad_norm": 9.916909217834473,
"learning_rate": 4.304792291567912e-06,
"loss": 0.0808,
"step": 3630
},
{
"epoch": 1.1258074438634267,
"grad_norm": 10.153642654418945,
"learning_rate": 4.230869909288737e-06,
"loss": 0.1023,
"step": 3660
},
{
"epoch": 1.1350353737311596,
"grad_norm": 10.010107040405273,
"learning_rate": 4.1571193281492025e-06,
"loss": 0.0979,
"step": 3690
},
{
"epoch": 1.1442633035988927,
"grad_norm": 12.91443157196045,
"learning_rate": 4.08355702186948e-06,
"loss": 0.1037,
"step": 3720
},
{
"epoch": 1.1534912334666256,
"grad_norm": 64.11460876464844,
"learning_rate": 4.010199422114649e-06,
"loss": 0.1009,
"step": 3750
},
{
"epoch": 1.1627191633343585,
"grad_norm": 6.108353614807129,
"learning_rate": 3.937062914824346e-06,
"loss": 0.1096,
"step": 3780
},
{
"epoch": 1.1719470932020917,
"grad_norm": 6.995621681213379,
"learning_rate": 3.864163836552625e-06,
"loss": 0.0899,
"step": 3810
},
{
"epoch": 1.1811750230698246,
"grad_norm": 9.924674034118652,
"learning_rate": 3.7915184708188512e-06,
"loss": 0.1087,
"step": 3840
},
{
"epoch": 1.1904029529375577,
"grad_norm": 38.43867111206055,
"learning_rate": 3.7191430444704353e-06,
"loss": 0.1065,
"step": 3870
},
{
"epoch": 1.1996308828052906,
"grad_norm": 127.10018920898438,
"learning_rate": 3.6470537240582403e-06,
"loss": 0.0967,
"step": 3900
},
{
"epoch": 1.2088588126730238,
"grad_norm": 5.650942325592041,
"learning_rate": 3.575266612225435e-06,
"loss": 0.091,
"step": 3930
},
{
"epoch": 1.2180867425407567,
"grad_norm": 36.26810836791992,
"learning_rate": 3.5037977441106434e-06,
"loss": 0.0988,
"step": 3960
},
{
"epoch": 1.2273146724084896,
"grad_norm": 8.099671363830566,
"learning_rate": 3.4326630837661623e-06,
"loss": 0.0914,
"step": 3990
},
{
"epoch": 1.2303906490310674,
"eval_accuracy": 0.9631609383041336,
"eval_exact_match": 0.7871715076071922,
"eval_loss": 0.13382020592689514,
"eval_runtime": 122.4574,
"eval_samples_per_second": 47.192,
"eval_steps_per_second": 5.904,
"step": 4000
},
{
"epoch": 1.2365426022762227,
"grad_norm": 2.012192726135254,
"learning_rate": 3.361878520592059e-06,
"loss": 0.0996,
"step": 4020
},
{
"epoch": 1.2457705321439556,
"grad_norm": 4.293970108032227,
"learning_rate": 3.2914598657869446e-06,
"loss": 0.0947,
"step": 4050
},
{
"epoch": 1.2549984620116887,
"grad_norm": 13.106005668640137,
"learning_rate": 3.2214228488162163e-06,
"loss": 0.0877,
"step": 4080
},
{
"epoch": 1.2642263918794217,
"grad_norm": 69.8658676147461,
"learning_rate": 3.1517831138985542e-06,
"loss": 0.0941,
"step": 4110
},
{
"epoch": 1.2734543217471548,
"grad_norm": 10.34127426147461,
"learning_rate": 3.082556216511462e-06,
"loss": 0.1074,
"step": 4140
},
{
"epoch": 1.2826822516148877,
"grad_norm": 10.491802215576172,
"learning_rate": 3.013757619916627e-06,
"loss": 0.0851,
"step": 4170
},
{
"epoch": 1.2919101814826206,
"grad_norm": 6.333405494689941,
