{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9763779527559056,
  "eval_steps": 5,
  "global_step": 189,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.015748031496062992,
      "grad_norm": 5.510926246643066,
      "learning_rate": 1.9894179894179895e-05,
      "loss": 0.9249,
      "step": 1
    },
    {
      "epoch": 0.031496062992125984,
      "grad_norm": 8.61505126953125,
      "learning_rate": 1.978835978835979e-05,
      "loss": 0.8445,
      "step": 2
    },
    {
      "epoch": 0.047244094488188976,
      "grad_norm": 7.036591529846191,
      "learning_rate": 1.9682539682539684e-05,
      "loss": 0.9654,
      "step": 3
    },
    {
      "epoch": 0.06299212598425197,
      "grad_norm": 5.803933143615723,
      "learning_rate": 1.9576719576719577e-05,
      "loss": 0.9276,
      "step": 4
    },
    {
      "epoch": 0.07874015748031496,
      "grad_norm": 5.716428756713867,
      "learning_rate": 1.947089947089947e-05,
      "loss": 0.9241,
      "step": 5
    },
    {
      "epoch": 0.07874015748031496,
      "eval_accuracy": 0.5678496868475992,
      "eval_loss": 0.6996241807937622,
      "eval_runtime": 108.7291,
      "eval_samples_per_second": 4.405,
      "eval_steps_per_second": 0.552,
      "step": 5
    },
    {
      "epoch": 0.09448818897637795,
      "grad_norm": 7.55866813659668,
      "learning_rate": 1.9365079365079367e-05,
      "loss": 0.9947,
      "step": 6
    },
    {
      "epoch": 0.11023622047244094,
      "grad_norm": 6.801171779632568,
      "learning_rate": 1.925925925925926e-05,
      "loss": 0.972,
      "step": 7
    },
    {
      "epoch": 0.12598425196850394,
      "grad_norm": 4.845946311950684,
      "learning_rate": 1.9153439153439156e-05,
      "loss": 0.6478,
      "step": 8
    },
    {
      "epoch": 0.14173228346456693,
      "grad_norm": 10.487945556640625,
      "learning_rate": 1.904761904761905e-05,
      "loss": 0.8597,
      "step": 9
    },
    {
      "epoch": 0.15748031496062992,
      "grad_norm": 5.452786445617676,
      "learning_rate": 1.8941798941798943e-05,
      "loss": 0.7708,
      "step": 10
    },
    {
      "epoch": 0.15748031496062992,
      "eval_accuracy": 0.6659707724425887,
      "eval_loss": 0.6283570528030396,
      "eval_runtime": 108.7155,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 10
    },
    {
      "epoch": 0.1732283464566929,
      "grad_norm": 4.522532939910889,
      "learning_rate": 1.8835978835978836e-05,
      "loss": 0.6848,
      "step": 11
    },
    {
      "epoch": 0.1889763779527559,
      "grad_norm": 6.4987688064575195,
      "learning_rate": 1.8730158730158732e-05,
      "loss": 0.6644,
      "step": 12
    },
    {
      "epoch": 0.2047244094488189,
      "grad_norm": 4.2297682762146,
      "learning_rate": 1.8624338624338625e-05,
      "loss": 0.7227,
      "step": 13
    },
    {
      "epoch": 0.2204724409448819,
      "grad_norm": 6.5658063888549805,
      "learning_rate": 1.851851851851852e-05,
      "loss": 0.6991,
      "step": 14
    },
    {
      "epoch": 0.23622047244094488,
      "grad_norm": 6.549685001373291,
      "learning_rate": 1.8412698412698415e-05,
      "loss": 0.7875,
      "step": 15
    },
    {
      "epoch": 0.23622047244094488,
      "eval_accuracy": 0.7244258872651357,
      "eval_loss": 0.5749094486236572,
      "eval_runtime": 108.7121,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 15
    },
    {
      "epoch": 0.25196850393700787,
      "grad_norm": 3.6349198818206787,
      "learning_rate": 1.8306878306878308e-05,
      "loss": 0.5732,
      "step": 16
    },
    {
      "epoch": 0.2677165354330709,
      "grad_norm": 4.741979598999023,
      "learning_rate": 1.82010582010582e-05,
      "loss": 0.5774,
      "step": 17
    },
    {
      "epoch": 0.28346456692913385,
      "grad_norm": 4.751223087310791,
      "learning_rate": 1.8095238095238097e-05,
      "loss": 0.5738,
      "step": 18
    },
    {
      "epoch": 0.2992125984251969,
      "grad_norm": 5.214819431304932,
      "learning_rate": 1.798941798941799e-05,
      "loss": 0.7182,
      "step": 19
    },
    {
      "epoch": 0.31496062992125984,
      "grad_norm": 5.566962718963623,
      "learning_rate": 1.7883597883597884e-05,
      "loss": 0.6575,
      "step": 20
    },
    {
      "epoch": 0.31496062992125984,
      "eval_accuracy": 0.7390396659707724,
      "eval_loss": 0.5360159873962402,
      "eval_runtime": 108.7252,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 20
    },
    {
      "epoch": 0.33070866141732286,
      "grad_norm": 4.060683727264404,
      "learning_rate": 1.7777777777777777e-05,
      "loss": 0.5976,
      "step": 21
    },
    {
      "epoch": 0.3464566929133858,
      "grad_norm": 5.9868621826171875,
      "learning_rate": 1.7671957671957673e-05,
      "loss": 0.7734,
      "step": 22
    },
    {
      "epoch": 0.36220472440944884,
      "grad_norm": 3.4295496940612793,
      "learning_rate": 1.7566137566137566e-05,
      "loss": 0.5543,
      "step": 23
    },
    {
      "epoch": 0.3779527559055118,
      "grad_norm": 4.587719917297363,
      "learning_rate": 1.7460317460317463e-05,
      "loss": 0.6497,
