{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.18707323917313629,
  "eval_steps": 200,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0023384154896642037,
      "grad_norm": 1.9981695413589478,
      "learning_rate": 4.4859813084112145e-06,
      "loss": 0.47716766357421875,
      "step": 25
    },
    {
      "epoch": 0.004676830979328407,
      "grad_norm": 1.9483280181884766,
      "learning_rate": 9.158878504672897e-06,
      "loss": 0.4592530059814453,
      "step": 50
    },
    {
      "epoch": 0.0070152464689926105,
      "grad_norm": 2.1290855407714844,
      "learning_rate": 1.3831775700934579e-05,
      "loss": 0.394501838684082,
      "step": 75
    },
    {
      "epoch": 0.009353661958656815,
      "grad_norm": 2.8394205570220947,
      "learning_rate": 1.8504672897196264e-05,
      "loss": 0.3752219009399414,
      "step": 100
    },
    {
      "epoch": 0.011692077448321018,
      "grad_norm": 2.5833706855773926,
      "learning_rate": 2.3177570093457944e-05,
      "loss": 0.40475303649902344,
      "step": 125
    },
    {
      "epoch": 0.014030492937985221,
      "grad_norm": 2.622011661529541,
      "learning_rate": 2.7850467289719627e-05,
      "loss": 0.45625301361083986,
      "step": 150
    },
    {
      "epoch": 0.016368908427649424,
      "grad_norm": 2.9406535625457764,
      "learning_rate": 3.2523364485981314e-05,
      "loss": 0.34918937683105467,
      "step": 175
    },
    {
      "epoch": 0.01870732391731363,
      "grad_norm": 3.8391785621643066,
      "learning_rate": 3.7196261682242994e-05,
      "loss": 0.41711517333984377,
      "step": 200
    },
    {
      "epoch": 0.01870732391731363,
      "eval_pairs_with_negatives_loss": 0.3914637565612793,
      "eval_pairs_with_negatives_runtime": 44.0353,
      "eval_pairs_with_negatives_samples_per_second": 28.523,
      "eval_pairs_with_negatives_steps_per_second": 3.565,
      "step": 200
    },
    {
      "epoch": 0.01870732391731363,
      "eval_positives_loss": 0.1480681300163269,
      "eval_positives_runtime": 51.7193,
      "eval_positives_samples_per_second": 42.537,
      "eval_positives_steps_per_second": 5.317,
      "step": 200
    },
    {
      "epoch": 0.02104573940697783,
      "grad_norm": 2.3628170490264893,
      "learning_rate": 4.186915887850468e-05,
      "loss": 0.4296648406982422,
      "step": 225
    },
    {
      "epoch": 0.023384154896642036,
      "grad_norm": 2.698835611343384,
      "learning_rate": 4.6542056074766354e-05,
      "loss": 0.43652393341064455,
      "step": 250
    },
    {
      "epoch": 0.02572257038630624,
      "grad_norm": 4.1166558265686035,
      "learning_rate": 5.121495327102804e-05,
      "loss": 0.4344004440307617,
      "step": 275
    },
    {
      "epoch": 0.028060985875970442,
      "grad_norm": 4.551556587219238,
      "learning_rate": 5.588785046728973e-05,
      "loss": 0.41842262268066405,
      "step": 300
    },
    {
      "epoch": 0.030399401365634647,
      "grad_norm": 2.3925657272338867,
      "learning_rate": 6.056074766355141e-05,
      "loss": 0.4197782897949219,
      "step": 325
    },
    {
      "epoch": 0.03273781685529885,
      "grad_norm": 3.1023190021514893,
      "learning_rate": 6.52336448598131e-05,
      "loss": 0.4292975616455078,
      "step": 350
    },
    {
      "epoch": 0.03507623234496305,
      "grad_norm": 4.1481032371521,
      "learning_rate": 6.990654205607478e-05,
      "loss": 0.4758753204345703,
