n44aa8fe29862 / trainer_state.json
insagur's picture
Add files using upload-large-folder tool
cba6eab verified
Raw
History Blame Contribute Delete
66.6 kB
{
"best_global_step": 1500,
"best_metric": 0.78310341,
"best_model_checkpoint": "/NHNHOME/data/sanghyeok/qwen-cua/runs/probe2000_idm_code_coord/v3-20260621-233930/checkpoint-1500",
"epoch": 0.34594594594594597,
"eval_steps": 500,
"global_step": 1500,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.00023063063063063062,
"grad_norm": 1590.688720703125,
"learning_rate": 1.0000000000000001e-07,
"loss": 49.45393371582031,
"loss_idm": 3.06640625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0621337890625,
"loss_idm_text": 0.0,
"loss_idm_type": 3.003211200237274,
"loss_sft": 0.013178251218050718,
"step": 1,
"token_acc": 0.855064946292334
},
{
"epoch": 0.0023063063063063064,
"grad_norm": 1611.734375,
"learning_rate": 1.0000000000000002e-06,
"loss": 47.76775105794271,
"loss_idm": 3.451388888888889,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.05978732638888889,
"loss_idm_text": 0.0,
"loss_idm_type": 2.8977474106682672,
"loss_sft": 0.013165522780683305,
"step": 10,
"token_acc": 0.8557971074689107
},
{
"epoch": 0.004612612612612613,
"grad_norm": 339.4031982421875,
"learning_rate": 2.0000000000000003e-06,
"loss": 24.062301635742188,
"loss_idm": 1.3859375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0595703125,
"loss_idm_text": 0.0,
"loss_idm_type": 1.4169265031814575,
"loss_sft": 0.013038619607686996,
"step": 20,
"token_acc": 0.8575614504505172
},
{
"epoch": 0.006918918918918919,
"grad_norm": 117.9345474243164,
"learning_rate": 3e-06,
"loss": 17.185517883300783,
"loss_idm": 0.934375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0529296875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.9935595035552979,
"loss_sft": 0.013021596148610115,
"step": 30,
"token_acc": 0.8576696365011294
},
{
"epoch": 0.009225225225225226,
"grad_norm": 174.67822265625,
"learning_rate": 4.000000000000001e-06,
"loss": 15.97247314453125,
"loss_idm": 0.9,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.049169921875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.921882438659668,
"loss_sft": 0.01299305036664009,
"step": 40,
"token_acc": 0.8573861458475359
},
{
"epoch": 0.011531531531531532,
"grad_norm": 383.7381591796875,
"learning_rate": 5e-06,
"loss": 12.401361846923828,
"loss_idm": 0.69296875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.044677734375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.7016407489776612,
"loss_sft": 0.013680103048682214,
"step": 50,
"token_acc": 0.8517720027290206
},
{
"epoch": 0.013837837837837838,
"grad_norm": 276.0512390136719,
"learning_rate": 6e-06,
"loss": 10.131932067871094,
"loss_idm": 0.559375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.04560546875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5592228412628174,
"loss_sft": 0.01410299502313137,
"step": 60,
"token_acc": 0.8488453901987737
},
{
"epoch": 0.016144144144144144,
"grad_norm": 236.4015655517578,
"learning_rate": 7e-06,
"loss": 10.888622283935547,
"loss_idm": 0.61328125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.03974609375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.6118354797363281,
"loss_sft": 0.013782905787229538,
"step": 70,
"token_acc": 0.8515422419612366
},
{
"epoch": 0.01845045045045045,
"grad_norm": 173.59402465820312,
"learning_rate": 8.000000000000001e-06,
"loss": 10.349563598632812,
"loss_idm": 0.5734375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.041650390625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5762068748474121,
"loss_sft": 0.013817435875535012,
"step": 80,
"token_acc": 0.8507013906275447
},
{
"epoch": 0.020756756756756756,
"grad_norm": 231.35693359375,
"learning_rate": 9e-06,
"loss": 10.654405212402343,
"loss_idm": 0.59375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0390625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5972403526306153,
"loss_sft": 0.01390336975455284,
"step": 90,
"token_acc": 0.8490709364931573
},
{
"epoch": 0.023063063063063063,
"grad_norm": 179.43270874023438,
"learning_rate": 1e-05,
"loss": 10.548146057128907,
"loss_idm": 0.584375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.03662109375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5928622126579285,
"loss_sft": 0.014141988009214401,
"step": 100,
"token_acc": 0.847036073395476
},
{
"epoch": 0.025369369369369368,
"grad_norm": 221.4450225830078,
"learning_rate": 9.999316524962347e-06,
"loss": 10.619392395019531,
"loss_idm": 0.5953125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.033154296875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.6010093808174133,
"loss_sft": 0.014010535180568695,
"step": 110,
"token_acc": 0.8487004546091859
},
{
"epoch": 0.027675675675675675,
"grad_norm": 186.53411865234375,
"learning_rate": 9.99726628670463e-06,
"loss": 10.216886138916015,
"loss_idm": 0.57109375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.03154296875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5778079628944397,
"loss_sft": 0.013904009386897087,
"step": 120,
"token_acc": 0.8502426663442629
},
{
"epoch": 0.029981981981981983,
"grad_norm": 157.05239868164062,
"learning_rate": 9.993849845741525e-06,
"loss": 10.855589294433594,
"loss_idm": 0.6046875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.03525390625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.6137941002845764,
"loss_sft": 0.014121369644999503,
"step": 130,
"token_acc": 0.8481210495695818
},
{
"epoch": 0.03228828828828829,
"grad_norm": 166.25057983398438,
"learning_rate": 9.989068136093873e-06,
"loss": 10.426358032226563,
"loss_idm": 0.58203125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.030859375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5915833473205566,
"loss_sft": 0.01406811736524105,
"step": 140,
"token_acc": 0.8483285794652639
},
{
"epoch": 0.034594594594594595,
"grad_norm": 140.43893432617188,
"learning_rate": 9.98292246503335e-06,
"loss": 10.939217376708985,
"loss_idm": 0.60859375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.03046875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.6234304666519165,
"loss_sft": 0.014257545396685601,
"step": 150,
"token_acc": 0.8462056666356372
},
{
"epoch": 0.0369009009009009,
"grad_norm": 101.10719299316406,
"learning_rate": 9.975414512725058e-06,
"loss": 9.8798828125,
"loss_idm": 0.54765625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.029736328125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5589980244636535,
"loss_sft": 0.013876666501164436,
"step": 160,
"token_acc": 0.8505486664698356
},
{
"epoch": 0.03920720720720721,
"grad_norm": 144.91943359375,
"learning_rate": 9.966546331768192e-06,
"loss": 10.420352172851562,
"loss_idm": 0.5953125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.029296875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5927331566810607,
"loss_sft": 0.014148441702127456,
"step": 170,
"token_acc": 0.8475072962442912
},
{
"epoch": 0.04151351351351351,
"grad_norm": 168.03384399414062,
"learning_rate": 9.956320346634877e-06,
"loss": 10.239119720458984,