      "step": 24
    },
    {
      "epoch": 0.3937007874015748,
      "grad_norm": 4.520890235900879,
      "learning_rate": 1.7354497354497356e-05,
      "loss": 0.6802,
      "step": 25
    },
    {
      "epoch": 0.3937007874015748,
      "eval_accuracy": 0.7432150313152401,
      "eval_loss": 0.5086582899093628,
      "eval_runtime": 108.7194,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 25
    },
    {
      "epoch": 0.4094488188976378,
      "grad_norm": 3.653116464614868,
      "learning_rate": 1.724867724867725e-05,
      "loss": 0.4888,
      "step": 26
    },
    {
      "epoch": 0.4251968503937008,
      "grad_norm": 4.042315483093262,
      "learning_rate": 1.7142857142857142e-05,
      "loss": 0.6004,
      "step": 27
    },
    {
      "epoch": 0.4409448818897638,
      "grad_norm": 5.317520618438721,
      "learning_rate": 1.7037037037037038e-05,
      "loss": 0.6253,
      "step": 28
    },
    {
      "epoch": 0.4566929133858268,
      "grad_norm": 3.8642020225524902,
      "learning_rate": 1.693121693121693e-05,
      "loss": 0.5778,
      "step": 29
    },
    {
      "epoch": 0.47244094488188976,
      "grad_norm": 2.3941361904144287,
      "learning_rate": 1.6825396825396828e-05,
      "loss": 0.3982,
      "step": 30
    },
    {
      "epoch": 0.47244094488188976,
      "eval_accuracy": 0.7578288100208769,
      "eval_loss": 0.4889708459377289,
      "eval_runtime": 108.7183,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 30
    },
    {
      "epoch": 0.4881889763779528,
      "grad_norm": 4.1248650550842285,
      "learning_rate": 1.671957671957672e-05,
      "loss": 0.777,
      "step": 31
    },
    {
      "epoch": 0.5039370078740157,
      "grad_norm": 3.369483470916748,
      "learning_rate": 1.6613756613756614e-05,
      "loss": 0.5675,
      "step": 32
    },
    {
      "epoch": 0.5196850393700787,
      "grad_norm": 3.8457119464874268,
      "learning_rate": 1.6507936507936507e-05,
      "loss": 0.6227,
      "step": 33
    },
    {
      "epoch": 0.5354330708661418,
      "grad_norm": 4.809354782104492,
      "learning_rate": 1.6402116402116404e-05,
      "loss": 0.7111,
      "step": 34
    },
    {
      "epoch": 0.5511811023622047,
      "grad_norm": 2.84769868850708,
      "learning_rate": 1.6296296296296297e-05,
      "loss": 0.4555,
      "step": 35
    },
    {
      "epoch": 0.5511811023622047,
      "eval_accuracy": 0.7599164926931107,
      "eval_loss": 0.4774630665779114,
      "eval_runtime": 108.7145,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 35
    },
    {
      "epoch": 0.5669291338582677,
      "grad_norm": 4.496406555175781,
      "learning_rate": 1.6190476190476193e-05,
      "loss": 0.6703,
      "step": 36
    },
    {
      "epoch": 0.5826771653543307,
      "grad_norm": 5.721245288848877,
      "learning_rate": 1.6084656084656086e-05,
      "loss": 0.7066,
      "step": 37
    },
    {
      "epoch": 0.5984251968503937,
      "grad_norm": 4.494580268859863,
      "learning_rate": 1.597883597883598e-05,
      "loss": 0.4907,
      "step": 38
    },
    {
      "epoch": 0.6141732283464567,
      "grad_norm": 2.8905560970306396,
      "learning_rate": 1.5873015873015872e-05,
      "loss": 0.5501,
      "step": 39
    },
    {
      "epoch": 0.6299212598425197,
      "grad_norm": 9.776362419128418,
      "learning_rate": 1.576719576719577e-05,
      "loss": 0.8838,
      "step": 40
    },
    {
      "epoch": 0.6299212598425197,
      "eval_accuracy": 0.7661795407098121,
      "eval_loss": 0.46829721331596375,
      "eval_runtime": 108.7189,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 40
    },
    {
      "epoch": 0.6456692913385826,
      "grad_norm": 3.8481881618499756,
      "learning_rate": 1.5661375661375662e-05,
      "loss": 0.5309,
      "step": 41
    },
    {
      "epoch": 0.6614173228346457,
      "grad_norm": 6.0327839851379395,
      "learning_rate": 1.555555555555556e-05,
      "loss": 0.6414,
      "step": 42
    },
    {
      "epoch": 0.6771653543307087,
      "grad_norm": 4.993657112121582,
      "learning_rate": 1.544973544973545e-05,
      "loss": 0.5727,
      "step": 43
    },
    {
      "epoch": 0.6929133858267716,
      "grad_norm": 4.3265252113342285,
      "learning_rate": 1.5343915343915344e-05,
      "loss": 0.4913,
      "step": 44
    },
    {
      "epoch": 0.7086614173228346,
      "grad_norm": 3.6012353897094727,
      "learning_rate": 1.523809523809524e-05,
      "loss": 0.4692,
      "step": 45
    },
    {
      "epoch": 0.7086614173228346,
      "eval_accuracy": 0.7661795407098121,
      "eval_loss": 0.4610559344291687,
      "eval_runtime": 108.7229,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 45
    },
    {
      "epoch": 0.7244094488188977,
      "grad_norm": 4.319406509399414,
      "learning_rate": 1.5132275132275134e-05,
      "loss": 0.5203,
      "step": 46
    },
    {
      "epoch": 0.7401574803149606,
      "grad_norm": 3.885263442993164,
      "learning_rate": 1.5026455026455027e-05,
      "loss": 0.5084,
      "step": 47
    },
    {
      "epoch": 0.7559055118110236,
      "grad_norm": 3.547327995300293,
