babylm-es / checkpoint-79326 /trainer_state.json
abritez's picture
Upload 21 files
83f5dbd verified
raw
history blame
29.2 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 5.815689149560117,
"eval_steps": 500,
"global_step": 79326,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.036656891495601175,
"grad_norm": 3.9739036560058594,
"learning_rate": 9.936968963517636e-05,
"loss": 7.911,
"step": 500
},
{
"epoch": 0.07331378299120235,
"grad_norm": 3.5818135738372803,
"learning_rate": 9.873937927035273e-05,
"loss": 7.4132,
"step": 1000
},
{
"epoch": 0.10997067448680352,
"grad_norm": 3.6124236583709717,
"learning_rate": 9.810906890552909e-05,
"loss": 7.2685,
"step": 1500
},
{
"epoch": 0.1466275659824047,
"grad_norm": 3.609344720840454,
"learning_rate": 9.747875854070544e-05,
"loss": 7.1646,
"step": 2000
},
{
"epoch": 0.18328445747800587,
"grad_norm": 3.878251075744629,
"learning_rate": 9.684844817588181e-05,
"loss": 7.112,
"step": 2500
},
{
"epoch": 0.21994134897360704,
"grad_norm": 3.640087604522705,
"learning_rate": 9.621813781105817e-05,
"loss": 6.9647,
"step": 3000
},
{
"epoch": 0.2565982404692082,
"grad_norm": 4.396575927734375,
"learning_rate": 9.558782744623454e-05,
"loss": 6.9285,
"step": 3500
},
{
"epoch": 0.2932551319648094,
"grad_norm": 4.27196741104126,
"learning_rate": 9.49575170814109e-05,
"loss": 6.8817,
"step": 4000
},
{
"epoch": 0.32991202346041054,
"grad_norm": 5.039891719818115,
"learning_rate": 9.432720671658725e-05,
"loss": 6.8308,
"step": 4500
},
{
"epoch": 0.36656891495601174,
"grad_norm": 6.211440563201904,
"learning_rate": 9.369689635176361e-05,
"loss": 6.7133,
"step": 5000
},
{
"epoch": 0.4032258064516129,
"grad_norm": 5.089609622955322,
"learning_rate": 9.306658598693997e-05,
"loss": 6.596,
"step": 5500
},
{
"epoch": 0.4398826979472141,
"grad_norm": 7.917068004608154,
"learning_rate": 9.243627562211632e-05,
"loss": 6.5651,
"step": 6000
},
{
"epoch": 0.47653958944281527,
"grad_norm": 4.634731292724609,
"learning_rate": 9.18059652572927e-05,
"loss": 6.4742,
"step": 6500
},
{
"epoch": 0.5131964809384164,
"grad_norm": 5.389685153961182,
"learning_rate": 9.117565489246905e-05,
"loss": 6.3593,
"step": 7000
},
{
"epoch": 0.5498533724340176,
"grad_norm": 6.881243705749512,
"learning_rate": 9.054534452764542e-05,
"loss": 6.2916,
"step": 7500
},
{
"epoch": 0.5865102639296188,
"grad_norm": 6.123986721038818,
"learning_rate": 8.991503416282178e-05,
"loss": 6.2029,
"step": 8000
},
{
"epoch": 0.6231671554252199,
"grad_norm": 6.262916088104248,
"learning_rate": 8.928472379799813e-05,
"loss": 6.1005,
"step": 8500
},
{
"epoch": 0.6598240469208211,
"grad_norm": 7.570896625518799,
"learning_rate": 8.86544134331745e-05,
"loss": 6.0054,
"step": 9000
},
{
"epoch": 0.6964809384164223,
"grad_norm": 6.367081642150879,
"learning_rate": 8.802410306835086e-05,
"loss": 5.8283,
"step": 9500