"loss_idm": 0.56328125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.028076171875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5830882906913757,
"loss_sft": 0.013758457824587822,
"step": 180,
"token_acc": 0.851148950239709
},
{
"epoch": 0.04381981981981982,
"grad_norm": 145.99766540527344,
"learning_rate": 9.944739353007344e-06,
"loss": 10.057091522216798,
"loss_idm": 0.5609375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.026904296875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5724347591400146,
"loss_sft": 0.014053185656666755,
"step": 190,
"token_acc": 0.8491978583015695
},
{
"epoch": 0.04612612612612613,
"grad_norm": 121.27855682373047,
"learning_rate": 9.931806517013612e-06,
"loss": 10.908540344238281,
"loss_idm": 0.61328125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.026611328125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.6256417632102966,
"loss_sft": 0.014259455353021621,
"step": 200,
"token_acc": 0.8467008716078382
},
{
"epoch": 0.048432432432432435,
"grad_norm": 147.652099609375,
"learning_rate": 9.917525374361913e-06,
"loss": 10.310504150390624,
"loss_idm": 0.5828125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.02548828125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5899111151695251,
"loss_sft": 0.014259019494056701,
"step": 210,
"token_acc": 0.8468536389223965
},
{
"epoch": 0.050738738738738735,
"grad_norm": 133.29347229003906,
"learning_rate": 9.901899829374048e-06,
"loss": 9.91939468383789,
"loss_idm": 0.55078125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.028125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5629432678222657,
"loss_sft": 0.013959363475441933,
"step": 220,
"token_acc": 0.8485922361662006
},
{
"epoch": 0.05304504504504504,
"grad_norm": 105.41419219970703,
"learning_rate": 9.884934153917998e-06,
"loss": 10.20447235107422,
"loss_idm": 0.57265625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0267578125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.582277512550354,
"loss_sft": 0.014081236347556115,
"step": 230,
"token_acc": 0.8486716679030506
},
{
"epoch": 0.05535135135135135,
"grad_norm": 108.32715606689453,
"learning_rate": 9.86663298624003e-06,
"loss": 10.037399291992188,
"loss_idm": 0.56328125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.025537109375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5717163324356079,
"loss_sft": 0.014507130160927773,
"step": 240,
"token_acc": 0.8445775773754048
},
{
"epoch": 0.05765765765765766,
"grad_norm": 101.92463684082031,
"learning_rate": 9.847001329696653e-06,
"loss": 10.207322692871093,
"loss_idm": 0.5765625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0251953125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5826281547546387,
"loss_sft": 0.014590685814619064,
"step": 250,
"token_acc": 0.84426266631776
},
{
"epoch": 0.059963963963963966,
"grad_norm": 120.43878173828125,
"learning_rate": 9.826044551386743e-06,
"loss": 10.321310424804688,
"loss_idm": 0.5796875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.02509765625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5899416208267212,
"loss_sft": 0.014607197791337966,
"step": 260,
"token_acc": 0.8437843269576525
},
{
"epoch": 0.06227027027027027,
"grad_norm": 112.54505920410156,
"learning_rate": 9.803768380684242e-06,
"loss": 9.775898742675782,
"loss_idm": 0.55390625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.02724609375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5538631200790405,
"loss_sft": 0.014404882118105889,
"step": 270,
"token_acc": 0.8459256892908046
},
{
"epoch": 0.06457657657657657,
"grad_norm": 94.44606018066406,
"learning_rate": 9.780178907671788e-06,
"loss": 10.339688873291015,
"loss_idm": 0.5859375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.026123046875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5898990511894227,
"loss_sft": 0.014457888901233673,
"step": 280,
"token_acc": 0.8451656208816066
},
{
"epoch": 0.06688288288288288,
"grad_norm": 93.52350616455078,
"learning_rate": 9.755282581475769e-06,
"loss": 9.293934631347657,
"loss_idm": 0.51640625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.02529296875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5259280681610108,
"loss_sft": 0.014395680651068688,
"step": 290,
"token_acc": 0.8458327979361675
},
{
"epoch": 0.06918918918918919,
"grad_norm": 113.6584701538086,
"learning_rate": 9.729086208503174e-06,
"loss": 9.748943328857422,
"loss_idm": 0.5515625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0262939453125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5521375179290772,
"loss_sft": 0.014811449870467185,
"step": 300,
"token_acc": 0.8420287020511943
},
{
"epoch": 0.0714954954954955,
"grad_norm": 81.37059783935547,
"learning_rate": 9.701596950580807e-06,
"loss": 9.784170532226563,
"loss_idm": 0.54453125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0253662109375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5561857938766479,
"loss_sft": 0.014250487089157104,
"step": 310,
"token_acc": 0.847669727830882
},
{
"epoch": 0.0738018018018018,
"grad_norm": 108.87104797363281,
"learning_rate": 9.672822322997305e-06,
"loss": 9.601348876953125,
"loss_idm": 0.5375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.023486328125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5469521760940552,
"loss_sft": 0.014529699832201004,
"step": 320,
"token_acc": 0.8448489136666951
},
{
"epoch": 0.07610810810810811,
"grad_norm": 130.01808166503906,
"learning_rate": 9.642770192448537e-06,
"loss": 9.793380737304688,
"loss_idm": 0.54375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.023828125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5585392713546753,
"loss_sft": 0.014446238800883294,
"step": 330,
"token_acc": 0.8454624950201325
},
{
"epoch": 0.07841441441441442,
"grad_norm": 102.37490844726562,
"learning_rate": 9.611448774886925e-06,
"loss": 10.019007873535156,
"loss_idm": 0.5578125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0240966796875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5720710635185242,
"loss_sft": 0.01458285003900528,
"step": 340,
"token_acc": 0.8438792841749271
},
{
"epoch": 0.08072072072072072,
"grad_norm": 102.40097045898438,
"learning_rate": 9.578866633275289e-06,
"loss": 9.723292541503906,
"loss_idm": 0.54921875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.02216796875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.555128300189972,
"loss_sft": 0.01469133198261261,
"step": 350,
"token_acc": 0.8427996218283734
},
{
"epoch": 0.08302702702702702,
"grad_norm": 114.85263061523438,
"learning_rate": 9.545032675245814e-06,
"loss": 9.351869201660156,
"loss_idm": 0.52578125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.02275390625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5321333289146424,
"loss_sft": 0.014326457306742668,
"step": 360,
"token_acc": 0.8459759580797804
},
{
"epoch": 0.08533333333333333,
"grad_norm": 111.28559112548828,
"learning_rate": 9.509956150664796e-06,
"loss": 9.204309844970703,
"loss_idm": 0.51796875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0224609375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5225722551345825,