      "learning_rate": 1.4920634920634922e-05,
      "loss": 0.442,
      "step": 48
    },
    {
      "epoch": 0.7716535433070866,
      "grad_norm": 3.8868982791900635,
      "learning_rate": 1.4814814814814815e-05,
      "loss": 0.5848,
      "step": 49
    },
    {
      "epoch": 0.7874015748031497,
      "grad_norm": 2.222346544265747,
      "learning_rate": 1.470899470899471e-05,
      "loss": 0.5455,
      "step": 50
    },
    {
      "epoch": 0.7874015748031497,
      "eval_accuracy": 0.7620041753653445,
      "eval_loss": 0.4531377851963043,
      "eval_runtime": 108.7528,
      "eval_samples_per_second": 4.404,
      "eval_steps_per_second": 0.552,
      "step": 50
    },
    {
      "epoch": 0.8031496062992126,
      "grad_norm": 3.129575252532959,
      "learning_rate": 1.4603174603174603e-05,
      "loss": 0.4861,
      "step": 51
    },
    {
      "epoch": 0.8188976377952756,
      "grad_norm": 4.924710750579834,
      "learning_rate": 1.44973544973545e-05,
      "loss": 0.5782,
      "step": 52
    },
    {
      "epoch": 0.8346456692913385,
      "grad_norm": 5.2157182693481445,
      "learning_rate": 1.4391534391534392e-05,
      "loss": 0.7203,
      "step": 53
    },
    {
      "epoch": 0.8503937007874016,
      "grad_norm": 4.697371959686279,
      "learning_rate": 1.4285714285714287e-05,
      "loss": 0.4261,
      "step": 54
    },
    {
      "epoch": 0.8661417322834646,
      "grad_norm": 2.8899056911468506,
      "learning_rate": 1.417989417989418e-05,
      "loss": 0.5696,
      "step": 55
    },
    {
      "epoch": 0.8661417322834646,
      "eval_accuracy": 0.7661795407098121,
      "eval_loss": 0.4459321200847626,
      "eval_runtime": 108.7951,
      "eval_samples_per_second": 4.403,
      "eval_steps_per_second": 0.551,
      "step": 55
    },
    {
      "epoch": 0.8818897637795275,
      "grad_norm": 4.532041072845459,
      "learning_rate": 1.4074074074074075e-05,
      "loss": 0.5723,
      "step": 56
    },
    {
      "epoch": 0.8976377952755905,
      "grad_norm": 2.3436343669891357,
      "learning_rate": 1.3968253968253968e-05,
      "loss": 0.3629,
      "step": 57
    },
    {
      "epoch": 0.9133858267716536,
      "grad_norm": 3.333158493041992,
      "learning_rate": 1.3862433862433865e-05,
      "loss": 0.5433,
      "step": 58
    },
    {
      "epoch": 0.9291338582677166,
      "grad_norm": 4.177884101867676,
      "learning_rate": 1.3756613756613758e-05,
      "loss": 0.3747,
      "step": 59
    },
    {
      "epoch": 0.9448818897637795,
      "grad_norm": 5.238712310791016,
      "learning_rate": 1.3650793650793652e-05,
      "loss": 0.7453,
      "step": 60
    },
    {
      "epoch": 0.9448818897637795,
      "eval_accuracy": 0.7766179540709812,
      "eval_loss": 0.4413756728172302,
      "eval_runtime": 108.7463,
      "eval_samples_per_second": 4.405,
      "eval_steps_per_second": 0.552,
      "step": 60
    },
    {
      "epoch": 0.9606299212598425,
      "grad_norm": 4.022979736328125,
      "learning_rate": 1.3544973544973545e-05,
      "loss": 0.6177,
      "step": 61
    },
    {
      "epoch": 0.9763779527559056,
      "grad_norm": 2.0528969764709473,
      "learning_rate": 1.343915343915344e-05,
      "loss": 0.3505,
      "step": 62
    },
    {
      "epoch": 0.9921259842519685,
      "grad_norm": 3.9705586433410645,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.5858,
      "step": 63
    },
    {
      "epoch": 1.0078740157480315,
      "grad_norm": 8.341585159301758,
      "learning_rate": 1.322751322751323e-05,
      "loss": 0.6721,
      "step": 64
    },
    {
      "epoch": 1.0236220472440944,
      "grad_norm": 4.031370162963867,
      "learning_rate": 1.3121693121693123e-05,
      "loss": 0.5369,
      "step": 65
    },
    {
      "epoch": 1.0236220472440944,
      "eval_accuracy": 0.7828810020876826,
      "eval_loss": 0.43705105781555176,
      "eval_runtime": 108.7278,
      "eval_samples_per_second": 4.405,
      "eval_steps_per_second": 0.552,
      "step": 65
    },
    {
      "epoch": 1.0393700787401574,
      "grad_norm": 2.898926019668579,
      "learning_rate": 1.3015873015873018e-05,
      "loss": 0.3628,
      "step": 66
    },
    {
      "epoch": 1.0551181102362204,
      "grad_norm": 2.9200918674468994,
      "learning_rate": 1.291005291005291e-05,
      "loss": 0.3311,
      "step": 67
    },
    {
      "epoch": 1.0708661417322836,
      "grad_norm": 4.506103992462158,
      "learning_rate": 1.2804232804232805e-05,
      "loss": 0.5813,
      "step": 68
    },
    {
      "epoch": 1.0866141732283465,
      "grad_norm": 4.187809944152832,
      "learning_rate": 1.2698412698412699e-05,
      "loss": 0.4802,
      "step": 69
    },
    {
      "epoch": 1.1023622047244095,
      "grad_norm": 3.5520920753479004,
      "learning_rate": 1.2592592592592593e-05,
      "loss": 0.3994,
      "step": 70
    },
    {
      "epoch": 1.1023622047244095,
      "eval_accuracy": 0.7849686847599165,
      "eval_loss": 0.43335652351379395,
      "eval_runtime": 108.738,
      "eval_samples_per_second": 4.405,
      "eval_steps_per_second": 0.552,
      "step": 70
    },
    {