},
{
"epoch": 0.7331378299120235,
"grad_norm": 6.622028350830078,
"learning_rate": 8.739379270352722e-05,
"loss": 5.7653,
"step": 10000
},
{
"epoch": 0.7697947214076246,
"grad_norm": 6.616864204406738,
"learning_rate": 8.676348233870359e-05,
"loss": 5.6764,
"step": 10500
},
{
"epoch": 0.8064516129032258,
"grad_norm": 7.048656463623047,
"learning_rate": 8.613317197387994e-05,
"loss": 5.5562,
"step": 11000
},
{
"epoch": 0.843108504398827,
"grad_norm": 6.860437870025635,
"learning_rate": 8.55028616090563e-05,
"loss": 5.4657,
"step": 11500
},
{
"epoch": 0.8797653958944281,
"grad_norm": 6.002134799957275,
"learning_rate": 8.487255124423267e-05,
"loss": 5.3649,
"step": 12000
},
{
"epoch": 0.9164222873900293,
"grad_norm": 7.97310209274292,
"learning_rate": 8.424224087940903e-05,
"loss": 5.3526,
"step": 12500
},
{
"epoch": 0.9530791788856305,
"grad_norm": 5.980320930480957,
"learning_rate": 8.361193051458538e-05,
"loss": 5.2898,
"step": 13000
},
{
"epoch": 0.9897360703812317,
"grad_norm": 7.705748558044434,
"learning_rate": 8.298162014976175e-05,
"loss": 5.1842,
"step": 13500
},
{
"epoch": 1.0,
"eval_loss": 5.10765266418457,
"eval_runtime": 51.2669,
"eval_samples_per_second": 751.187,
"eval_steps_per_second": 46.95,
"step": 13640
},
{
"epoch": 1.0263929618768328,
"grad_norm": 7.542203426361084,
"learning_rate": 8.23513097849381e-05,
"loss": 5.1511,
"step": 14000
},
{
"epoch": 1.063049853372434,
"grad_norm": 6.65974235534668,
"learning_rate": 8.172099942011447e-05,
"loss": 5.1524,
"step": 14500
},
{
"epoch": 1.099706744868035,
"grad_norm": 6.591290473937988,
"learning_rate": 8.109068905529082e-05,
"loss": 5.0419,
"step": 15000
},
{
"epoch": 1.1363636363636362,
"grad_norm": 6.578639507293701,
"learning_rate": 8.046037869046718e-05,
"loss": 4.9851,
"step": 15500
},
{
"epoch": 1.1730205278592376,
"grad_norm": 6.83113956451416,
"learning_rate": 7.983006832564355e-05,
"loss": 4.9795,
"step": 16000
},
{
"epoch": 1.2096774193548387,
"grad_norm": 6.948678493499756,
"learning_rate": 7.919975796081991e-05,
"loss": 4.9811,
"step": 16500
},
{
"epoch": 1.2463343108504399,
"grad_norm": 6.949957847595215,
"learning_rate": 7.856944759599628e-05,
"loss": 4.9127,
"step": 17000
},
{
"epoch": 1.282991202346041,
"grad_norm": 6.94000768661499,
"learning_rate": 7.793913723117263e-05,
"loss": 4.9336,
"step": 17500
},
{
"epoch": 1.3196480938416422,
"grad_norm": 6.532993316650391,
"learning_rate": 7.730882686634899e-05,
"loss": 4.8546,
"step": 18000
},
{
"epoch": 1.3563049853372435,
"grad_norm": 7.553442001342773,
"learning_rate": 7.667851650152536e-05,
"loss": 4.7791,
"step": 18500
},
{
"epoch": 1.3929618768328447,
"grad_norm": 6.4063720703125,
"learning_rate": 7.604820613670172e-05,
"loss": 4.8218,
"step": 19000
},
{
"epoch": 1.4296187683284458,
"grad_norm": 8.8884916305542,
"learning_rate": 7.541789577187807e-05,