"loss_sft": 0.014612309634685516,
"step": 370,
"token_acc": 0.8429501243235337
},
{
"epoch": 0.08763963963963964,
"grad_norm": 159.0714111328125,
"learning_rate": 9.473646649103819e-06,
"loss": 10.069181823730469,
"loss_idm": 0.55625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0218017578125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.577532160282135,
"loss_sft": 0.014597192034125328,
"step": 380,
"token_acc": 0.8435810512784968
},
{
"epoch": 0.08994594594594595,
"grad_norm": 109.76734161376953,
"learning_rate": 9.43611409721806e-06,
"loss": 9.046263885498046,
"loss_idm": 0.51171875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.024365234375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5109918117523193,
"loss_sft": 0.01441824659705162,
"step": 390,
"token_acc": 0.8451461865571329
},
{
"epoch": 0.09225225225225225,
"grad_norm": 114.03449249267578,
"learning_rate": 9.397368756032445e-06,
"loss": 8.885872650146485,
"loss_idm": 0.49765625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0222412109375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5030822515487671,
"loss_sft": 0.014403243735432624,
"step": 400,
"token_acc": 0.8454346932611032
},
{
"epoch": 0.09455855855855856,
"grad_norm": 95.63912200927734,
"learning_rate": 9.357421218136387e-06,
"loss": 9.656077575683593,
"loss_idm": 0.53828125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.02294921875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5517061471939086,
"loss_sft": 0.013985250145196915,
"step": 410,
"token_acc": 0.8492683217596042
},
{
"epoch": 0.09686486486486487,
"grad_norm": 104.48960876464844,
"learning_rate": 9.31628240478787e-06,
"loss": 9.242665100097657,
"loss_idm": 0.51328125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0219482421875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.525335168838501,
"loss_sft": 0.014777269214391708,
"step": 420,
"token_acc": 0.8418984633339817
},
{
"epoch": 0.09917117117117118,
"grad_norm": 147.1559600830078,
"learning_rate": 9.273963562927695e-06,
"loss": 9.189979553222656,
"loss_idm": 0.5125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.02255859375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5218879580497742,
"loss_sft": 0.014679400622844696,
"step": 430,
"token_acc": 0.8437356123388582
},
{
"epoch": 0.10147747747747747,
"grad_norm": 93.56449127197266,
"learning_rate": 9.230476262104678e-06,
"loss": 9.361734008789062,
"loss_idm": 0.52421875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0218505859375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5325615644454956,
"loss_sft": 0.014835886657238007,
"step": 440,
"token_acc": 0.8411223489145692
},
{
"epoch": 0.10378378378378378,
"grad_norm": 82.03609466552734,
"learning_rate": 9.185832391312644e-06,
"loss": 8.240397644042968,
"loss_idm": 0.45546875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0206298828125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.46435696482658384,
"loss_sft": 0.014625262469053268,
"step": 450,
"token_acc": 0.8430563318063227
},
{
"epoch": 0.10609009009009009,
"grad_norm": 101.49946594238281,
"learning_rate": 9.140044155740102e-06,
"loss": 8.768144226074218,
"loss_idm": 0.48203125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0253173828125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.49250184297561644,
"loss_sft": 0.014648660272359847,
"step": 460,
"token_acc": 0.8423310494803166
},
{
"epoch": 0.1083963963963964,
"grad_norm": 119.373291015625,
"learning_rate": 9.093124073433464e-06,
"loss": 9.084942626953126,
"loss_idm": 0.509375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.02119140625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5161856293678284,
"loss_sft": 0.01479191929101944,
"step": 470,
"token_acc": 0.842016730205665
},
{
"epoch": 0.1107027027027027,
"grad_norm": 110.53947448730469,
"learning_rate": 9.045084971874738e-06,
"loss": 8.959126281738282,
"loss_idm": 0.5015625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0206787109375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5086698532104492,
"loss_sft": 0.014888135716319084,
"step": 480,
"token_acc": 0.8410448617813955
},
{
"epoch": 0.11300900900900901,
"grad_norm": 88.150146484375,
"learning_rate": 8.995939984474624e-06,
"loss": 9.11611557006836,
"loss_idm": 0.51484375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.021240234375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.517701530456543,
"loss_sft": 0.014868151023983955,
"step": 490,
"token_acc": 0.8409646758677575
},
{
"epoch": 0.11531531531531532,
"grad_norm": 105.73524475097656,
"learning_rate": 8.94570254698197e-06,
"loss": 8.90081558227539,
"loss_idm": 0.49921875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0226318359375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5032265663146973,
"loss_sft": 0.01476171426475048,
"step": 500,
"token_acc": 0.8424521747539707
},
{
"epoch": 0.11531531531531532,
"eval_loss": 0.873286783695221,
"eval_runtime": 169.8257,
"eval_samples_per_second": 10.275,
"eval_steps_per_second": 2.573,
"eval_token_acc": 0.8295710593091115,
"step": 500
},
{
"epoch": 0.11762162162162162,
"grad_norm": 79.52201843261719,
"learning_rate": 8.894386393810563e-06,
"loss": 8.980358123779297,
"loss_idm": 0.501171875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0220947265625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5085306227207184,
"loss_sft": 0.014769895002245903,
"step": 510,
"token_acc": 0.8416994573590164
},
{
"epoch": 0.11992792792792793,
"grad_norm": 87.73248291015625,
"learning_rate": 8.842005554284296e-06,
"loss": 8.839938354492187,
"loss_idm": 0.4921875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0215576171875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5001400589942933,
"loss_sft": 0.014775922894477845,
"step": 520,
"token_acc": 0.8430164098119748
},
{
"epoch": 0.12223423423423424,
"grad_norm": 108.73457336425781,
"learning_rate": 8.788574348801676e-06,
"loss": 9.055592346191407,
"loss_idm": 0.50859375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.021142578125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5146129488945007,
"loss_sft": 0.014646383747458459,
"step": 530,
"token_acc": 0.843842382599978
},
{
"epoch": 0.12454054054054053,
"grad_norm": 98.13040924072266,
"learning_rate": 8.734107384920771e-06,
"loss": 8.952853393554687,
"loss_idm": 0.50625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0204833984375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5087056159973145,
"loss_sft": 0.014654205366969108,
"step": 540,
"token_acc": 0.8428865032158228
},
{
"epoch": 0.12684684684684686,
"grad_norm": 120.66363525390625,
"learning_rate": 8.67861955336566e-06,
"loss": 8.789979553222656,
"loss_idm": 0.48984375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.02001953125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.49941076040267945,
"loss_sft": 0.014644744619727135,
"step": 550,
"token_acc": 0.8429765894640101
},
{
"epoch": 0.12915315315315315,
"grad_norm": 105.6294174194336,
"learning_rate": 8.622126023955446e-06,
"loss": 8.306752014160157,