      "epoch": 1.1181102362204725,
      "grad_norm": 3.6081998348236084,
      "learning_rate": 1.2486772486772486e-05,
      "loss": 0.5266,
      "step": 71
    },
    {
      "epoch": 1.1338582677165354,
      "grad_norm": 3.6554276943206787,
      "learning_rate": 1.2380952380952383e-05,
      "loss": 0.5231,
      "step": 72
    },
    {
      "epoch": 1.1496062992125984,
      "grad_norm": 3.551367998123169,
      "learning_rate": 1.2275132275132276e-05,
      "loss": 0.4538,
      "step": 73
    },
    {
      "epoch": 1.1653543307086613,
      "grad_norm": 4.252958297729492,
      "learning_rate": 1.216931216931217e-05,
      "loss": 0.4688,
      "step": 74
    },
    {
      "epoch": 1.1811023622047245,
      "grad_norm": 4.337672710418701,
      "learning_rate": 1.2063492063492064e-05,
      "loss": 0.4235,
      "step": 75
    },
    {
      "epoch": 1.1811023622047245,
      "eval_accuracy": 0.791231732776618,
      "eval_loss": 0.429808109998703,
      "eval_runtime": 108.723,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 75
    },
    {
      "epoch": 1.1968503937007875,
      "grad_norm": 2.607356548309326,
      "learning_rate": 1.1957671957671959e-05,
      "loss": 0.3639,
      "step": 76
    },
    {
      "epoch": 1.2125984251968505,
      "grad_norm": 3.198551654815674,
      "learning_rate": 1.1851851851851852e-05,
      "loss": 0.4066,
      "step": 77
    },
    {
      "epoch": 1.2283464566929134,
      "grad_norm": 4.820532321929932,
      "learning_rate": 1.1746031746031748e-05,
      "loss": 0.5906,
      "step": 78
    },
    {
      "epoch": 1.2440944881889764,
      "grad_norm": 3.5706419944763184,
      "learning_rate": 1.1640211640211641e-05,
      "loss": 0.5065,
      "step": 79
    },
    {
      "epoch": 1.2598425196850394,
      "grad_norm": 4.763455867767334,
      "learning_rate": 1.1534391534391536e-05,
      "loss": 0.4811,
      "step": 80
    },
    {
      "epoch": 1.2598425196850394,
      "eval_accuracy": 0.791231732776618,
      "eval_loss": 0.4265703856945038,
      "eval_runtime": 108.7158,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 80
    },
    {
      "epoch": 1.2755905511811023,
      "grad_norm": 5.053676605224609,
      "learning_rate": 1.1428571428571429e-05,
      "loss": 0.5317,
      "step": 81
    },
    {
      "epoch": 1.2913385826771653,
      "grad_norm": 4.484920024871826,
      "learning_rate": 1.1322751322751324e-05,
      "loss": 0.5474,
      "step": 82
    },
    {
      "epoch": 1.3070866141732282,
      "grad_norm": 4.059377193450928,
      "learning_rate": 1.1216931216931217e-05,
      "loss": 0.4936,
      "step": 83
    },
    {
      "epoch": 1.3228346456692912,
      "grad_norm": 4.017063617706299,
      "learning_rate": 1.1111111111111113e-05,
      "loss": 0.3574,
      "step": 84
    },
    {
      "epoch": 1.3385826771653544,
      "grad_norm": 3.275650978088379,
      "learning_rate": 1.1005291005291006e-05,
      "loss": 0.5072,
      "step": 85
    },
    {
      "epoch": 1.3385826771653544,
      "eval_accuracy": 0.791231732776618,
      "eval_loss": 0.4252822697162628,
      "eval_runtime": 108.7176,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 85
    },
    {
      "epoch": 1.3543307086614174,
      "grad_norm": 5.258458614349365,
      "learning_rate": 1.0899470899470901e-05,
      "loss": 0.6735,
      "step": 86
    },
    {
      "epoch": 1.3700787401574803,
      "grad_norm": 3.070061445236206,
      "learning_rate": 1.0793650793650794e-05,
      "loss": 0.4365,
      "step": 87
    },
    {
      "epoch": 1.3858267716535433,
      "grad_norm": 3.556974172592163,
      "learning_rate": 1.0687830687830689e-05,
      "loss": 0.5113,
      "step": 88
    },
    {
      "epoch": 1.4015748031496063,
      "grad_norm": 2.5072743892669678,
      "learning_rate": 1.0582010582010582e-05,
      "loss": 0.286,
      "step": 89
    },
    {
      "epoch": 1.4173228346456692,
      "grad_norm": 4.407125949859619,
      "learning_rate": 1.0476190476190477e-05,
      "loss": 0.4405,
      "step": 90
    },
    {
      "epoch": 1.4173228346456692,
      "eval_accuracy": 0.7849686847599165,
      "eval_loss": 0.42280957102775574,
      "eval_runtime": 108.7146,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 90
    },
    {
      "epoch": 1.4330708661417324,
      "grad_norm": 6.310215473175049,
      "learning_rate": 1.037037037037037e-05,
      "loss": 0.6205,
      "step": 91
    },
    {
      "epoch": 1.4488188976377954,
      "grad_norm": 3.586291551589966,
      "learning_rate": 1.0264550264550266e-05,
      "loss": 0.4571,
      "step": 92
    },
    {
      "epoch": 1.4645669291338583,
      "grad_norm": 4.950135707855225,
      "learning_rate": 1.015873015873016e-05,
      "loss": 0.7159,
      "step": 93
    },
    {
      "epoch": 1.4803149606299213,
      "grad_norm": 2.9908485412597656,
      "learning_rate": 1.0052910052910054e-05,
      "loss": 0.4769,
      "step": 94
    },
    {
      "epoch": 1.4960629921259843,
      "grad_norm": 4.945335865020752,
      "learning_rate": 9.947089947089947e-06,
      "loss": 0.5349,
      "step": 95
    },
    {