"loss": 4.756,
"step": 19500
},
{
"epoch": 1.466275659824047,
"grad_norm": 7.412301540374756,
"learning_rate": 7.478758540705444e-05,
"loss": 4.7164,
"step": 20000
},
{
"epoch": 1.502932551319648,
"grad_norm": 5.796535491943359,
"learning_rate": 7.41572750422308e-05,
"loss": 4.7209,
"step": 20500
},
{
"epoch": 1.5395894428152492,
"grad_norm": 6.591666221618652,
"learning_rate": 7.352696467740716e-05,
"loss": 4.6933,
"step": 21000
},
{
"epoch": 1.5762463343108504,
"grad_norm": 8.342963218688965,
"learning_rate": 7.289665431258353e-05,
"loss": 4.6505,
"step": 21500
},
{
"epoch": 1.6129032258064515,
"grad_norm": 7.765003681182861,
"learning_rate": 7.226634394775988e-05,
"loss": 4.5958,
"step": 22000
},
{
"epoch": 1.6495601173020527,
"grad_norm": 8.44519329071045,
"learning_rate": 7.163603358293624e-05,
"loss": 4.5874,
"step": 22500
},
{
"epoch": 1.6862170087976538,
"grad_norm": 10.166460037231445,
"learning_rate": 7.100572321811261e-05,
"loss": 4.588,
"step": 23000
},
{
"epoch": 1.7228739002932552,
"grad_norm": 6.965490341186523,
"learning_rate": 7.037541285328895e-05,
"loss": 4.6057,
"step": 23500
},
{
"epoch": 1.7595307917888563,
"grad_norm": 7.495148181915283,
"learning_rate": 6.974510248846532e-05,
"loss": 4.5788,
"step": 24000
},
{
"epoch": 1.7961876832844574,
"grad_norm": 8.987942695617676,
"learning_rate": 6.911479212364168e-05,
"loss": 4.5289,
"step": 24500
},
{
"epoch": 1.8328445747800588,
"grad_norm": 6.871776580810547,
"learning_rate": 6.848448175881804e-05,
"loss": 4.5326,
"step": 25000
},
{
"epoch": 1.86950146627566,
"grad_norm": 6.7767744064331055,
"learning_rate": 6.78541713939944e-05,
"loss": 4.4945,
"step": 25500
},
{
"epoch": 1.906158357771261,
"grad_norm": 10.17478084564209,
"learning_rate": 6.722386102917076e-05,
"loss": 4.4387,
"step": 26000
},
{
"epoch": 1.9428152492668622,
"grad_norm": 8.18852424621582,
"learning_rate": 6.659355066434713e-05,
"loss": 4.4369,
"step": 26500
},
{
"epoch": 1.9794721407624634,
"grad_norm": 10.012794494628906,
"learning_rate": 6.596324029952349e-05,
"loss": 4.3956,
"step": 27000
},
{
"epoch": 2.0,
"eval_loss": 4.3589653968811035,
"eval_runtime": 52.0275,
"eval_samples_per_second": 740.205,
"eval_steps_per_second": 46.264,
"step": 27280
},
{
"epoch": 2.0161290322580645,
"grad_norm": 7.973904132843018,
"learning_rate": 6.533292993469985e-05,
"loss": 4.4279,
"step": 27500
},
{
"epoch": 2.0527859237536656,
"grad_norm": 7.848902225494385,
"learning_rate": 6.470261956987622e-05,
"loss": 4.3801,
"step": 28000
},
{
"epoch": 2.089442815249267,
"grad_norm": 7.455059051513672,
"learning_rate": 6.407230920505257e-05,
"loss": 4.3868,
"step": 28500
},
{
"epoch": 2.126099706744868,
"grad_norm": 8.987055778503418,
"learning_rate": 6.344199884022893e-05,
"loss": 4.3451,
"step": 29000
},
{
"epoch": 2.162756598240469,
"grad_norm": 7.139368534088135,