"loss_idm": 0.47109375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0207763671875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4683968424797058,
"loss_sft": 0.014503800868988037,
"step": 560,
"token_acc": 0.844274669579772
},
{
"epoch": 0.13145945945945947,
"grad_norm": 87.99810028076172,
"learning_rate": 8.564642241456986e-06,
"loss": 9.134395599365234,
"loss_idm": 0.5078125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0187744140625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5218061327934265,
"loss_sft": 0.014652644842863083,
"step": 570,
"token_acc": 0.8432568832791741
},
{
"epoch": 0.13376576576576577,
"grad_norm": 94.22286224365234,
"learning_rate": 8.506183921362443e-06,
"loss": 9.002159118652344,
"loss_idm": 0.50390625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0199462890625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5122206807136536,
"loss_sft": 0.01476593129336834,
"step": 580,
"token_acc": 0.8415455047829566
},
{
"epoch": 0.13607207207207206,
"grad_norm": 96.09319305419922,
"learning_rate": 8.446767045592829e-06,
"loss": 8.899197387695313,
"loss_idm": 0.48984375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.020263671875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5060975551605225,
"loss_sft": 0.014485136792063712,
"step": 590,
"token_acc": 0.8453194613833785
},
{
"epoch": 0.13837837837837838,
"grad_norm": 91.08513641357422,
"learning_rate": 8.386407858128707e-06,
"loss": 8.522463989257812,
"loss_idm": 0.47265625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0227294921875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4799580335617065,
"loss_sft": 0.01448415480554104,
"step": 600,
"token_acc": 0.8451588424854661
},
{
"epoch": 0.14068468468468467,
"grad_norm": 91.31858825683594,
"learning_rate": 8.325122860569241e-06,
"loss": 8.168463897705077,
"loss_idm": 0.4453125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0220458984375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.45817803144454955,
"loss_sft": 0.01455080546438694,
"step": 610,
"token_acc": 0.8434149712322799
},
{
"epoch": 0.142990990990991,
"grad_norm": 92.19358825683594,
"learning_rate": 8.262928807620843e-06,
"loss": 9.086221313476562,
"loss_idm": 0.5046875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0214111328125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5163055658340454,
"loss_sft": 0.014690371602773667,
"step": 620,
"token_acc": 0.8422049082316321
},
{
"epoch": 0.1452972972972973,
"grad_norm": 85.43914794921875,
"learning_rate": 8.199842702516584e-06,
"loss": 8.663780212402344,
"loss_idm": 0.47265625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0191650390625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.49150881767272947,
"loss_sft": 0.01495927982032299,
"step": 630,
"token_acc": 0.8403844066897375
},
{
"epoch": 0.1476036036036036,
"grad_norm": 100.33062744140625,
"learning_rate": 8.135881792367686e-06,
"loss": 8.27200927734375,
"loss_idm": 0.4625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0180419921875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.46880000829696655,
"loss_sft": 0.014622250199317932,
"step": 640,
"token_acc": 0.8430498513734264
},
{
"epoch": 0.1499099099099099,
"grad_norm": 102.80355834960938,
"learning_rate": 8.071063563448341e-06,
"loss": 8.619700622558593,
"loss_idm": 0.4796875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0202880859375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4885416507720947,
"loss_sft": 0.014433516189455986,
"step": 650,
"token_acc": 0.845255299241181
},
{
"epoch": 0.15221621621621623,
"grad_norm": 88.60161590576172,
"learning_rate": 8.005405736415127e-06,
"loss": 8.179424285888672,
"loss_idm": 0.460546875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0212158203125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4592796742916107,
"loss_sft": 0.014803672954440118,
"step": 660,
"token_acc": 0.8411518466838719
},
{
"epoch": 0.15452252252252252,
"grad_norm": 96.72508239746094,
"learning_rate": 7.938926261462366e-06,
"loss": 8.372673797607423,
"loss_idm": 0.45859375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0193115234375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4736079216003418,
"loss_sft": 0.014764882624149323,
"step": 670,
"token_acc": 0.8415420786843315
},
{
"epoch": 0.15682882882882884,
"grad_norm": 109.25978088378906,
"learning_rate": 7.871643313414718e-06,
"loss": 8.376298522949218,
"loss_idm": 0.46875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0222412109375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.47051931619644166,
"loss_sft": 0.014839059859514236,
"step": 680,
"token_acc": 0.8409984786981344
},
{
"epoch": 0.15913513513513514,
"grad_norm": 103.52886962890625,
"learning_rate": 7.803575286758365e-06,
"loss": 7.899095916748047,
"loss_idm": 0.44375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0193359375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.44420482516288756,
"loss_sft": 0.014591414108872413,
"step": 690,
"token_acc": 0.8434285377428007
},
{
"epoch": 0.16144144144144143,
"grad_norm": 116.31279754638672,
"learning_rate": 7.734740790612137e-06,
"loss": 7.9990592956542965,
"loss_idm": 0.44296875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.01865234375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4515872120857239,
"loss_sft": 0.014406795054674149,
"step": 700,
"token_acc": 0.8455261067379216
},
{
"epoch": 0.16374774774774775,
"grad_norm": 104.14566802978516,
"learning_rate": 7.66515864363997e-06,
"loss": 8.241651916503907,
"loss_idm": 0.4578125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0193603515625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.46591639518737793,
"loss_sft": 0.014483961462974548,
"step": 710,
"token_acc": 0.8448154530557234
},
{
"epoch": 0.16605405405405405,
"grad_norm": 104.71328735351562,
"learning_rate": 7.594847868906076e-06,
"loss": 8.089508819580079,
"loss_idm": 0.45234375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0209228515625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4547982931137085,
"loss_sft": 0.01439845934510231,
"step": 720,
"token_acc": 0.8451516312280465
},
{
"epoch": 0.16836036036036037,
"grad_norm": 107.36788177490234,
"learning_rate": 7.52382768867422e-06,
"loss": 8.494679260253907,
"loss_idm": 0.471875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0205322265625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.48007341623306277,
"loss_sft": 0.014671024680137635,
"step": 730,
"token_acc": 0.8430621123298506
},
{
"epoch": 0.17066666666666666,
"grad_norm": 84.9700927734375,
"learning_rate": 7.452117519152542e-06,
"loss": 7.954332733154297,
"loss_idm": 0.4421875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.020751953125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4460784077644348,
"loss_sft": 0.014631084725260734,
"step": 740,
"token_acc": 0.84308325249699
},
{
"epoch": 0.17297297297297298,
"grad_norm": 126.57601928710938,
"learning_rate": 7.379736965185369e-06,
"loss": 8.158969116210937,
"loss_idm": 0.45859375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0198974609375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.459906268119812,