      "step": 375
    },
    {
      "epoch": 0.03741464783462726,
      "grad_norm": 3.8799962997436523,
      "learning_rate": 7.457943925233644e-05,
      "loss": 0.3311753845214844,
      "step": 400
    },
    {
      "epoch": 0.03741464783462726,
      "eval_pairs_with_negatives_loss": 0.3694910407066345,
      "eval_pairs_with_negatives_runtime": 44.2166,
      "eval_pairs_with_negatives_samples_per_second": 28.406,
      "eval_pairs_with_negatives_steps_per_second": 3.551,
      "step": 400
    },
    {
      "epoch": 0.03741464783462726,
      "eval_positives_loss": 0.11797073483467102,
      "eval_positives_runtime": 51.3355,
      "eval_positives_samples_per_second": 42.855,
      "eval_positives_steps_per_second": 5.357,
      "step": 400
    },
    {
      "epoch": 0.03975306332429146,
      "grad_norm": 3.7068564891815186,
      "learning_rate": 7.925233644859813e-05,
      "loss": 0.388726806640625,
      "step": 425
    },
    {
      "epoch": 0.04209147881395566,
      "grad_norm": 2.3407328128814697,
      "learning_rate": 8.392523364485981e-05,
      "loss": 0.4401784133911133,
      "step": 450
    },
    {
      "epoch": 0.04442989430361987,
      "grad_norm": 3.0923948287963867,
      "learning_rate": 8.85981308411215e-05,
      "loss": 0.41050430297851564,
      "step": 475
    },
    {
      "epoch": 0.04676830979328407,
      "grad_norm": 2.626753568649292,
      "learning_rate": 9.327102803738317e-05,
      "loss": 0.3923164749145508,
      "step": 500
    },
    {
      "epoch": 0.04910672528294827,
      "grad_norm": 2.729557991027832,
      "learning_rate": 9.794392523364486e-05,
      "loss": 0.3163204765319824,
      "step": 525
    },
    {
      "epoch": 0.05144514077261248,
      "grad_norm": 3.1612608432769775,
      "learning_rate": 9.986215045293424e-05,
      "loss": 0.35652976989746094,
      "step": 550
    },
    {
      "epoch": 0.05378355626227668,
      "grad_norm": 2.224350690841675,
      "learning_rate": 9.961599054745964e-05,
      "loss": 0.37072269439697264,
      "step": 575
    },
    {
      "epoch": 0.056121971751940884,
      "grad_norm": 3.207382917404175,
      "learning_rate": 9.936983064198503e-05,
      "loss": 0.3007513427734375,
      "step": 600
    },
    {
      "epoch": 0.056121971751940884,
      "eval_pairs_with_negatives_loss": 0.3388254940509796,
      "eval_pairs_with_negatives_runtime": 43.9558,
      "eval_pairs_with_negatives_samples_per_second": 28.574,
      "eval_pairs_with_negatives_steps_per_second": 3.572,
      "step": 600
    },
    {
      "epoch": 0.056121971751940884,
      "eval_positives_loss": 0.10862970352172852,
      "eval_positives_runtime": 51.7355,
      "eval_positives_samples_per_second": 42.524,
      "eval_positives_steps_per_second": 5.315,
      "step": 600
    },
    {
      "epoch": 0.058460387241605086,
      "grad_norm": 4.494985103607178,
      "learning_rate": 9.912367073651045e-05,
      "loss": 0.35943916320800784,
      "step": 625
    },
    {
      "epoch": 0.060798802731269294,
      "grad_norm": 3.7057838439941406,
      "learning_rate": 9.887751083103585e-05,
      "loss": 0.39355968475341796,
      "step": 650
    },
    {
      "epoch": 0.06313721822093349,
      "grad_norm": 1.6454180479049683,
      "learning_rate": 9.863135092556125e-05,
      "loss": 0.32073760986328126,
      "step": 675
    },
    {
      "epoch": 0.0654756337105977,
      "grad_norm": 3.901522636413574,