"learning_rate": 6.28116884754053e-05,
"loss": 4.3601,
"step": 29500
},
{
"epoch": 2.19941348973607,
"grad_norm": 7.4019317626953125,
"learning_rate": 6.218137811058166e-05,
"loss": 4.3565,
"step": 30000
},
{
"epoch": 2.236070381231672,
"grad_norm": 8.461463928222656,
"learning_rate": 6.155106774575801e-05,
"loss": 4.2893,
"step": 30500
},
{
"epoch": 2.2727272727272725,
"grad_norm": 8.040552139282227,
"learning_rate": 6.0920757380934376e-05,
"loss": 4.2938,
"step": 31000
},
{
"epoch": 2.309384164222874,
"grad_norm": 8.279471397399902,
"learning_rate": 6.029044701611074e-05,
"loss": 4.2922,
"step": 31500
},
{
"epoch": 2.346041055718475,
"grad_norm": 7.4184699058532715,
"learning_rate": 5.96601366512871e-05,
"loss": 4.2655,
"step": 32000
},
{
"epoch": 2.3826979472140764,
"grad_norm": 9.337892532348633,
"learning_rate": 5.902982628646345e-05,
"loss": 4.2428,
"step": 32500
},
{
"epoch": 2.4193548387096775,
"grad_norm": 8.708230018615723,
"learning_rate": 5.8399515921639816e-05,
"loss": 4.2081,
"step": 33000
},
{
"epoch": 2.4560117302052786,
"grad_norm": 7.0168938636779785,
"learning_rate": 5.776920555681617e-05,
"loss": 4.2089,
"step": 33500
},
{
"epoch": 2.4926686217008798,
"grad_norm": 9.225556373596191,
"learning_rate": 5.7138895191992536e-05,
"loss": 4.226,
"step": 34000
},
{
"epoch": 2.529325513196481,
"grad_norm": 7.094930648803711,
"learning_rate": 5.65085848271689e-05,
"loss": 4.1786,
"step": 34500
},
{
"epoch": 2.565982404692082,
"grad_norm": 7.8396711349487305,
"learning_rate": 5.587827446234526e-05,
"loss": 4.2071,
"step": 35000
},
{
"epoch": 2.602639296187683,
"grad_norm": 8.818264961242676,
"learning_rate": 5.524796409752162e-05,
"loss": 4.1563,
"step": 35500
},
{
"epoch": 2.6392961876832843,
"grad_norm": 7.717746734619141,
"learning_rate": 5.461765373269798e-05,
"loss": 4.165,
"step": 36000
},
{
"epoch": 2.6759530791788855,
"grad_norm": 6.068331718444824,
"learning_rate": 5.3987343367874346e-05,
"loss": 4.1111,
"step": 36500
},
{
"epoch": 2.712609970674487,
"grad_norm": 8.796777725219727,
"learning_rate": 5.33570330030507e-05,
"loss": 4.1574,
"step": 37000
},
{
"epoch": 2.7492668621700878,
"grad_norm": 8.890911102294922,
"learning_rate": 5.2726722638227066e-05,
"loss": 4.1259,
"step": 37500
},
{
"epoch": 2.7859237536656893,
"grad_norm": 8.333062171936035,
"learning_rate": 5.209641227340343e-05,
"loss": 4.1107,
"step": 38000
},
{
"epoch": 2.8225806451612905,
"grad_norm": 8.116930961608887,
"learning_rate": 5.146610190857979e-05,
"loss": 4.1155,
"step": 38500
},
{
"epoch": 2.8592375366568916,
"grad_norm": 11.292548179626465,
"learning_rate": 5.083579154375615e-05,
"loss": 4.0813,
"step": 39000
},
{
"epoch": 2.8958944281524928,
"grad_norm": 8.076095581054688,
"learning_rate": 5.020548117893251e-05,
"loss": 4.1091,
"step": 39500
},
{
"epoch": 2.932551319648094,
"grad_norm": 6.882072925567627,