"loss_sft": 0.014672993496060372,
"step": 750,
"token_acc": 0.8428532139788794
},
{
"epoch": 0.17527927927927928,
"grad_norm": 92.83258819580078,
"learning_rate": 7.30670581489344e-06,
"loss": 8.280567932128907,
"loss_idm": 0.46171875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.019189453125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.46855597496032714,
"loss_sft": 0.014394653588533401,
"step": 760,
"token_acc": 0.8446953594377756
},
{
"epoch": 0.1775855855855856,
"grad_norm": 111.44788360595703,
"learning_rate": 7.233044034264034e-06,
"loss": 7.862223815917969,
"loss_idm": 0.4359375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0191162109375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4427155554294586,
"loss_sft": 0.014474740624427796,
"step": 770,
"token_acc": 0.8450180938562969
},
{
"epoch": 0.1798918918918919,
"grad_norm": 115.48260498046875,
"learning_rate": 7.158771761692464e-06,
"loss": 7.9805961608886715,
"loss_idm": 0.44453125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0202392578125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.44799396991729734,
"loss_sft": 0.01469801552593708,
"step": 780,
"token_acc": 0.8424233709705329
},
{
"epoch": 0.18219819819819819,
"grad_norm": 94.31640625,
"learning_rate": 7.083909302476453e-06,
"loss": 9.633565521240234,
"loss_idm": 0.5421875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0183837890625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5536909341812134,
"loss_sft": 0.014749213308095931,
"step": 790,
"token_acc": 0.8422723821357079
},
{
"epoch": 0.1845045045045045,
"grad_norm": 96.44784545898438,
"learning_rate": 7.008477123264849e-06,
"loss": 7.915254974365235,
"loss_idm": 0.4421875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0171142578125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4477837860584259,
"loss_sft": 0.014466813579201699,
"step": 800,
"token_acc": 0.8447225350327021
},
{
"epoch": 0.1868108108108108,
"grad_norm": 84.42618560791016,
"learning_rate": 6.932495846462262e-06,
"loss": 8.682425689697265,
"loss_idm": 0.4859375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.017529296875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.49512912034988404,
"loss_sft": 0.014556843042373657,
"step": 810,
"token_acc": 0.8437006309304983
},
{
"epoch": 0.18911711711711712,
"grad_norm": 114.39678192138672,
"learning_rate": 6.855986244591104e-06,
"loss": 7.2060691833496096,
"loss_idm": 0.394921875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.019873046875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.40005741119384763,
"loss_sft": 0.014659629762172699,
"step": 820,
"token_acc": 0.8422255628738242
},
{
"epoch": 0.19142342342342342,
"grad_norm": 125.64624786376953,
"learning_rate": 6.778969234612583e-06,
"loss": 8.314984130859376,
"loss_idm": 0.465234375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.018212890625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4713524401187897,
"loss_sft": 0.014716658741235733,
"step": 830,
"token_acc": 0.8419139432862319
},
{
"epoch": 0.19372972972972974,
"grad_norm": 82.18567657470703,
"learning_rate": 6.701465872208216e-06,
"loss": 8.177500152587891,
"loss_idm": 0.46015625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.018994140625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.46165603399276733,
"loss_sft": 0.014731941372156143,
"step": 840,
"token_acc": 0.8417456898652139
},
{
"epoch": 0.19603603603603603,
"grad_norm": 76.3083724975586,
"learning_rate": 6.6234973460234184e-06,
"loss": 6.866006469726562,
"loss_idm": 0.375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0199951171875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.37900240421295167,
"loss_sft": 0.014587616920471192,
"step": 850,
"token_acc": 0.8433828349109465
},
{
"epoch": 0.19834234234234235,
"grad_norm": 87.86666107177734,
"learning_rate": 6.545084971874738e-06,
"loss": 9.54729766845703,
"loss_idm": 0.52734375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.020166015625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.5463018894195557,
"loss_sft": 0.014688147231936455,
"step": 860,
"token_acc": 0.843223021860908
},
{
"epoch": 0.20064864864864865,
"grad_norm": 79.54348754882812,
"learning_rate": 6.466250186922325e-06,
"loss": 7.474161529541016,
"loss_idm": 0.41015625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0190673828125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4184339165687561,
"loss_sft": 0.014442858472466468,
"step": 870,
"token_acc": 0.8446895228961879
},
{
"epoch": 0.20295495495495494,
"grad_norm": 101.24223327636719,
"learning_rate": 6.387014543809224e-06,
"loss": 7.865777587890625,
"loss_idm": 0.435546875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.02099609375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4404805421829224,
"loss_sft": 0.014567815512418748,
"step": 880,
"token_acc": 0.8441662849959559
},
{
"epoch": 0.20526126126126126,
"grad_norm": 113.0296401977539,
"learning_rate": 6.3073997047691e-06,
"loss": 7.874870300292969,
"loss_idm": 0.440625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0199462890625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4417296290397644,
"loss_sft": 0.014670121297240258,
"step": 890,
"token_acc": 0.8433289124014535
},
{
"epoch": 0.20756756756756756,
"grad_norm": 103.46904754638672,
"learning_rate": 6.227427435703997e-06,
"loss": 8.081219482421876,
"loss_idm": 0.444140625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0183837890625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.45728793144226076,
"loss_sft": 0.014389193430542947,
"step": 900,
"token_acc": 0.8454201846502517
},
{
"epoch": 0.20987387387387388,
"grad_norm": 88.96587371826172,
"learning_rate": 6.147119600233758e-06,
"loss": 8.543597412109374,
"loss_idm": 0.4796875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.019921875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4841804504394531,
"loss_sft": 0.014478649944067001,
"step": 910,
"token_acc": 0.8443082839472513
},
{
"epoch": 0.21218018018018017,
"grad_norm": 117.9122085571289,
"learning_rate": 6.066498153718735e-06,
"loss": 7.415792083740234,
"loss_idm": 0.4140625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0222412109375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4112999439239502,
"loss_sft": 0.014585891366004944,
"step": 920,
"token_acc": 0.8436708029516633
},
{
"epoch": 0.2144864864864865,
"grad_norm": 86.82301330566406,
"learning_rate": 5.985585137257401e-06,
"loss": 7.67747802734375,
"loss_idm": 0.427734375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0196044921875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4302915811538696,
"loss_sft": 0.014485430717468262,
"step": 930,
"token_acc": 0.8436629761319727
},
{
"epoch": 0.2167927927927928,
"grad_norm": 95.94313049316406,
"learning_rate": 5.904402671660551e-06,
"loss": 7.310446166992188,
"loss_idm": 0.4046875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0189697265625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.40787339210510254,
"loss_sft": 0.014458712935447694,
"step": 940,