      "learning_rate": 9.838519102008666e-05,
      "loss": 0.3371485900878906,
      "step": 700
    },
    {
      "epoch": 0.0678140492002619,
      "grad_norm": 2.2362253665924072,
      "learning_rate": 9.813903111461206e-05,
      "loss": 0.3385154724121094,
      "step": 725
    },
    {
      "epoch": 0.0701524646899261,
      "grad_norm": 2.490204334259033,
      "learning_rate": 9.789287120913746e-05,
      "loss": 0.2718297004699707,
      "step": 750
    },
    {
      "epoch": 0.07249088017959031,
      "grad_norm": 2.2875030040740967,
      "learning_rate": 9.764671130366287e-05,
      "loss": 0.44294830322265627,
      "step": 775
    },
    {
      "epoch": 0.07482929566925452,
      "grad_norm": 1.40684175491333,
      "learning_rate": 9.740055139818827e-05,
      "loss": 0.2684394645690918,
      "step": 800
    },
    {
      "epoch": 0.07482929566925452,
      "eval_pairs_with_negatives_loss": 0.34527480602264404,
      "eval_pairs_with_negatives_runtime": 44.0724,
      "eval_pairs_with_negatives_samples_per_second": 28.499,
      "eval_pairs_with_negatives_steps_per_second": 3.562,
      "step": 800
    },
    {
      "epoch": 0.07482929566925452,
      "eval_positives_loss": 0.10431129485368729,
      "eval_positives_runtime": 51.6255,
      "eval_positives_samples_per_second": 42.615,
      "eval_positives_steps_per_second": 5.327,
      "step": 800
    },
    {
      "epoch": 0.07716771115891871,
      "grad_norm": 2.8712947368621826,
      "learning_rate": 9.715439149271367e-05,
      "loss": 0.25394817352294924,
      "step": 825
    },
    {
      "epoch": 0.07950612664858292,
      "grad_norm": 1.8017457723617554,
      "learning_rate": 9.690823158723908e-05,
      "loss": 0.32388198852539063,
      "step": 850
    },
    {
      "epoch": 0.08184454213824713,
      "grad_norm": 2.945971727371216,
      "learning_rate": 9.666207168176448e-05,
      "loss": 0.29435272216796876,
      "step": 875
    },
    {
      "epoch": 0.08418295762791132,
      "grad_norm": 2.1370229721069336,
      "learning_rate": 9.641591177628988e-05,
      "loss": 0.30667646408081056,
      "step": 900
    },
    {
      "epoch": 0.08652137311757553,
      "grad_norm": 3.4880053997039795,
      "learning_rate": 9.616975187081529e-05,
      "loss": 0.3113353729248047,
      "step": 925
    },
    {
      "epoch": 0.08885978860723974,
      "grad_norm": 1.221985101699829,
      "learning_rate": 9.592359196534069e-05,
      "loss": 0.33869354248046873,
      "step": 950
    },
    {
      "epoch": 0.09119820409690393,
      "grad_norm": 2.5686190128326416,
      "learning_rate": 9.567743205986609e-05,
      "loss": 0.2734596633911133,
      "step": 975
    },
    {
      "epoch": 0.09353661958656814,
      "grad_norm": 1.7126892805099487,
      "learning_rate": 9.54312721543915e-05,
      "loss": 0.2984904098510742,
      "step": 1000
    },
    {
      "epoch": 0.09353661958656814,
      "eval_pairs_with_negatives_loss": 0.3210560381412506,
      "eval_pairs_with_negatives_runtime": 44.9784,
      "eval_pairs_with_negatives_samples_per_second": 27.925,
      "eval_pairs_with_negatives_steps_per_second": 3.491,
      "step": 1000
    },
    {
      "epoch": 0.09353661958656814,
      "eval_positives_loss": 0.08913413435220718,
      "eval_positives_runtime": 52.3889,
      "eval_positives_samples_per_second": 41.994,
      "eval_positives_steps_per_second": 5.249,
      "step": 1000