"learning_rate": 2.9454026917058824e-06,
"loss": 0.1051,
"step": 4200
},
{
"epoch": 1.3011381113503537,
"grad_norm": 11.53557300567627,
"learning_rate": 2.8775067003685366e-06,
"loss": 0.095,
"step": 4230
},
{
"epoch": 1.3103660412180869,
"grad_norm": 11.399128913879395,
"learning_rate": 2.8100848118808415e-06,
"loss": 0.0901,
"step": 4260
},
{
"epoch": 1.3195939710858198,
"grad_norm": 12.951314926147461,
"learning_rate": 2.743152086318348e-06,
"loss": 0.0921,
"step": 4290
},
{
"epoch": 1.3288219009535527,
"grad_norm": 12.634110450744629,
"learning_rate": 2.676723474491948e-06,
"loss": 0.0892,
"step": 4320
},
{
"epoch": 1.3380498308212858,
"grad_norm": 49.48850631713867,
"learning_rate": 2.610813814608271e-06,
"loss": 0.0881,
"step": 4350
},
{
"epoch": 1.3472777606890187,
"grad_norm": 21.164209365844727,
"learning_rate": 2.545437828955291e-06,
"loss": 0.0937,
"step": 4380
},
{
"epoch": 1.3565056905567516,
"grad_norm": 12.729103088378906,
"learning_rate": 2.480610120613778e-06,
"loss": 0.0835,
"step": 4410
},
{
"epoch": 1.3657336204244848,
"grad_norm": 12.80295467376709,
"learning_rate": 2.4163451701954145e-06,
"loss": 0.0884,
"step": 4440
},
{
"epoch": 1.374961550292218,
"grad_norm": 7.701408386230469,
"learning_rate": 2.352657332608233e-06,
"loss": 0.088,
"step": 4470
},
{
"epoch": 1.3841894801599508,
"grad_norm": 17.93044090270996,
"learning_rate": 2.289560833850168e-06,
"loss": 0.0965,
"step": 4500
},
{
"epoch": 1.3934174100276837,
"grad_norm": 7.396183490753174,
"learning_rate": 2.2270697678313687e-06,
"loss": 0.0873,
"step": 4530
},
{
"epoch": 1.4026453398954168,
"grad_norm": 8.487717628479004,
"learning_rate": 2.1651980932260454e-06,
"loss": 0.0852,
"step": 4560
},
{
"epoch": 1.4118732697631498,
"grad_norm": 5.432651042938232,
"learning_rate": 2.103959630354498e-06,
"loss": 0.0975,
"step": 4590
},
{
"epoch": 1.4211011996308829,
"grad_norm": 11.554012298583984,
"learning_rate": 2.0433680580960753e-06,
"loss": 0.1049,
"step": 4620
},
{
"epoch": 1.4303291294986158,
"grad_norm": 4.724039077758789,
"learning_rate": 1.9834369108336987e-06,
"loss": 0.0921,
"step": 4650
},
{
"epoch": 1.439557059366349,
"grad_norm": 6.700310707092285,
"learning_rate": 1.924179575430689e-06,
"loss": 0.1047,
"step": 4680
},
{
"epoch": 1.4487849892340818,
"grad_norm": 9.561090469360352,
"learning_rate": 1.8656092882405296e-06,
"loss": 0.0757,
"step": 4710
},
{
"epoch": 1.4580129191018147,
"grad_norm": 9.078134536743164,
"learning_rate": 1.8077391321502513e-06,
"loss": 0.086,
"step": 4740
},
{
"epoch": 1.4672408489695479,
"grad_norm": 6.513936519622803,
"learning_rate": 1.7505820336580959e-06,
"loss": 0.0757,
"step": 4770
},
{
"epoch": 1.4764687788372808,
"grad_norm": 9.984694480895996,
"learning_rate": 1.6941507599861245e-06,
"loss": 0.081,
"step": 4800
},
{
"epoch": 1.4764687788372808,
"eval_accuracy": 0.9634118308930201,
"eval_exact_match": 0.7880359612724758,