"learning_rate": 4.957517081410887e-05,
"loss": 4.082,
"step": 40000
},
{
"epoch": 2.969208211143695,
"grad_norm": 8.762887954711914,
"learning_rate": 4.894486044928523e-05,
"loss": 4.0442,
"step": 40500
},
{
"epoch": 3.0,
"eval_loss": 3.9874393939971924,
"eval_runtime": 51.3732,
"eval_samples_per_second": 749.632,
"eval_steps_per_second": 46.853,
"step": 40920
},
{
"epoch": 3.005865102639296,
"grad_norm": 6.891060829162598,
"learning_rate": 4.831455008446159e-05,
"loss": 4.0661,
"step": 41000
},
{
"epoch": 3.0425219941348973,
"grad_norm": 9.079097747802734,
"learning_rate": 4.768423971963795e-05,
"loss": 3.999,
"step": 41500
},
{
"epoch": 3.0791788856304985,
"grad_norm": 9.445289611816406,
"learning_rate": 4.7053929354814315e-05,
"loss": 4.045,
"step": 42000
},
{
"epoch": 3.1158357771260996,
"grad_norm": 7.992264270782471,
"learning_rate": 4.642361898999067e-05,
"loss": 4.0385,
"step": 42500
},
{
"epoch": 3.1524926686217007,
"grad_norm": 8.101459503173828,
"learning_rate": 4.5793308625167035e-05,
"loss": 3.9818,
"step": 43000
},
{
"epoch": 3.189149560117302,
"grad_norm": 7.8059492111206055,
"learning_rate": 4.51629982603434e-05,
"loss": 3.9905,
"step": 43500
},
{
"epoch": 3.225806451612903,
"grad_norm": 10.241538047790527,
"learning_rate": 4.4532687895519755e-05,
"loss": 3.9456,
"step": 44000
},
{
"epoch": 3.2624633431085046,
"grad_norm": 8.502156257629395,
"learning_rate": 4.390237753069611e-05,
"loss": 3.9347,
"step": 44500
},
{
"epoch": 3.2991202346041058,
"grad_norm": 7.6501641273498535,
"learning_rate": 4.3272067165872475e-05,
"loss": 3.9469,
"step": 45000
},
{
"epoch": 3.335777126099707,
"grad_norm": 8.457347869873047,
"learning_rate": 4.264175680104884e-05,
"loss": 3.9397,
"step": 45500
},
{
"epoch": 3.372434017595308,
"grad_norm": 8.348608016967773,
"learning_rate": 4.20114464362252e-05,
"loss": 3.9413,
"step": 46000
},
{
"epoch": 3.409090909090909,
"grad_norm": 10.004755020141602,
"learning_rate": 4.138113607140156e-05,
"loss": 3.9057,
"step": 46500
},
{
"epoch": 3.4457478005865103,
"grad_norm": 7.643105983734131,
"learning_rate": 4.075082570657792e-05,
"loss": 3.9165,
"step": 47000
},
{
"epoch": 3.4824046920821115,
"grad_norm": 9.214630126953125,
"learning_rate": 4.0120515341754285e-05,
"loss": 3.8871,
"step": 47500
},
{
"epoch": 3.5190615835777126,
"grad_norm": 7.3644700050354,
"learning_rate": 3.949020497693065e-05,
"loss": 3.9115,
"step": 48000
},
{
"epoch": 3.5557184750733137,
"grad_norm": 8.032934188842773,
"learning_rate": 3.8859894612107e-05,
"loss": 3.8775,
"step": 48500
},
{
"epoch": 3.592375366568915,
"grad_norm": 8.48763370513916,
"learning_rate": 3.822958424728336e-05,
"loss": 3.8719,
"step": 49000
},
{
"epoch": 3.629032258064516,
"grad_norm": 7.765323162078857,
"learning_rate": 3.7599273882459725e-05,
"loss": 3.8727,
"step": 49500
},
{
"epoch": 3.665689149560117,