    },
    {
      "epoch": 0.09587503507623235,
      "grad_norm": 1.6900842189788818,
      "learning_rate": 9.51851122489169e-05,
      "loss": 0.355332145690918,
      "step": 1025
    },
    {
      "epoch": 0.09821345056589655,
      "grad_norm": 0.7989633083343506,
      "learning_rate": 9.49389523434423e-05,
      "loss": 0.25679166793823244,
      "step": 1050
    },
    {
      "epoch": 0.10055186605556075,
      "grad_norm": 1.7311954498291016,
      "learning_rate": 9.469279243796771e-05,
      "loss": 0.344718017578125,
      "step": 1075
    },
    {
      "epoch": 0.10289028154522496,
      "grad_norm": 1.9694781303405762,
      "learning_rate": 9.444663253249311e-05,
      "loss": 0.3239378356933594,
      "step": 1100
    },
    {
      "epoch": 0.10522869703488916,
      "grad_norm": 2.0283656120300293,
      "learning_rate": 9.420047262701852e-05,
      "loss": 0.3015079879760742,
      "step": 1125
    },
    {
      "epoch": 0.10756711252455337,
      "grad_norm": 2.0422067642211914,
      "learning_rate": 9.395431272154392e-05,
      "loss": 0.3864572906494141,
      "step": 1150
    },
    {
      "epoch": 0.10990552801421756,
      "grad_norm": 2.2227938175201416,
      "learning_rate": 9.370815281606933e-05,
      "loss": 0.2981968116760254,
      "step": 1175
    },
    {
      "epoch": 0.11224394350388177,
      "grad_norm": 1.9632631540298462,
      "learning_rate": 9.346199291059473e-05,
      "loss": 0.3105415725708008,
      "step": 1200
    },
    {
      "epoch": 0.11224394350388177,
      "eval_pairs_with_negatives_loss": 0.3232487738132477,
      "eval_pairs_with_negatives_runtime": 44.5297,
      "eval_pairs_with_negatives_samples_per_second": 28.206,
      "eval_pairs_with_negatives_steps_per_second": 3.526,
      "step": 1200
    },
    {
      "epoch": 0.11224394350388177,
      "eval_positives_loss": 0.08286190032958984,
      "eval_positives_runtime": 52.0911,
      "eval_positives_samples_per_second": 42.234,
      "eval_positives_steps_per_second": 5.279,
      "step": 1200
    },
    {
      "epoch": 0.11458235899354598,
      "grad_norm": 0.5930349826812744,
      "learning_rate": 9.321583300512013e-05,
      "loss": 0.29635957717895506,
      "step": 1225
    },
    {
      "epoch": 0.11692077448321017,
      "grad_norm": 1.5982012748718262,
      "learning_rate": 9.296967309964554e-05,
      "loss": 0.24172809600830078,
      "step": 1250
    },
    {
      "epoch": 0.11925918997287438,
      "grad_norm": 0.5673118829727173,
      "learning_rate": 9.272351319417094e-05,
      "loss": 0.26859012603759763,
      "step": 1275
    },
    {
      "epoch": 0.12159760546253859,
      "grad_norm": 1.4948972463607788,
      "learning_rate": 9.247735328869634e-05,
      "loss": 0.2932207489013672,
      "step": 1300
    },
    {
      "epoch": 0.12393602095220278,
      "grad_norm": 1.545386791229248,
      "learning_rate": 9.223119338322175e-05,
      "loss": 0.2383422088623047,
      "step": 1325
    },
    {
      "epoch": 0.12627443644186698,
      "grad_norm": 2.5136566162109375,
      "learning_rate": 9.198503347774715e-05,
      "loss": 0.31081596374511716,
      "step": 1350
    },
    {
      "epoch": 0.12861285193153119,
      "grad_norm": 1.8467180728912354,
      "learning_rate": 9.173887357227255e-05,
      "loss": 0.321625862121582,
      "step": 1375
    },
    {
      "epoch": 0.1309512674211954,
      "grad_norm": 1.3382201194763184,