      "epoch": 1.4960629921259843,
      "eval_accuracy": 0.7870563674321504,
      "eval_loss": 0.41962236166000366,
      "eval_runtime": 108.7194,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 95
    },
    {
      "epoch": 1.5118110236220472,
      "grad_norm": 4.648383140563965,
      "learning_rate": 9.841269841269842e-06,
      "loss": 0.5035,
      "step": 96
    },
    {
      "epoch": 1.5275590551181102,
      "grad_norm": 4.447684288024902,
      "learning_rate": 9.735449735449735e-06,
      "loss": 0.5128,
      "step": 97
    },
    {
      "epoch": 1.5433070866141732,
      "grad_norm": 3.652973175048828,
      "learning_rate": 9.62962962962963e-06,
      "loss": 0.3454,
      "step": 98
    },
    {
      "epoch": 1.5590551181102361,
      "grad_norm": 3.083529472351074,
      "learning_rate": 9.523809523809525e-06,
      "loss": 0.4522,
      "step": 99
    },
    {
      "epoch": 1.574803149606299,
      "grad_norm": 2.6377124786376953,
      "learning_rate": 9.417989417989418e-06,
      "loss": 0.3342,
      "step": 100
    },
    {
      "epoch": 1.574803149606299,
      "eval_accuracy": 0.7828810020876826,
      "eval_loss": 0.4169768989086151,
      "eval_runtime": 108.7386,
      "eval_samples_per_second": 4.405,
      "eval_steps_per_second": 0.552,
      "step": 100
    },
    {
      "epoch": 1.590551181102362,
      "grad_norm": 5.530861854553223,
      "learning_rate": 9.312169312169313e-06,
      "loss": 0.6154,
      "step": 101
    },
    {
      "epoch": 1.6062992125984252,
      "grad_norm": 2.849217176437378,
      "learning_rate": 9.206349206349207e-06,
      "loss": 0.3823,
      "step": 102
    },
    {
      "epoch": 1.6220472440944882,
      "grad_norm": 2.8741447925567627,
      "learning_rate": 9.1005291005291e-06,
      "loss": 0.2884,
      "step": 103
    },
    {
      "epoch": 1.6377952755905512,
      "grad_norm": 5.099402904510498,
      "learning_rate": 8.994708994708995e-06,
      "loss": 0.4426,
      "step": 104
    },
    {
      "epoch": 1.6535433070866141,
      "grad_norm": 3.130911350250244,
      "learning_rate": 8.888888888888888e-06,
      "loss": 0.5271,
      "step": 105
    },
    {
      "epoch": 1.6535433070866141,
      "eval_accuracy": 0.7933194154488518,
      "eval_loss": 0.41489914059638977,
      "eval_runtime": 108.7149,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 105
    },
    {
      "epoch": 1.6692913385826773,
      "grad_norm": 3.4253950119018555,
      "learning_rate": 8.783068783068783e-06,
      "loss": 0.4482,
      "step": 106
    },
    {
      "epoch": 1.6850393700787403,
      "grad_norm": 3.2515480518341064,
      "learning_rate": 8.677248677248678e-06,
      "loss": 0.5227,
      "step": 107
    },
    {
      "epoch": 1.7007874015748032,
      "grad_norm": 3.6166484355926514,
      "learning_rate": 8.571428571428571e-06,
      "loss": 0.4545,
      "step": 108
    },
    {
      "epoch": 1.7165354330708662,
      "grad_norm": 2.7220921516418457,
      "learning_rate": 8.465608465608466e-06,
      "loss": 0.3609,
      "step": 109
    },
    {
      "epoch": 1.7322834645669292,
      "grad_norm": 2.6449429988861084,
      "learning_rate": 8.35978835978836e-06,
      "loss": 0.3463,
      "step": 110
    },
    {
      "epoch": 1.7322834645669292,
      "eval_accuracy": 0.7974947807933194,
      "eval_loss": 0.41358983516693115,
      "eval_runtime": 108.7211,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 110
    },
    {
      "epoch": 1.7480314960629921,
      "grad_norm": 3.9665286540985107,
      "learning_rate": 8.253968253968254e-06,
      "loss": 0.5416,
      "step": 111
    },
    {
      "epoch": 1.763779527559055,
      "grad_norm": 3.658632516860962,
      "learning_rate": 8.148148148148148e-06,
      "loss": 0.423,
      "step": 112
    },
    {
      "epoch": 1.779527559055118,
      "grad_norm": 3.2784206867218018,
      "learning_rate": 8.042328042328043e-06,
      "loss": 0.4253,
      "step": 113
    },
    {
      "epoch": 1.795275590551181,
      "grad_norm": 2.654160737991333,
      "learning_rate": 7.936507936507936e-06,
      "loss": 0.4836,
      "step": 114
    },
    {
      "epoch": 1.811023622047244,
      "grad_norm": 4.960519313812256,
      "learning_rate": 7.830687830687831e-06,
      "loss": 0.4867,
      "step": 115
    },
    {
      "epoch": 1.811023622047244,
      "eval_accuracy": 0.7995824634655533,
      "eval_loss": 0.4127565622329712,
      "eval_runtime": 108.7361,
      "eval_samples_per_second": 4.405,
      "eval_steps_per_second": 0.552,
      "step": 115
    },
    {
      "epoch": 1.826771653543307,
      "grad_norm": 2.9885411262512207,
      "learning_rate": 7.724867724867726e-06,
      "loss": 0.3218,
      "step": 116
    },
    {
      "epoch": 1.84251968503937,
      "grad_norm": 3.868762254714966,
      "learning_rate": 7.61904761904762e-06,
      "loss": 0.5335,
      "step": 117
    },
    {
      "epoch": 1.858267716535433,
      "grad_norm": 3.111746072769165,
      "learning_rate": 7.5132275132275136e-06,
      "loss": 0.4498,
      "step": 118
    },
    {
      "epoch": 1.874015748031496,
      "grad_norm": 3.94144868850708,