"grad_norm": 9.104835510253906,
"learning_rate": 3.696896351763609e-05,
"loss": 3.847,
"step": 50000
},
{
"epoch": 3.7023460410557183,
"grad_norm": 7.518016815185547,
"learning_rate": 3.6338653152812445e-05,
"loss": 3.8503,
"step": 50500
},
{
"epoch": 3.73900293255132,
"grad_norm": 9.076459884643555,
"learning_rate": 3.570834278798881e-05,
"loss": 3.8496,
"step": 51000
},
{
"epoch": 3.7756598240469206,
"grad_norm": 9.251144409179688,
"learning_rate": 3.507803242316517e-05,
"loss": 3.8394,
"step": 51500
},
{
"epoch": 3.812316715542522,
"grad_norm": 9.594878196716309,
"learning_rate": 3.444772205834153e-05,
"loss": 3.8104,
"step": 52000
},
{
"epoch": 3.8489736070381233,
"grad_norm": 7.103595733642578,
"learning_rate": 3.381741169351789e-05,
"loss": 3.8329,
"step": 52500
},
{
"epoch": 3.8856304985337244,
"grad_norm": 7.397706985473633,
"learning_rate": 3.3187101328694254e-05,
"loss": 3.8172,
"step": 53000
},
{
"epoch": 3.9222873900293256,
"grad_norm": 8.312359809875488,
"learning_rate": 3.255679096387061e-05,
"loss": 3.8246,
"step": 53500
},
{
"epoch": 3.9589442815249267,
"grad_norm": 7.361260414123535,
"learning_rate": 3.192648059904697e-05,
"loss": 3.8078,
"step": 54000
},
{
"epoch": 3.995601173020528,
"grad_norm": 9.614564895629883,
"learning_rate": 3.129617023422333e-05,
"loss": 3.7865,
"step": 54500
},
{
"epoch": 4.0,
"eval_loss": 3.7747433185577393,
"eval_runtime": 51.2384,
"eval_samples_per_second": 751.604,
"eval_steps_per_second": 46.976,
"step": 54560
},
{
"epoch": 4.032258064516129,
"grad_norm": 9.029661178588867,
"learning_rate": 3.0665859869399694e-05,
"loss": 3.7978,
"step": 55000
},
{
"epoch": 4.068914956011731,
"grad_norm": 9.358321189880371,
"learning_rate": 3.0035549504576054e-05,
"loss": 3.821,
"step": 55500
},
{
"epoch": 4.105571847507331,
"grad_norm": 9.089591979980469,
"learning_rate": 2.9405239139752418e-05,
"loss": 3.7474,
"step": 56000
},
{
"epoch": 4.142228739002933,
"grad_norm": 9.139451026916504,
"learning_rate": 2.8774928774928778e-05,
"loss": 3.7571,
"step": 56500
},
{
"epoch": 4.178885630498534,
"grad_norm": 9.728781700134277,
"learning_rate": 2.8144618410105137e-05,
"loss": 3.8133,
"step": 57000
},
{
"epoch": 4.215542521994135,
"grad_norm": 8.330384254455566,
"learning_rate": 2.75143080452815e-05,
"loss": 3.7412,
"step": 57500
},
{
"epoch": 4.252199413489736,
"grad_norm": 9.436174392700195,
"learning_rate": 2.6883997680457857e-05,
"loss": 3.7338,
"step": 58000
},
{
"epoch": 4.288856304985337,
"grad_norm": 12.170429229736328,
"learning_rate": 2.6253687315634217e-05,
"loss": 3.7215,
"step": 58500
},
{
"epoch": 4.325513196480938,
"grad_norm": 8.019417762756348,
"learning_rate": 2.562337695081058e-05,
"loss": 3.7096,
"step": 59000
},
{
"epoch": 4.36217008797654,
"grad_norm": 7.6509480476379395,
"learning_rate": 2.499306658598694e-05,
"loss": 3.7279,
"step": 59500