"token_acc": 0.845064712919843
},
{
"epoch": 0.2190990990990991,
"grad_norm": 90.85774230957031,
"learning_rate": 5.82297295140367e-06,
"loss": 7.37610092163086,
"loss_idm": 0.41015625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.019921875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.41144019961357114,
"loss_sft": 0.014352937042713166,
"step": 950,
"token_acc": 0.8450359871010459
},
{
"epoch": 0.2214054054054054,
"grad_norm": 91.5264892578125,
"learning_rate": 5.74131823855921e-06,
"loss": 7.979737854003906,
"loss_idm": 0.439453125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0195068359375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.450022554397583,
"loss_sft": 0.014237866923213004,
"step": 960,
"token_acc": 0.8463489009812982
},
{
"epoch": 0.22371171171171172,
"grad_norm": 89.29737854003906,
"learning_rate": 5.659460856710346e-06,
"loss": 7.60855712890625,
"loss_idm": 0.41953125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0177490234375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4289973974227905,
"loss_sft": 0.014107778668403625,
"step": 970,
"token_acc": 0.8469193499209564
},
{
"epoch": 0.22601801801801802,
"grad_norm": 106.5138168334961,
"learning_rate": 5.577423184847932e-06,
"loss": 8.291693115234375,
"loss_idm": 0.46171875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0203857421875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4682368040084839,
"loss_sft": 0.014279923215508462,
"step": 980,
"token_acc": 0.8464010922588995
},
{
"epoch": 0.2283243243243243,
"grad_norm": 117.88599395751953,
"learning_rate": 5.495227651252315e-06,
"loss": 7.344970703125,
"loss_idm": 0.402734375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0205810546875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.40857953429222105,
"loss_sft": 0.014561992883682252,
"step": 990,
"token_acc": 0.8436562465051372
},
{
"epoch": 0.23063063063063063,
"grad_norm": 107.79817199707031,
"learning_rate": 5.412896727361663e-06,
"loss": 7.102320861816406,
"loss_idm": 0.397265625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0206787109375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.3930091679096222,
"loss_sft": 0.014505467936396598,
"step": 1000,
"token_acc": 0.8436447490694896
},
{
"epoch": 0.23063063063063063,
"eval_loss": 0.826145350933075,
"eval_runtime": 168.0761,
"eval_samples_per_second": 10.382,
"eval_steps_per_second": 2.6,
"eval_token_acc": 0.8308116622599108,
"step": 1000
},
{
"epoch": 0.23293693693693693,
"grad_norm": 88.33614349365234,
"learning_rate": 5.3304529216284974e-06,
"loss": 7.70971450805664,
"loss_idm": 0.42890625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0201171875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4321062684059143,
"loss_sft": 0.014323608577251434,
"step": 1010,
"token_acc": 0.8463429671668039
},
{
"epoch": 0.23524324324324325,
"grad_norm": 103.5328369140625,
"learning_rate": 5.247918773366112e-06,
"loss": 7.400896453857422,
"loss_idm": 0.411328125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.018603515625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.41432325839996337,
"loss_sft": 0.014487444236874581,
"step": 1020,
"token_acc": 0.8436925959265521
},
{
"epoch": 0.23754954954954954,
"grad_norm": 95.37596893310547,
"learning_rate": 5.165316846586541e-06,
"loss": 7.066916656494141,
"loss_idm": 0.398828125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.01865234375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.3934369027614594,
"loss_sft": 0.01427248790860176,
"step": 1030,
"token_acc": 0.8449177570364499
},
{
"epoch": 0.23985585585585587,
"grad_norm": 106.63520050048828,
"learning_rate": 5.082669723831793e-06,
"loss": 7.509798431396485,
"loss_idm": 0.41484375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0177490234375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.42152470350265503,
"loss_sft": 0.014691157266497612,
"step": 1040,
"token_acc": 0.8420689476522522
},
{
"epoch": 0.24216216216216216,
"grad_norm": 110.2610855102539,
"learning_rate": 5e-06,
"loss": 6.679129028320313,
"loss_idm": 0.367578125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0177490234375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.3698115944862366,
"loss_sft": 0.01454925611615181,
"step": 1050,
"token_acc": 0.8429293549000219
},
{
"epoch": 0.24446846846846848,
"grad_norm": 116.46825408935547,
"learning_rate": 4.917330276168208e-06,
"loss": 7.56883544921875,
"loss_idm": 0.426171875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0185791015625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4245879828929901,
"loss_sft": 0.014570457488298416,
"step": 1060,
"token_acc": 0.843766214243804
},
{
"epoch": 0.24677477477477477,
"grad_norm": 90.62398529052734,
"learning_rate": 4.8346831534134595e-06,
"loss": 8.461653137207032,
"loss_idm": 0.48203125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.019677734375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4794220805168152,
"loss_sft": 0.014480674639344215,
"step": 1070,
"token_acc": 0.8445398105746493
},
{
"epoch": 0.24908108108108107,
"grad_norm": 97.12873840332031,
"learning_rate": 4.752081226633888e-06,
"loss": 7.096697998046875,
"loss_idm": 0.391015625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.017236328125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.39673874378204343,
"loss_sft": 0.014352129772305489,
"step": 1080,
"token_acc": 0.8445942044328101
},
{
"epoch": 0.25138738738738736,
"grad_norm": 148.5345458984375,
"learning_rate": 4.669547078371503e-06,
"loss": 8.134896850585937,
"loss_idm": 0.453125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.01767578125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4617579817771912,
"loss_sft": 0.01423180066049099,
"step": 1090,
"token_acc": 0.8462392484597084
},
{
"epoch": 0.2536936936936937,
"grad_norm": 99.68243408203125,
"learning_rate": 4.587103272638339e-06,
"loss": 6.537295532226563,
"loss_idm": 0.363671875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.01689453125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.3626400887966156,
"loss_sft": 0.014130813628435135,
"step": 1100,
"token_acc": 0.847496286866115
},
{
"epoch": 0.256,
"grad_norm": 106.85647583007812,
"learning_rate": 4.504772348747687e-06,
"loss": 7.332121276855469,
"loss_idm": 0.40234375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.017724609375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.41151710152626036,
"loss_sft": 0.014223045855760574,
"step": 1110,
"token_acc": 0.8462705823615773
},
{
"epoch": 0.2583063063063063,
"grad_norm": 107.44613647460938,
"learning_rate": 4.42257681515207e-06,
"loss": 6.910240173339844,
"loss_idm": 0.381640625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.018505859375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.3840563654899597,
"loss_sft": 0.014243298768997192,
"step": 1120,
"token_acc": 0.8465594450949473
},
{
"epoch": 0.2606126126126126,
"grad_norm": 107.26278686523438,
"learning_rate": 4.340539143289655e-06,
"loss": 7.2168525695800785,
"loss_idm": 0.403125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0174560546875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.40406553745269774,