      "learning_rate": 7.4074074074074075e-06,
      "loss": 0.3923,
      "step": 119
    },
    {
      "epoch": 1.889763779527559,
      "grad_norm": 3.6796834468841553,
      "learning_rate": 7.301587301587301e-06,
      "loss": 0.3221,
      "step": 120
    },
    {
      "epoch": 1.889763779527559,
      "eval_accuracy": 0.7995824634655533,
      "eval_loss": 0.4124543368816376,
      "eval_runtime": 108.7265,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 120
    },
    {
      "epoch": 1.905511811023622,
      "grad_norm": 3.597407341003418,
      "learning_rate": 7.195767195767196e-06,
      "loss": 0.5133,
      "step": 121
    },
    {
      "epoch": 1.9212598425196852,
      "grad_norm": 2.6824302673339844,
      "learning_rate": 7.08994708994709e-06,
      "loss": 0.4173,
      "step": 122
    },
    {
      "epoch": 1.9370078740157481,
      "grad_norm": 3.5606563091278076,
      "learning_rate": 6.984126984126984e-06,
      "loss": 0.5625,
      "step": 123
    },
    {
      "epoch": 1.952755905511811,
      "grad_norm": 2.981217622756958,
      "learning_rate": 6.878306878306879e-06,
      "loss": 0.3826,
      "step": 124
    },
    {
      "epoch": 1.968503937007874,
      "grad_norm": 2.708043336868286,
      "learning_rate": 6.772486772486773e-06,
      "loss": 0.3542,
      "step": 125
    },
    {
      "epoch": 1.968503937007874,
      "eval_accuracy": 0.7995824634655533,
      "eval_loss": 0.4116251766681671,
      "eval_runtime": 108.7821,
      "eval_samples_per_second": 4.403,
      "eval_steps_per_second": 0.552,
      "step": 125
    },
    {
      "epoch": 1.984251968503937,
      "grad_norm": 3.5302915573120117,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.5818,
      "step": 126
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.032024383544922,
      "learning_rate": 6.560846560846561e-06,
      "loss": 0.5345,
      "step": 127
    },
    {
      "epoch": 2.015748031496063,
      "grad_norm": 1.919519305229187,
      "learning_rate": 6.455026455026455e-06,
      "loss": 0.312,
      "step": 128
    },
    {
      "epoch": 2.031496062992126,
      "grad_norm": 3.025320291519165,
      "learning_rate": 6.349206349206349e-06,
      "loss": 0.4013,
      "step": 129
    },
    {
      "epoch": 2.047244094488189,
      "grad_norm": 4.498544216156006,
      "learning_rate": 6.243386243386243e-06,
      "loss": 0.5465,
      "step": 130
    },
    {
      "epoch": 2.047244094488189,
      "eval_accuracy": 0.7995824634655533,
      "eval_loss": 0.4106995761394501,
      "eval_runtime": 108.7564,
      "eval_samples_per_second": 4.404,
      "eval_steps_per_second": 0.552,
      "step": 130
    },
    {
      "epoch": 2.062992125984252,
      "grad_norm": 4.507568359375,
      "learning_rate": 6.137566137566138e-06,
      "loss": 0.4377,
      "step": 131
    },
    {
      "epoch": 2.078740157480315,
      "grad_norm": 3.309709310531616,
      "learning_rate": 6.031746031746032e-06,
      "loss": 0.4066,
      "step": 132
    },
    {
      "epoch": 2.094488188976378,
      "grad_norm": 3.5611140727996826,
      "learning_rate": 5.925925925925926e-06,
      "loss": 0.4138,
      "step": 133
    },
    {
      "epoch": 2.1102362204724407,
      "grad_norm": 5.7032670974731445,
      "learning_rate": 5.820105820105821e-06,
      "loss": 0.592,
      "step": 134
    },
    {
      "epoch": 2.1259842519685037,
      "grad_norm": 2.3446578979492188,
      "learning_rate": 5.7142857142857145e-06,
      "loss": 0.3427,
      "step": 135
    },
    {
      "epoch": 2.1259842519685037,
      "eval_accuracy": 0.7995824634655533,
      "eval_loss": 0.41008052229881287,
      "eval_runtime": 108.7205,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 135
    },
    {
      "epoch": 2.141732283464567,
      "grad_norm": 3.090672254562378,
      "learning_rate": 5.6084656084656084e-06,
      "loss": 0.3996,
      "step": 136
    },
    {
      "epoch": 2.15748031496063,
      "grad_norm": 2.8654699325561523,
      "learning_rate": 5.502645502645503e-06,
      "loss": 0.3874,
      "step": 137
    },
    {
      "epoch": 2.173228346456693,
      "grad_norm": 2.6311564445495605,
      "learning_rate": 5.396825396825397e-06,
      "loss": 0.4439,
      "step": 138
    },
    {
      "epoch": 2.188976377952756,
      "grad_norm": 3.5829622745513916,
      "learning_rate": 5.291005291005291e-06,
      "loss": 0.4736,
      "step": 139
    },
    {
      "epoch": 2.204724409448819,
      "grad_norm": 3.072617769241333,
      "learning_rate": 5.185185185185185e-06,
      "loss": 0.4787,
      "step": 140
    },
    {
      "epoch": 2.204724409448819,
      "eval_accuracy": 0.8037578288100209,
      "eval_loss": 0.40874549746513367,
      "eval_runtime": 108.7212,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 140
    },
    {
      "epoch": 2.220472440944882,
      "grad_norm": 3.076615571975708,
      "learning_rate": 5.07936507936508e-06,
      "loss": 0.3407,
      "step": 141
    },
    {
      "epoch": 2.236220472440945,
      "grad_norm": 3.896620273590088,
      "learning_rate": 4.973544973544974e-06,
      "loss": 0.2494,