      "learning_rate": 9.149271366679796e-05,
      "loss": 0.2083415222167969,
      "step": 1400
    },
    {
      "epoch": 0.1309512674211954,
      "eval_pairs_with_negatives_loss": 0.3091065585613251,
      "eval_pairs_with_negatives_runtime": 44.3305,
      "eval_pairs_with_negatives_samples_per_second": 28.333,
      "eval_pairs_with_negatives_steps_per_second": 3.542,
      "step": 1400
    },
    {
      "epoch": 0.1309512674211954,
      "eval_positives_loss": 0.08935973793268204,
      "eval_positives_runtime": 51.7119,
      "eval_positives_samples_per_second": 42.543,
      "eval_positives_steps_per_second": 5.318,
      "step": 1400
    },
    {
      "epoch": 0.1332896829108596,
      "grad_norm": 1.0311458110809326,
      "learning_rate": 9.124655376132336e-05,
      "loss": 0.2932785987854004,
      "step": 1425
    },
    {
      "epoch": 0.1356280984005238,
      "grad_norm": 2.180497884750366,
      "learning_rate": 9.100039385584876e-05,
      "loss": 0.2038213348388672,
      "step": 1450
    },
    {
      "epoch": 0.13796651389018802,
      "grad_norm": 2.1046860218048096,
      "learning_rate": 9.075423395037416e-05,
      "loss": 0.2514521217346191,
      "step": 1475
    },
    {
      "epoch": 0.1403049293798522,
      "grad_norm": 2.4484400749206543,
      "learning_rate": 9.050807404489957e-05,
      "loss": 0.26426057815551757,
      "step": 1500
    },
    {
      "epoch": 0.1426433448695164,
      "grad_norm": 0.5209934711456299,
      "learning_rate": 9.026191413942497e-05,
      "loss": 0.2484210968017578,
      "step": 1525
    },
    {
      "epoch": 0.14498176035918062,
      "grad_norm": 2.0274691581726074,
      "learning_rate": 9.001575423395037e-05,
      "loss": 0.32157943725585936,
      "step": 1550
    },
    {
      "epoch": 0.14732017584884483,
      "grad_norm": 2.850849151611328,
      "learning_rate": 8.976959432847578e-05,
      "loss": 0.32649993896484375,
      "step": 1575
    },
    {
      "epoch": 0.14965859133850903,
      "grad_norm": 3.255429267883301,
      "learning_rate": 8.952343442300118e-05,
      "loss": 0.26256885528564455,
      "step": 1600
    },
    {
      "epoch": 0.14965859133850903,
      "eval_pairs_with_negatives_loss": 0.3166193962097168,
      "eval_pairs_with_negatives_runtime": 44.3277,
      "eval_pairs_with_negatives_samples_per_second": 28.334,
      "eval_pairs_with_negatives_steps_per_second": 3.542,
      "step": 1600
    },
    {
      "epoch": 0.14965859133850903,
      "eval_positives_loss": 0.077468141913414,
      "eval_positives_runtime": 51.8074,
      "eval_positives_samples_per_second": 42.465,
      "eval_positives_steps_per_second": 5.308,
      "step": 1600
    },
    {
      "epoch": 0.15199700682817324,
      "grad_norm": 2.324928045272827,
      "learning_rate": 8.927727451752658e-05,
      "loss": 0.28111976623535156,
      "step": 1625
    },
    {
      "epoch": 0.15433542231783742,
      "grad_norm": 1.722293496131897,
      "learning_rate": 8.9031114612052e-05,
      "loss": 0.27923690795898437,
      "step": 1650
    },
    {
      "epoch": 0.15667383780750163,
      "grad_norm": 2.5481624603271484,
      "learning_rate": 8.87849547065774e-05,
      "loss": 0.2887708282470703,
      "step": 1675
    },
    {
      "epoch": 0.15901225329716584,
      "grad_norm": 1.4688785076141357,
      "learning_rate": 8.853879480110279e-05,
      "loss": 0.32429874420166016,
      "step": 1700
    },
