{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.5152129817444218,
  "eval_steps": 31,
  "global_step": 310,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.008113590263691683,
      "eval_loss": 5.708795547485352,
      "eval_runtime": 34.9917,
      "eval_samples_per_second": 5.944,
      "eval_steps_per_second": 0.743,
      "step": 1
    },
    {
      "epoch": 0.02434077079107505,
      "grad_norm": 27.649152755737305,
      "learning_rate": 3e-05,
      "loss": 22.4374,
      "step": 3
    },
    {
      "epoch": 0.0486815415821501,
      "grad_norm": 24.12411880493164,
      "learning_rate": 6e-05,
      "loss": 20.3477,
      "step": 6
    },
    {
      "epoch": 0.07302231237322515,
      "grad_norm": 22.957651138305664,
      "learning_rate": 9e-05,
      "loss": 13.9237,
      "step": 9
    },
    {
      "epoch": 0.0973630831643002,
      "grad_norm": 22.14281463623047,
      "learning_rate": 9.999238475781957e-05,
      "loss": 8.2192,
      "step": 12
    },
    {
      "epoch": 0.12170385395537525,
      "grad_norm": 15.811911582946777,
      "learning_rate": 9.99524110790929e-05,
      "loss": 5.4619,
      "step": 15
    },
    {
      "epoch": 0.1460446247464503,
      "grad_norm": 17.17688751220703,
      "learning_rate": 9.987820251299122e-05,
      "loss": 4.1998,
      "step": 18
    },
    {
      "epoch": 0.17038539553752535,
      "grad_norm": 9.06302547454834,
      "learning_rate": 9.976980991835894e-05,
      "loss": 3.5102,
      "step": 21
    },
    {
      "epoch": 0.1947261663286004,
      "grad_norm": 9.071863174438477,
      "learning_rate": 9.962730758206611e-05,
      "loss": 2.9501,
      "step": 24
    },
    {
      "epoch": 0.21906693711967545,
      "grad_norm": 10.715943336486816,
      "learning_rate": 9.945079316809585e-05,
      "loss": 2.8262,
      "step": 27
    },
    {
      "epoch": 0.2434077079107505,
      "grad_norm": 6.811004161834717,
      "learning_rate": 9.924038765061042e-05,
      "loss": 2.6217,
      "step": 30
    },
    {
      "epoch": 0.2515212981744422,
      "eval_loss": 0.6682190299034119,
      "eval_runtime": 35.2448,
      "eval_samples_per_second": 5.902,
      "eval_steps_per_second": 0.738,
      "step": 31
    },
    {
      "epoch": 0.26774847870182555,
      "grad_norm": 5.427604675292969,
      "learning_rate": 9.899623523104149e-05,
      "loss": 2.4006,
      "step": 33
    },
    {
      "epoch": 0.2920892494929006,
      "grad_norm": 7.537416458129883,
      "learning_rate": 9.871850323926177e-05,
      "loss": 2.7397,
      "step": 36
    },
    {
      "epoch": 0.31643002028397565,
      "grad_norm": 8.674805641174316,
      "learning_rate": 9.84073820189054e-05,
      "loss": 2.4828,
      "step": 39
    },
    {
      "epoch": 0.3407707910750507,
      "grad_norm": 8.167730331420898,
      "learning_rate": 9.806308479691595e-05,
      "loss": 2.5302,
      "step": 42
    },
    {
      "epoch": 0.36511156186612576,
      "grad_norm": 10.937932014465332,
      "learning_rate": 9.768584753741134e-05,
      "loss": 2.3383,
      "step": 45
    },
    {
      "epoch": 0.3894523326572008,
      "grad_norm": 3.8246636390686035,
      "learning_rate": 9.727592877996585e-05,
      "loss": 2.212,
      "step": 48
    },
    {
      "epoch": 0.41379310344827586,
      "grad_norm": 4.667800426483154,
      "learning_rate": 9.683360946241989e-05,
      "loss": 2.3763,
      "step": 51
    },
    {
      "epoch": 0.4381338742393509,
      "grad_norm": 4.990477561950684,
      "learning_rate": 9.635919272833938e-05,
      "loss": 2.3995,
      "step": 54
    },
    {
      "epoch": 0.46247464503042596,
      "grad_norm": 9.09072494506836,
      "learning_rate": 9.58530037192562e-05,
      "loss": 2.545,
      "step": 57
    },
    {
      "epoch": 0.486815415821501,
      "grad_norm": 5.338695049285889,
      "learning_rate": 9.53153893518325e-05,
      "loss": 2.3183,
      "step": 60
    },
    {
      "epoch": 0.5030425963488844,
      "eval_loss": 0.5645254850387573,
      "eval_runtime": 13.4155,
      "eval_samples_per_second": 15.504,
      "eval_steps_per_second": 1.938,
      "step": 62
    },
    {
      "epoch": 0.5111561866125761,