"loss_sft": 0.014284095168113709,
"step": 1130,
"token_acc": 0.8454709336305773
},
{
"epoch": 0.26291891891891894,
"grad_norm": 116.3701400756836,
"learning_rate": 4.25868176144079e-06,
"loss": 7.558566284179688,
"loss_idm": 0.419921875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0177978515625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4247333943843842,
"loss_sft": 0.014513048902153968,
"step": 1140,
"token_acc": 0.8439526822829935
},
{
"epoch": 0.26522522522522524,
"grad_norm": 106.37321472167969,
"learning_rate": 4.17702704859633e-06,
"loss": 7.477565765380859,
"loss_idm": 0.41640625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0197021484375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4179769098758698,
"loss_sft": 0.01451987624168396,
"step": 1150,
"token_acc": 0.8434178920433081
},
{
"epoch": 0.26753153153153153,
"grad_norm": 115.76299285888672,
"learning_rate": 4.0955973283394525e-06,
"loss": 7.254231262207031,
"loss_idm": 0.402734375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0183837890625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.405047333240509,
"loss_sft": 0.014345861971378326,
"step": 1160,
"token_acc": 0.8443389144238174
},
{
"epoch": 0.2698378378378378,
"grad_norm": 158.42250061035156,
"learning_rate": 4.0144148627426e-06,
"loss": 7.521492004394531,
"loss_idm": 0.41796875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0178955078125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4225184381008148,
"loss_sft": 0.01449190303683281,
"step": 1170,
"token_acc": 0.8435488968221455
},
{
"epoch": 0.2721441441441441,
"grad_norm": 103.64552307128906,
"learning_rate": 3.9335018462812664e-06,
"loss": 8.284562683105468,
"loss_idm": 0.45546875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0184814453125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.46931086778640746,
"loss_sft": 0.014623204618692398,
"step": 1180,
"token_acc": 0.8422325811350072
},
{
"epoch": 0.27445045045045047,
"grad_norm": 90.13056945800781,
"learning_rate": 3.852880399766243e-06,
"loss": 7.880992126464844,
"loss_idm": 0.440234375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0189208984375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.44379000663757323,
"loss_sft": 0.014453979581594468,
"step": 1190,
"token_acc": 0.8441096901131333
},
{
"epoch": 0.27675675675675676,
"grad_norm": 104.42198944091797,
"learning_rate": 3.7725725642960047e-06,
"loss": 6.547203826904297,
"loss_idm": 0.3640625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.01826171875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.36079212427139284,
"loss_sft": 0.014533436298370362,
"step": 1200,
"token_acc": 0.8435185033768867
},
{
"epoch": 0.27906306306306305,
"grad_norm": 109.83740997314453,
"learning_rate": 3.6926002952309015e-06,
"loss": 7.021842956542969,
"loss_idm": 0.386328125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.01845703125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.39095706343650816,
"loss_sft": 0.014352500438690186,
"step": 1210,
"token_acc": 0.8457043054504614
},
{
"epoch": 0.28136936936936935,
"grad_norm": 92.76325988769531,
"learning_rate": 3.6129854561907786e-06,
"loss": 6.612565612792968,
"loss_idm": 0.363671875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0177490234375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.3655923068523407,
"loss_sft": 0.014593320712447167,
"step": 1220,
"token_acc": 0.8422301537496077
},
{
"epoch": 0.2836756756756757,
"grad_norm": 108.94733428955078,
"learning_rate": 3.533749813077677e-06,
"loss": 6.7885498046875,
"loss_idm": 0.3734375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.01787109375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.3769371509552002,
"loss_sft": 0.014226069301366806,
"step": 1230,
"token_acc": 0.8456893225753114
},
{
"epoch": 0.285981981981982,
"grad_norm": 130.9772186279297,
"learning_rate": 3.4549150281252635e-06,
"loss": 6.741462707519531,
"loss_idm": 0.37109375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0176513671875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.37425958514213564,
"loss_sft": 0.014150273799896241,
"step": 1240,
"token_acc": 0.8465476829923336
},
{
"epoch": 0.2882882882882883,
"grad_norm": 119.88499450683594,
"learning_rate": 3.3765026539765832e-06,
"loss": 7.194857788085938,
"loss_idm": 0.400390625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.018896484375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.4020317316055298,
"loss_sft": 0.013935164362192155,
"step": 1250,
"token_acc": 0.8496373374216677
},
{
"epoch": 0.2905945945945946,
"grad_norm": 107.04531860351562,
"learning_rate": 3.298534127791785e-06,
"loss": 7.381626892089844,
"loss_idm": 0.409765625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.01904296875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.413124680519104,
"loss_sft": 0.014187094569206238,
"step": 1260,
"token_acc": 0.8464285982464
},
{
"epoch": 0.2929009009009009,
"grad_norm": 133.18710327148438,
"learning_rate": 3.2210307653874175e-06,
"loss": 6.52490234375,
"loss_idm": 0.360546875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.016943359375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.36171150803565977,
"loss_sft": 0.01412605307996273,
"step": 1270,
"token_acc": 0.8482812389612749
},
{
"epoch": 0.2952072072072072,
"grad_norm": 133.87335205078125,
"learning_rate": 3.1440137554088957e-06,
"loss": 7.00994873046875,
"loss_idm": 0.389453125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0179931640625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.39080272912979125,
"loss_sft": 0.014256376028060912,
"step": 1280,
"token_acc": 0.8455885394740364
},
{
"epoch": 0.2975135135135135,
"grad_norm": 118.04527282714844,
"learning_rate": 3.06750415353774e-06,
"loss": 6.74976806640625,
"loss_idm": 0.37421875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.017529296875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.3751899302005768,
"loss_sft": 0.014237936958670616,
"step": 1290,
"token_acc": 0.8457528871889156
},
{
"epoch": 0.2998198198198198,
"grad_norm": 152.24542236328125,
"learning_rate": 2.991522876735154e-06,
"loss": 6.73217544555664,
"loss_idm": 0.375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0180419921875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.373495876789093,
"loss_sft": 0.014310174062848091,
"step": 1300,
"token_acc": 0.8459135928839514
},
{
"epoch": 0.3021261261261261,
"grad_norm": 143.8121337890625,
"learning_rate": 2.9160906975235493e-06,
"loss": 6.565863037109375,
"loss_idm": 0.365234375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.019970703125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.36084928512573244,
"loss_sft": 0.014364153146743774,
"step": 1310,
"token_acc": 0.8448077517079684
},
{
"epoch": 0.30443243243243245,
"grad_norm": 115.51368713378906,
"learning_rate": 2.8412282383075362e-06,
"loss": 6.6176704406738285,
"loss_idm": 0.365234375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.015966796875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.36805868744850156,