},
{
"epoch": 4.39882697947214,
"grad_norm": 8.255581855773926,
"learning_rate": 2.43627562211633e-05,
"loss": 3.749,
"step": 60000
},
{
"epoch": 4.435483870967742,
"grad_norm": 9.91395378112793,
"learning_rate": 2.3732445856339664e-05,
"loss": 3.6773,
"step": 60500
},
{
"epoch": 4.472140762463344,
"grad_norm": 9.325223922729492,
"learning_rate": 2.3102135491516024e-05,
"loss": 3.7204,
"step": 61000
},
{
"epoch": 4.508797653958944,
"grad_norm": 9.393867492675781,
"learning_rate": 2.2471825126692384e-05,
"loss": 3.6893,
"step": 61500
},
{
"epoch": 4.545454545454545,
"grad_norm": 8.97889232635498,
"learning_rate": 2.1841514761868744e-05,
"loss": 3.6765,
"step": 62000
},
{
"epoch": 4.5821114369501466,
"grad_norm": 7.196798324584961,
"learning_rate": 2.1211204397045107e-05,
"loss": 3.6679,
"step": 62500
},
{
"epoch": 4.618768328445748,
"grad_norm": 9.22948932647705,
"learning_rate": 2.0580894032221467e-05,
"loss": 3.6591,
"step": 63000
},
{
"epoch": 4.655425219941349,
"grad_norm": 8.780097961425781,
"learning_rate": 1.9950583667397827e-05,
"loss": 3.6391,
"step": 63500
},
{
"epoch": 4.69208211143695,
"grad_norm": 10.625210762023926,
"learning_rate": 1.9320273302574187e-05,
"loss": 3.6819,
"step": 64000
},
{
"epoch": 4.728739002932551,
"grad_norm": 7.736958980560303,
"learning_rate": 1.868996293775055e-05,
"loss": 3.6486,
"step": 64500
},
{
"epoch": 4.765395894428153,
"grad_norm": 8.629966735839844,
"learning_rate": 1.805965257292691e-05,
"loss": 3.6629,
"step": 65000
},
{
"epoch": 4.802052785923753,
"grad_norm": 7.5463762283325195,
"learning_rate": 1.7429342208103274e-05,
"loss": 3.6398,
"step": 65500
},
{
"epoch": 4.838709677419355,
"grad_norm": 7.877403736114502,
"learning_rate": 1.679903184327963e-05,
"loss": 3.6593,
"step": 66000
},
{
"epoch": 4.875366568914956,
"grad_norm": 8.22919750213623,
"learning_rate": 1.6168721478455993e-05,
"loss": 3.6139,
"step": 66500
},
{
"epoch": 4.912023460410557,
"grad_norm": 9.631176948547363,
"learning_rate": 1.5538411113632353e-05,
"loss": 3.6452,
"step": 67000
},
{
"epoch": 4.948680351906159,
"grad_norm": 8.381847381591797,
"learning_rate": 1.4908100748808715e-05,
"loss": 3.6273,
"step": 67500
},
{
"epoch": 4.9853372434017595,
"grad_norm": 9.712096214294434,
"learning_rate": 1.4277790383985077e-05,
"loss": 3.5952,
"step": 68000
},
{
"epoch": 5.0,
"eval_loss": 3.599792003631592,
"eval_runtime": 51.3354,
"eval_samples_per_second": 750.184,
"eval_steps_per_second": 46.888,
"step": 68200
},
{
"epoch": 5.021994134897361,
"grad_norm": 9.814529418945312,
"learning_rate": 1.3647480019161435e-05,
"loss": 3.6124,
"step": 68500
},
{
"epoch": 5.058651026392962,
"grad_norm": 9.232531547546387,
"learning_rate": 1.3017169654337797e-05,
"loss": 3.6207,
"step": 69000
},
{
"epoch": 5.095307917888563,
"grad_norm": 9.92640495300293,
"learning_rate": 1.2386859289514158e-05,