"eval_loss": 0.13220299780368805,
"eval_runtime": 123.5912,
"eval_samples_per_second": 46.759,
"eval_steps_per_second": 5.85,
"step": 4800
},
{
"epoch": 1.485696708705014,
"grad_norm": 9.214570999145508,
"learning_rate": 1.638457916228382e-06,
"loss": 0.0924,
"step": 4830
},
{
"epoch": 1.4949246385727468,
"grad_norm": 20.25838851928711,
"learning_rate": 1.5835159425352975e-06,
"loss": 0.0992,
"step": 4860
},
{
"epoch": 1.50415256844048,
"grad_norm": 14.184989929199219,
"learning_rate": 1.5293371113349166e-06,
"loss": 0.0824,
"step": 4890
},
{
"epoch": 1.5133804983082129,
"grad_norm": 9.43948745727539,
"learning_rate": 1.4759335245915945e-06,
"loss": 0.0973,
"step": 4920
},
{
"epoch": 1.5226084281759458,
"grad_norm": 6.130595684051514,
"learning_rate": 1.4233171111027827e-06,
"loss": 0.0943,
"step": 4950
},
{
"epoch": 1.531836358043679,
"grad_norm": 26.128459930419922,
"learning_rate": 1.3714996238344724e-06,
"loss": 0.087,
"step": 4980
},
{
"epoch": 1.541064287911412,
"grad_norm": 7.372842788696289,
"learning_rate": 1.3204926372959376e-06,
"loss": 0.0966,
"step": 5010
},
{
"epoch": 1.5502922177791447,
"grad_norm": 8.696173667907715,
"learning_rate": 1.270307544954319e-06,
"loss": 0.085,
"step": 5040
},
{
"epoch": 1.5595201476468779,
"grad_norm": 15.655683517456055,
"learning_rate": 1.2209555566896652e-06,
"loss": 0.0936,
"step": 5070
},
{
"epoch": 1.568748077514611,
"grad_norm": 26.870996475219727,
"learning_rate": 1.1724476962909664e-06,
"loss": 0.0943,
"step": 5100
},
{
"epoch": 1.577976007382344,
"grad_norm": 4.403360843658447,
"learning_rate": 1.1247947989937685e-06,
"loss": 0.0819,
"step": 5130
},
{
"epoch": 1.5872039372500768,
"grad_norm": 3.324984073638916,
"learning_rate": 1.0780075090598974e-06,
"loss": 0.0908,
"step": 5160
},
{
"epoch": 1.59643186711781,
"grad_norm": 10.39843463897705,
"learning_rate": 1.0320962773998417e-06,
"loss": 0.0821,
"step": 5190
},
{
"epoch": 1.605659796985543,
"grad_norm": 6.516891002655029,
"learning_rate": 9.870713592383214e-07,
"loss": 0.0836,
"step": 5220
},
{
"epoch": 1.6148877268532758,
"grad_norm": 3.791872262954712,
"learning_rate": 9.429428118235806e-07,
"loss": 0.0872,
"step": 5250
},
{
"epoch": 1.6241156567210089,
"grad_norm": 5.973302364349365,
"learning_rate": 8.997204921808833e-07,
"loss": 0.0951,
"step": 5280
},
{
"epoch": 1.633343586588742,
"grad_norm": 6.1672587394714355,
"learning_rate": 8.57414054910749e-07,
"loss": 0.0955,
"step": 5310
},
{
"epoch": 1.642571516456475,
"grad_norm": 7.527838230133057,
"learning_rate": 8.160329500323933e-07,
"loss": 0.0903,
"step": 5340
},
{
"epoch": 1.6517994463242078,
"grad_norm": 94.20953369140625,
"learning_rate": 7.755864208728736e-07,
"loss": 0.0823,
"step": 5370
},
{
"epoch": 1.661027376191941,
"grad_norm": 17.918153762817383,
"learning_rate": 7.360835020023976e-07,
"loss": 0.0785,
"step": 5400
},
{
"epoch": 1.670255306059674,