"loss_sft": 0.01430610753595829,
"step": 1320,
"token_acc": 0.8448409814290369
},
{
"epoch": 0.30673873873873875,
"grad_norm": 103.80223846435547,
"learning_rate": 2.766955965735968e-06,
"loss": 6.629913330078125,
"loss_idm": 0.36796875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.018212890625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.3660490334033966,
"loss_sft": 0.014540331438183784,
"step": 1330,
"token_acc": 0.8437379067938862
},
{
"epoch": 0.30904504504504504,
"grad_norm": 123.66497039794922,
"learning_rate": 2.693294185106562e-06,
"loss": 6.788928985595703,
"loss_idm": 0.38046875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.01904296875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.37561420202255247,
"loss_sft": 0.014337588101625443,
"step": 1340,
"token_acc": 0.8445305506848307
},
{
"epoch": 0.31135135135135134,
"grad_norm": 90.85652160644531,
"learning_rate": 2.6202630348146323e-06,
"loss": 6.739347839355469,
"loss_idm": 0.371484375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.016357421875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.3759130001068115,
"loss_sft": 0.014109160378575325,
"step": 1350,
"token_acc": 0.846979458728549
},
{
"epoch": 0.3136576576576577,
"grad_norm": 120.79927825927734,
"learning_rate": 2.5478824808474613e-06,
"loss": 6.946615600585938,
"loss_idm": 0.388671875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.01796875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.38742896914482117,
"loss_sft": 0.014025571942329406,
"step": 1360,
"token_acc": 0.8474245017667137
},
{
"epoch": 0.315963963963964,
"grad_norm": 127.58772277832031,
"learning_rate": 2.476172311325783e-06,
"loss": 6.843831634521484,
"loss_idm": 0.379296875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0177978515625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.38061107993125914,
"loss_sft": 0.014309322834014893,
"step": 1370,
"token_acc": 0.8454213580875642
},
{
"epoch": 0.3182702702702703,
"grad_norm": 151.91925048828125,
"learning_rate": 2.4051521310939258e-06,
"loss": 6.953953552246094,
"loss_idm": 0.3859375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.016845703125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.38837702870368956,
"loss_sft": 0.014433859288692475,
"step": 1380,
"token_acc": 0.8446607330752671
},
{
"epoch": 0.32057657657657657,
"grad_norm": 142.9082489013672,
"learning_rate": 2.3348413563600324e-06,
"loss": 7.533473205566406,
"loss_idm": 0.41640625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.01787109375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.42370715737342834,
"loss_sft": 0.01422720868140459,
"step": 1390,
"token_acc": 0.8468790057744192
},
{
"epoch": 0.32288288288288286,
"grad_norm": 87.2136001586914,
"learning_rate": 2.265259209387867e-06,
"loss": 6.789948272705078,
"loss_idm": 0.373046875,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0173095703125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.3781958520412445,
"loss_sft": 0.014054833725094796,
"step": 1400,
"token_acc": 0.8476745137822668
},
{
"epoch": 0.3251891891891892,
"grad_norm": 109.24415588378906,
"learning_rate": 2.1964247132416373e-06,
"loss": 6.430654144287109,
"loss_idm": 0.356640625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0153076171875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.3571660578250885,
"loss_sft": 0.01431470774114132,
"step": 1410,
"token_acc": 0.8448665901524277
},
{
"epoch": 0.3274954954954955,
"grad_norm": 116.79399871826172,
"learning_rate": 2.1283566865852824e-06,
"loss": 6.299882507324218,
"loss_idm": 0.344140625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.017724609375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.34696500301361083,
"loss_sft": 0.014206652343273164,
"step": 1420,
"token_acc": 0.846456135209066
},
{
"epoch": 0.3298018018018018,
"grad_norm": 149.03395080566406,
"learning_rate": 2.061073738537635e-06,
"loss": 6.599853515625,
"loss_idm": 0.36328125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0202392578125,
"loss_idm_text": 0.0,
"loss_idm_type": 0.36307021975517273,
"loss_sft": 0.014167750626802445,
"step": 1430,
"token_acc": 0.8466417422677717
},
{
"epoch": 0.3321081081081081,
"grad_norm": 132.9679718017578,
"learning_rate": 1.9945942635848745e-06,
"loss": 6.743374633789062,
"loss_idm": 0.374609375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0191162109375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.3731849730014801,
"loss_sft": 0.014212041348218917,
"step": 1440,
"token_acc": 0.8470092899704256
},
{
"epoch": 0.33441441441441444,
"grad_norm": 119.5426254272461,
"learning_rate": 1.928936436551661e-06,
"loss": 6.725025939941406,
"loss_idm": 0.37265625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.017724609375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.3735596239566803,
"loss_sft": 0.014224140346050263,
"step": 1450,
"token_acc": 0.8464133441783575
},
{
"epoch": 0.33672072072072073,
"grad_norm": 139.33950805664062,
"learning_rate": 1.864118207632315e-06,
"loss": 6.592694091796875,
"loss_idm": 0.36484375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0165771484375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.36636746525764463,
"loss_sft": 0.014215507730841636,
"step": 1460,
"token_acc": 0.8457222099999259
},
{
"epoch": 0.33902702702702703,
"grad_norm": 126.1017837524414,
"learning_rate": 1.8001572974834169e-06,
"loss": 6.243590545654297,
"loss_idm": 0.34140625,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0188232421875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.3425869166851044,
"loss_sft": 0.014009960740804673,
"step": 1470,
"token_acc": 0.8481653395222281
},
{
"epoch": 0.3413333333333333,
"grad_norm": 143.65353393554688,
"learning_rate": 1.7370711923791567e-06,
"loss": 6.1479026794433596,
"loss_idm": 0.3359375,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0189697265625,
"loss_idm_text": 0.0,
"loss_idm_type": 0.3357097864151001,
"loss_sft": 0.014131288602948188,
"step": 1480,
"token_acc": 0.8479393838271811
},
{
"epoch": 0.3436396396396396,
"grad_norm": 121.66058349609375,
"learning_rate": 1.6748771394307584e-06,
"loss": 6.679605865478516,
"loss_idm": 0.370703125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0171875,
"loss_idm_text": 0.0,
"loss_idm_type": 0.37073889970779417,
"loss_sft": 0.01429830901324749,
"step": 1490,
"token_acc": 0.8450069266156981
},
{
"epoch": 0.34594594594594597,
"grad_norm": 155.4982452392578,
"learning_rate": 1.6135921418712959e-06,
"loss": 6.242042160034179,
"loss_idm": 0.3453125,
"loss_idm_anl": 0.0,
"loss_idm_coord": 0.0177490234375,
"loss_idm_text": 0.0,
"loss_idm_type": 0.34358363747596743,
"loss_sft": 0.01402926966547966,
"step": 1500,
"token_acc": 0.8482708050343194
},
{
"epoch": 0.34594594594594597,
"eval_loss": 0.7831034064292908,
"eval_runtime": 233.7646,
"eval_samples_per_second": 7.465,
"eval_steps_per_second": 1.869,
"eval_token_acc": 0.8332306899706579,
"step": 1500
}
],
"logging_steps": 10,
"max_steps": 2000,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 1.8759920794589987e+19,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}