"loss": 3.6176,
"step": 69500
},
{
"epoch": 5.131964809384164,
"grad_norm": 9.3385591506958,
"learning_rate": 1.1756548924690518e-05,
"loss": 3.6151,
"step": 70000
},
{
"epoch": 5.168621700879766,
"grad_norm": 8.617002487182617,
"learning_rate": 1.112623855986688e-05,
"loss": 3.621,
"step": 70500
},
{
"epoch": 5.205278592375366,
"grad_norm": 8.497230529785156,
"learning_rate": 1.049592819504324e-05,
"loss": 3.6085,
"step": 71000
},
{
"epoch": 5.241935483870968,
"grad_norm": 8.745635986328125,
"learning_rate": 9.8656178302196e-06,
"loss": 3.5636,
"step": 71500
},
{
"epoch": 5.278592375366569,
"grad_norm": 7.370077610015869,
"learning_rate": 9.235307465395961e-06,
"loss": 3.5896,
"step": 72000
},
{
"epoch": 5.31524926686217,
"grad_norm": 9.268915176391602,
"learning_rate": 8.604997100572321e-06,
"loss": 3.5571,
"step": 72500
},
{
"epoch": 5.351906158357771,
"grad_norm": 8.88790225982666,
"learning_rate": 7.974686735748683e-06,
"loss": 3.5424,
"step": 73000
},
{
"epoch": 5.3885630498533725,
"grad_norm": 6.400046348571777,
"learning_rate": 7.344376370925044e-06,
"loss": 3.5482,
"step": 73500
},
{
"epoch": 5.425219941348973,
"grad_norm": 8.085587501525879,
"learning_rate": 6.714066006101405e-06,
"loss": 3.5732,
"step": 74000
},
{
"epoch": 5.461876832844575,
"grad_norm": 8.79615306854248,
"learning_rate": 6.083755641277765e-06,
"loss": 3.5689,
"step": 74500
},
{
"epoch": 5.4985337243401755,
"grad_norm": 9.852959632873535,
"learning_rate": 5.453445276454126e-06,
"loss": 3.542,
"step": 75000
},
{
"epoch": 5.535190615835777,
"grad_norm": 9.936841011047363,
"learning_rate": 4.823134911630487e-06,
"loss": 3.5407,
"step": 75500
},
{
"epoch": 5.571847507331379,
"grad_norm": 9.2828950881958,
"learning_rate": 4.192824546806848e-06,
"loss": 3.547,
"step": 76000
},
{
"epoch": 5.608504398826979,
"grad_norm": 8.38242244720459,
"learning_rate": 3.5625141819832086e-06,
"loss": 3.5334,
"step": 76500
},
{
"epoch": 5.645161290322581,
"grad_norm": 7.721536159515381,
"learning_rate": 2.9322038171595694e-06,
"loss": 3.5499,
"step": 77000
},
{
"epoch": 5.681818181818182,
"grad_norm": 8.407939910888672,
"learning_rate": 2.30189345233593e-06,
"loss": 3.4661,
"step": 77500
},
{
"epoch": 5.718475073313783,
"grad_norm": 9.764538764953613,
"learning_rate": 1.671583087512291e-06,
"loss": 3.5483,
"step": 78000
},
{
"epoch": 5.755131964809384,
"grad_norm": 8.228140830993652,
"learning_rate": 1.0412727226886518e-06,
"loss": 3.5607,
"step": 78500
},
{
"epoch": 5.7917888563049855,
"grad_norm": 11.577902793884277,
"learning_rate": 4.1096235786501275e-07,
"loss": 3.5589,
"step": 79000
}
],
"logging_steps": 500,
"max_steps": 79326,
"num_input_tokens_seen": 0,
"num_train_epochs": 6,
"save_steps": 8192,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 2.5090344956154816e+16,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}