    {
      "epoch": 0.16135066878683005,
      "grad_norm": 0.970000684261322,
      "learning_rate": 8.82926348956282e-05,
      "loss": 0.23183115005493163,
      "step": 1725
    },
    {
      "epoch": 0.16368908427649426,
      "grad_norm": 2.141195297241211,
      "learning_rate": 8.80464749901536e-05,
      "loss": 0.2943249702453613,
      "step": 1750
    },
    {
      "epoch": 0.16602749976615844,
      "grad_norm": 0.9625000357627869,
      "learning_rate": 8.7800315084679e-05,
      "loss": 0.24936399459838868,
      "step": 1775
    },
    {
      "epoch": 0.16836591525582265,
      "grad_norm": 2.191002368927002,
      "learning_rate": 8.755415517920442e-05,
      "loss": 0.3477742004394531,
      "step": 1800
    },
    {
      "epoch": 0.16836591525582265,
      "eval_pairs_with_negatives_loss": 0.311288446187973,
      "eval_pairs_with_negatives_runtime": 44.4068,
      "eval_pairs_with_negatives_samples_per_second": 28.284,
      "eval_pairs_with_negatives_steps_per_second": 3.535,
      "step": 1800
    },
    {
      "epoch": 0.16836591525582265,
      "eval_positives_loss": 0.07508163899183273,
      "eval_positives_runtime": 51.7306,
      "eval_positives_samples_per_second": 42.528,
      "eval_positives_steps_per_second": 5.316,
      "step": 1800
    },
    {
      "epoch": 0.17070433074548685,
      "grad_norm": 2.958533525466919,
      "learning_rate": 8.730799527372981e-05,
      "loss": 0.3265147018432617,
      "step": 1825
    },
    {
      "epoch": 0.17304274623515106,
      "grad_norm": 2.005192756652832,
      "learning_rate": 8.706183536825521e-05,
      "loss": 0.29331169128417967,
      "step": 1850
    },
    {
      "epoch": 0.17538116172481527,
      "grad_norm": 2.02463960647583,
      "learning_rate": 8.681567546278063e-05,
      "loss": 0.2671446990966797,
      "step": 1875
    },
    {
      "epoch": 0.17771957721447948,
      "grad_norm": 2.153892755508423,
      "learning_rate": 8.656951555730604e-05,
      "loss": 0.29273582458496095,
      "step": 1900
    },
    {
      "epoch": 0.18005799270414366,
      "grad_norm": 2.7679359912872314,
      "learning_rate": 8.632335565183144e-05,
      "loss": 0.29389562606811526,
      "step": 1925
    },
    {
      "epoch": 0.18239640819380787,
      "grad_norm": 2.963479995727539,
      "learning_rate": 8.607719574635684e-05,
      "loss": 0.2356496810913086,
      "step": 1950
    },
    {
      "epoch": 0.18473482368347208,
      "grad_norm": 2.6095352172851562,
      "learning_rate": 8.583103584088225e-05,
      "loss": 0.24131065368652344,
      "step": 1975
    },
    {
      "epoch": 0.18707323917313629,
      "grad_norm": 2.1474149227142334,
      "learning_rate": 8.558487593540765e-05,
      "loss": 0.20256967544555665,
      "step": 2000
    },
    {
      "epoch": 0.18707323917313629,
      "eval_pairs_with_negatives_loss": 0.2920902371406555,
      "eval_pairs_with_negatives_runtime": 44.1734,
      "eval_pairs_with_negatives_samples_per_second": 28.433,
      "eval_pairs_with_negatives_steps_per_second": 3.554,
      "step": 2000
    },
    {
      "epoch": 0.18707323917313629,
      "eval_positives_loss": 0.06497267633676529,
      "eval_positives_runtime": 51.883,
      "eval_positives_samples_per_second": 42.403,
      "eval_positives_steps_per_second": 5.3,
      "step": 2000
    }
  ],
  "logging_steps": 25,
  "max_steps": 10691,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}