"grad_norm": 6.371737957000732,
"learning_rate": 6.975330172162687e-07,
"loss": 0.0864,
"step": 5430
},
{
"epoch": 1.679483235927407,
"grad_norm": 26.493167877197266,
"learning_rate": 6.599435775639046e-07,
"loss": 0.1011,
"step": 5460
},
{
"epoch": 1.68871116579514,
"grad_norm": 5.014997482299805,
"learning_rate": 6.233235794253817e-07,
"loss": 0.0815,
"step": 5490
},
{
"epoch": 1.697939095662873,
"grad_norm": 6.553559303283691,
"learning_rate": 5.876812026359341e-07,
"loss": 0.0955,
"step": 5520
},
{
"epoch": 1.707167025530606,
"grad_norm": 8.305948257446289,
"learning_rate": 5.530244086588094e-07,
"loss": 0.0934,
"step": 5550
},
{
"epoch": 1.7163949553983389,
"grad_norm": 6.3951640129089355,
"learning_rate": 5.193609388069204e-07,
"loss": 0.0816,
"step": 5580
},
{
"epoch": 1.7225469086434944,
"eval_accuracy": 0.964552263058983,
"eval_exact_match": 0.7951244813278008,
"eval_loss": 0.12911444902420044,
"eval_runtime": 122.3841,
"eval_samples_per_second": 47.22,
"eval_steps_per_second": 5.908,
"step": 5600
},
{
"epoch": 1.725622885266072,
"grad_norm": 16.649213790893555,
"learning_rate": 4.86698312513656e-07,
"loss": 0.0866,
"step": 5610
},
{
"epoch": 1.7348508151338051,
"grad_norm": 21.06963539123535,
"learning_rate": 4.5504382565326355e-07,
"loss": 0.083,
"step": 5640
},
{
"epoch": 1.744078745001538,
"grad_norm": 4.905073165893555,
"learning_rate": 4.244045489111598e-07,
"loss": 0.0965,
"step": 5670
},
{
"epoch": 1.753306674869271,
"grad_norm": 29.119220733642578,
"learning_rate": 3.9478732620455315e-07,
"loss": 0.1041,
"step": 5700
},
{
"epoch": 1.762534604737004,
"grad_norm": 24.03986930847168,
"learning_rate": 3.6619877315370644e-07,
"loss": 0.0916,
"step": 5730
},
{
"epoch": 1.771762534604737,
"grad_norm": 9.187104225158691,
"learning_rate": 3.3864527560420744e-07,
"loss": 0.0861,
"step": 5760
},
{
"epoch": 1.78099046447247,
"grad_norm": 63.82421112060547,
"learning_rate": 3.1213298820055084e-07,
"loss": 0.0785,
"step": 5790
},
{
"epoch": 1.790218394340203,
"grad_norm": 31.934934616088867,
"learning_rate": 2.8666783301137813e-07,
"loss": 0.0819,
"step": 5820
},
{
"epoch": 1.7994463242079362,
"grad_norm": 10.350878715515137,
"learning_rate": 2.6225549820665555e-07,
"loss": 0.087,
"step": 5850
},
{
"epoch": 1.808674254075669,
"grad_norm": 11.43632698059082,
"learning_rate": 2.389014367871073e-07,
"loss": 0.1064,
"step": 5880
},
{
"epoch": 1.817902183943402,
"grad_norm": 5.604170799255371,
"learning_rate": 2.1661086536617027e-07,
"loss": 0.0943,
"step": 5910
},
{
"epoch": 1.827130113811135,
"grad_norm": 18.148305892944336,
"learning_rate": 1.953887630047613e-07,
"loss": 0.0919,
"step": 5940
},
{
"epoch": 1.836358043678868,
"grad_norm": 11.869507789611816,
"learning_rate": 1.752398700990937e-07,
"loss": 0.0895,
"step": 5970
},
{
"epoch": 1.845585973546601,
"grad_norm": 9.245363235473633,
"learning_rate": 1.5616868732182046e-07,