      "step": 142
    },
    {
      "epoch": 2.251968503937008,
      "grad_norm": 2.670351266860962,
      "learning_rate": 4.867724867724868e-06,
      "loss": 0.2989,
      "step": 143
    },
    {
      "epoch": 2.267716535433071,
      "grad_norm": 3.2045724391937256,
      "learning_rate": 4.761904761904762e-06,
      "loss": 0.5389,
      "step": 144
    },
    {
      "epoch": 2.283464566929134,
      "grad_norm": 2.343071699142456,
      "learning_rate": 4.656084656084656e-06,
      "loss": 0.4229,
      "step": 145
    },
    {
      "epoch": 2.283464566929134,
      "eval_accuracy": 0.8016701461377871,
      "eval_loss": 0.40726396441459656,
      "eval_runtime": 108.722,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 145
    },
    {
      "epoch": 2.2992125984251968,
      "grad_norm": 4.241441249847412,
      "learning_rate": 4.55026455026455e-06,
      "loss": 0.3901,
      "step": 146
    },
    {
      "epoch": 2.3149606299212597,
      "grad_norm": 3.265329360961914,
      "learning_rate": 4.444444444444444e-06,
      "loss": 0.5216,
      "step": 147
    },
    {
      "epoch": 2.3307086614173227,
      "grad_norm": 3.340651273727417,
      "learning_rate": 4.338624338624339e-06,
      "loss": 0.4494,
      "step": 148
    },
    {
      "epoch": 2.3464566929133857,
      "grad_norm": 3.6961324214935303,
      "learning_rate": 4.232804232804233e-06,
      "loss": 0.5125,
      "step": 149
    },
    {
      "epoch": 2.362204724409449,
      "grad_norm": 3.192697286605835,
      "learning_rate": 4.126984126984127e-06,
      "loss": 0.4514,
      "step": 150
    },
    {
      "epoch": 2.362204724409449,
      "eval_accuracy": 0.8037578288100209,
      "eval_loss": 0.40633270144462585,
      "eval_runtime": 108.7184,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 150
    },
    {
      "epoch": 2.377952755905512,
      "grad_norm": 2.322136878967285,
      "learning_rate": 4.0211640211640215e-06,
      "loss": 0.2585,
      "step": 151
    },
    {
      "epoch": 2.393700787401575,
      "grad_norm": 3.902898073196411,
      "learning_rate": 3.9153439153439155e-06,
      "loss": 0.4682,
      "step": 152
    },
    {
      "epoch": 2.409448818897638,
      "grad_norm": 4.637646675109863,
      "learning_rate": 3.80952380952381e-06,
      "loss": 0.4195,
      "step": 153
    },
    {
      "epoch": 2.425196850393701,
      "grad_norm": 3.0741162300109863,
      "learning_rate": 3.7037037037037037e-06,
      "loss": 0.3182,
      "step": 154
    },
    {
      "epoch": 2.440944881889764,
      "grad_norm": 3.675708055496216,
      "learning_rate": 3.597883597883598e-06,
      "loss": 0.5116,
      "step": 155
    },
    {
      "epoch": 2.440944881889764,
      "eval_accuracy": 0.8037578288100209,
      "eval_loss": 0.40514081716537476,
      "eval_runtime": 108.7241,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 155
    },
    {
      "epoch": 2.456692913385827,
      "grad_norm": 5.543003082275391,
      "learning_rate": 3.492063492063492e-06,
      "loss": 0.4976,
      "step": 156
    },
    {
      "epoch": 2.47244094488189,
      "grad_norm": 4.1292595863342285,
      "learning_rate": 3.3862433862433864e-06,
      "loss": 0.415,
      "step": 157
    },
    {
      "epoch": 2.4881889763779528,
      "grad_norm": 4.066864967346191,
      "learning_rate": 3.2804232804232807e-06,
      "loss": 0.4172,
      "step": 158
    },
    {
      "epoch": 2.5039370078740157,
      "grad_norm": 3.1275038719177246,
      "learning_rate": 3.1746031746031746e-06,
      "loss": 0.4113,
      "step": 159
    },
    {
      "epoch": 2.5196850393700787,
      "grad_norm": 4.276533603668213,
      "learning_rate": 3.068783068783069e-06,
      "loss": 0.3234,
      "step": 160
    },
    {
      "epoch": 2.5196850393700787,
      "eval_accuracy": 0.8058455114822547,
      "eval_loss": 0.40446344017982483,
      "eval_runtime": 108.7229,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 160
    },
    {
      "epoch": 2.5354330708661417,
      "grad_norm": 5.183084011077881,
      "learning_rate": 2.962962962962963e-06,
      "loss": 0.5434,
      "step": 161
    },
    {
      "epoch": 2.5511811023622046,
      "grad_norm": 2.6602373123168945,
      "learning_rate": 2.8571428571428573e-06,
      "loss": 0.3082,
      "step": 162
    },
    {
      "epoch": 2.5669291338582676,
      "grad_norm": 3.0452637672424316,
      "learning_rate": 2.7513227513227516e-06,
      "loss": 0.3836,
      "step": 163
    },
    {
      "epoch": 2.5826771653543306,
      "grad_norm": 2.222245931625366,
      "learning_rate": 2.6455026455026455e-06,
      "loss": 0.3992,
      "step": 164
    },
    {
      "epoch": 2.5984251968503935,
      "grad_norm": 3.8691773414611816,
      "learning_rate": 2.53968253968254e-06,
      "loss": 0.3993,
      "step": 165
    },
    {
      "epoch": 2.5984251968503935,
      "eval_accuracy": 0.8058455114822547,
      "eval_loss": 0.40404006838798523,
      "eval_runtime": 108.7316,
      "eval_samples_per_second": 4.405,
      "eval_steps_per_second": 0.552,
      "step": 165
    },
    {
      "epoch": 2.6141732283464565,
      "grad_norm": 3.0435469150543213,
      "learning_rate": 2.433862433862434e-06,
      "loss": 0.4177,
      "step": 166
    },
    {
      "epoch": 2.6299212598425195,
      "grad_norm": 2.3059229850769043,
      "learning_rate": 2.328042328042328e-06,
      "loss": 0.2918,
      "step": 167
    },
    {
      "epoch": 2.6456692913385824,
      "grad_norm": 3.192077875137329,
      "learning_rate": 2.222222222222222e-06,
      "loss": 0.3203,
      "step": 168
    },
    {
      "epoch": 2.661417322834646,
      "grad_norm": 3.519620418548584,
      "learning_rate": 2.1164021164021164e-06,
      "loss": 0.2996,
      "step": 169
    },
    {
      "epoch": 2.677165354330709,
      "grad_norm": 3.1694862842559814,
      "learning_rate": 2.0105820105820108e-06,
      "loss": 0.3264,
      "step": 170
    },
    {
      "epoch": 2.677165354330709,
      "eval_accuracy": 0.8058455114822547,
      "eval_loss": 0.40368661284446716,
      "eval_runtime": 108.7143,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 170
    },
    {
      "epoch": 2.6929133858267718,
      "grad_norm": 3.3654487133026123,
      "learning_rate": 1.904761904761905e-06,
      "loss": 0.3288,
      "step": 171
    },
    {
      "epoch": 2.7086614173228347,
      "grad_norm": 4.05765438079834,
      "learning_rate": 1.798941798941799e-06,
      "loss": 0.4648,
      "step": 172
    },
    {
      "epoch": 2.7244094488188977,
      "grad_norm": 4.653472900390625,
      "learning_rate": 1.6931216931216932e-06,
      "loss": 0.5563,
      "step": 173
    },
    {
      "epoch": 2.7401574803149606,
      "grad_norm": 4.682580471038818,
      "learning_rate": 1.5873015873015873e-06,
      "loss": 0.4531,
      "step": 174
    },
    {
      "epoch": 2.7559055118110236,
      "grad_norm": 2.616373300552368,
      "learning_rate": 1.4814814814814815e-06,
      "loss": 0.3316,
      "step": 175
    },
    {
      "epoch": 2.7559055118110236,
      "eval_accuracy": 0.8037578288100209,
      "eval_loss": 0.4034806787967682,
      "eval_runtime": 108.7256,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 175
    },
    {
      "epoch": 2.7716535433070866,
      "grad_norm": 4.228743076324463,
      "learning_rate": 1.3756613756613758e-06,
      "loss": 0.5363,
      "step": 176
    },
    {
      "epoch": 2.7874015748031495,
      "grad_norm": 4.0345845222473145,
      "learning_rate": 1.26984126984127e-06,
      "loss": 0.3705,
      "step": 177
    },
    {
      "epoch": 2.8031496062992125,
      "grad_norm": 3.482527494430542,
      "learning_rate": 1.164021164021164e-06,
      "loss": 0.3347,
      "step": 178
    },
    {
      "epoch": 2.8188976377952755,
      "grad_norm": 2.3308494091033936,
      "learning_rate": 1.0582010582010582e-06,
      "loss": 0.3579,
      "step": 179
    },
    {
      "epoch": 2.8346456692913384,
      "grad_norm": 4.280152797698975,
      "learning_rate": 9.523809523809525e-07,
      "loss": 0.4855,
      "step": 180
    },
    {
      "epoch": 2.8346456692913384,
      "eval_accuracy": 0.8037578288100209,
      "eval_loss": 0.4035497307777405,
      "eval_runtime": 108.7255,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 180
    },
    {
      "epoch": 2.850393700787402,
      "grad_norm": 3.315803289413452,
      "learning_rate": 8.465608465608466e-07,
      "loss": 0.4837,
      "step": 181
    },
    {
      "epoch": 2.866141732283465,
      "grad_norm": 2.9597713947296143,
      "learning_rate": 7.407407407407407e-07,
      "loss": 0.3521,
      "step": 182
    },
    {
      "epoch": 2.8818897637795278,
      "grad_norm": 4.547348976135254,
      "learning_rate": 6.34920634920635e-07,
      "loss": 0.4159,
      "step": 183
    },
    {
      "epoch": 2.8976377952755907,
      "grad_norm": 3.0314743518829346,
      "learning_rate": 5.291005291005291e-07,
      "loss": 0.3792,
      "step": 184
    },
    {
      "epoch": 2.9133858267716537,
      "grad_norm": 4.220534801483154,
      "learning_rate": 4.232804232804233e-07,
      "loss": 0.536,
      "step": 185
    },
    {
      "epoch": 2.9133858267716537,
      "eval_accuracy": 0.8058455114822547,
      "eval_loss": 0.4035782217979431,
      "eval_runtime": 108.7202,
      "eval_samples_per_second": 4.406,
      "eval_steps_per_second": 0.552,
      "step": 185
    },
    {
      "epoch": 2.9291338582677167,
      "grad_norm": 3.666822910308838,
      "learning_rate": 3.174603174603175e-07,
      "loss": 0.2668,
      "step": 186
    },
    {
      "epoch": 2.9448818897637796,
      "grad_norm": 5.5073137283325195,
      "learning_rate": 2.1164021164021165e-07,
      "loss": 0.4872,
      "step": 187
    },
    {
      "epoch": 2.9606299212598426,
      "grad_norm": 2.3061676025390625,
      "learning_rate": 1.0582010582010582e-07,
      "loss": 0.3233,
      "step": 188
    },
    {
      "epoch": 2.9763779527559056,
      "grad_norm": 3.6440348625183105,
      "learning_rate": 0.0,
      "loss": 0.4471,
      "step": 189
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 189,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 20,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}