"loss": 0.0872,
"step": 6000
},
{
"epoch": 1.854813903414334,
"grad_norm": 12.95739459991455,
"learning_rate": 1.3817947461670532e-07,
"loss": 0.0806,
"step": 6030
},
{
"epoch": 1.8640418332820672,
"grad_norm": 16.675260543823242,
"learning_rate": 1.2127625024708657e-07,
"loss": 0.0939,
"step": 6060
},
{
"epoch": 1.8732697631498,
"grad_norm": 8.455509185791016,
"learning_rate": 1.054627898983046e-07,
"loss": 0.0813,
"step": 6090
},
{
"epoch": 1.882497693017533,
"grad_norm": 13.262840270996094,
"learning_rate": 9.074262583433413e-08,
"loss": 0.0865,
"step": 6120
},
{
"epoch": 1.8917256228852661,
"grad_norm": 14.098983764648438,
"learning_rate": 7.711904610877175e-08,
"loss": 0.1136,
"step": 6150
},
{
"epoch": 1.900953552752999,
"grad_norm": 10.868261337280273,
"learning_rate": 6.45950938303852e-08,
"loss": 0.0903,
"step": 6180
},
{
"epoch": 1.910181482620732,
"grad_norm": 45.73233413696289,
"learning_rate": 5.317356648337268e-08,
"loss": 0.0806,
"step": 6210
},
{
"epoch": 1.919409412488465,
"grad_norm": 26.091588973999023,
"learning_rate": 4.285701530248321e-08,
"loss": 0.0847,
"step": 6240
},
{
"epoch": 1.9286373423561982,
"grad_norm": 19.761188507080078,
"learning_rate": 3.3647744703148575e-08,
"loss": 0.0989,
"step": 6270
},
{
"epoch": 1.9378652722239311,
"grad_norm": 32.86652374267578,
"learning_rate": 2.5547811766744035e-08,
"loss": 0.0824,
"step": 6300
},
{
"epoch": 1.947093202091664,
"grad_norm": 18.7479305267334,
"learning_rate": 1.8559025781094697e-08,
"loss": 0.0808,
"step": 6330
},
{
"epoch": 1.9563211319593972,
"grad_norm": 6.697521686553955,
"learning_rate": 1.2682947836335502e-08,
"loss": 0.0995,
"step": 6360
},
{
"epoch": 1.96554906182713,
"grad_norm": 13.512880325317383,
"learning_rate": 7.920890476209032e-09,
"loss": 0.088,
"step": 6390
},
{
"epoch": 1.9686250384497077,
"eval_accuracy": 0.9644171218664658,
"eval_exact_match": 0.7947786998616874,
"eval_loss": 0.1286803036928177,
"eval_runtime": 122.5547,
"eval_samples_per_second": 47.154,
"eval_steps_per_second": 5.899,
"step": 6400
},
{
"epoch": 1.974776991694863,
"grad_norm": 6.648954391479492,
"learning_rate": 4.27391740488059e-09,
"loss": 0.0888,
"step": 6420
},
{
"epoch": 1.9840049215625961,
"grad_norm": 9.480125427246094,
"learning_rate": 1.7428432493393765e-09,
"loss": 0.0825,
"step": 6450
},
{
"epoch": 1.9932328514303292,
"grad_norm": 8.281903266906738,
"learning_rate": 3.282333774334845e-10,
"loss": 0.1059,
"step": 6480
},
{
"epoch": 2.0,
"step": 6502,
"total_flos": 5.340659231903187e+18,
"train_loss": 0.1384785544208584,
"train_runtime": 10804.874,
"train_samples_per_second": 19.256,
"train_steps_per_second": 0.602
}
],
"logging_steps": 30,
"max_steps": 6502,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 800,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 5.340659231903187e+18,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}