      "grad_norm": 3.9001667499542236,
      "learning_rate": 9.474671808010126e-05,
      "loss": 2.3843,
      "step": 63
    },
    {
      "epoch": 0.5354969574036511,
      "grad_norm": 6.379414081573486,
      "learning_rate": 9.414737964294636e-05,
      "loss": 2.2963,
      "step": 66
    },
    {
      "epoch": 0.5598377281947262,
      "grad_norm": 4.97390079498291,
      "learning_rate": 9.351778479699499e-05,
      "loss": 2.2517,
      "step": 69
    },
    {
      "epoch": 0.5841784989858012,
      "grad_norm": 4.063937187194824,
      "learning_rate": 9.285836503510562e-05,
      "loss": 2.2437,
      "step": 72
    },
    {
      "epoch": 0.6085192697768763,
      "grad_norm": 3.602717161178589,
      "learning_rate": 9.21695722906443e-05,
      "loss": 2.1695,
      "step": 75
    },
    {
      "epoch": 0.6328600405679513,
      "grad_norm": 4.555128574371338,
      "learning_rate": 9.145187862775209e-05,
      "loss": 2.1648,
      "step": 78
    },
    {
      "epoch": 0.6572008113590264,
      "grad_norm": 5.0544867515563965,
      "learning_rate": 9.070577591781597e-05,
      "loss": 2.0922,
      "step": 81
    },
    {
      "epoch": 0.6815415821501014,
      "grad_norm": 4.92878532409668,
      "learning_rate": 8.993177550236464e-05,
      "loss": 2.141,
      "step": 84
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 3.6681647300720215,
      "learning_rate": 8.91304078426207e-05,
      "loss": 2.1638,
      "step": 87
    },
    {
      "epoch": 0.7302231237322515,
      "grad_norm": 4.451002597808838,
      "learning_rate": 8.83022221559489e-05,
      "loss": 2.1766,
      "step": 90
    },
    {
      "epoch": 0.7545638945233266,
      "grad_norm": 5.902685165405273,
      "learning_rate": 8.744778603945011e-05,
      "loss": 2.1005,
      "step": 93
    },
    {
      "epoch": 0.7545638945233266,
      "eval_loss": 0.5568718314170837,
      "eval_runtime": 13.4611,
      "eval_samples_per_second": 15.452,
      "eval_steps_per_second": 1.931,
      "step": 93
    },
    {
      "epoch": 0.7789046653144016,
      "grad_norm": 4.811842918395996,
      "learning_rate": 8.656768508095853e-05,
      "loss": 2.2144,
      "step": 96
    },
    {
      "epoch": 0.8032454361054767,
      "grad_norm": 4.506133556365967,
      "learning_rate": 8.566252245770909e-05,
      "loss": 2.0727,
      "step": 99
    },
    {
      "epoch": 0.8275862068965517,
      "grad_norm": 3.898118734359741,
      "learning_rate": 8.473291852294987e-05,
      "loss": 1.9802,
      "step": 102
    },
    {
      "epoch": 0.8519269776876268,
      "grad_norm": 4.056739807128906,
      "learning_rate": 8.377951038078302e-05,
      "loss": 2.034,
      "step": 105
    },
    {
      "epoch": 0.8762677484787018,
      "grad_norm": 4.857217788696289,
      "learning_rate": 8.280295144952536e-05,
      "loss": 1.9623,
      "step": 108
    },
    {
      "epoch": 0.9006085192697769,
      "grad_norm": 4.711864948272705,
      "learning_rate": 8.18039110138882e-05,
      "loss": 2.3024,
      "step": 111
    },
    {
      "epoch": 0.9249492900608519,
      "grad_norm": 4.590112686157227,
      "learning_rate": 8.07830737662829e-05,
      "loss": 1.9634,
      "step": 114
    },
    {
      "epoch": 0.949290060851927,
      "grad_norm": 6.305273056030273,
      "learning_rate": 7.974113933756707e-05,
      "loss": 2.0567,
      "step": 117
    },
    {
      "epoch": 0.973630831643002,
      "grad_norm": 3.979123115539551,
      "learning_rate": 7.86788218175523e-05,
      "loss": 2.024,
      "step": 120
    },
    {
      "epoch": 0.9979716024340771,
      "grad_norm": 3.4800965785980225,
      "learning_rate": 7.75968492656029e-05,
      "loss": 2.2663,
      "step": 123
    },
    {
      "epoch": 1.0060851926977687,
      "eval_loss": 0.5130282640457153,
      "eval_runtime": 13.4323,
      "eval_samples_per_second": 15.485,
      "eval_steps_per_second": 1.936,
      "step": 124
    },
    {
      "epoch": 1.0223123732251522,
      "grad_norm": 4.823460578918457,
      "learning_rate": 7.649596321166024e-05,
      "loss": 1.8243,
      "step": 126
    },
    {
      "epoch": 1.0466531440162272,
      "grad_norm": 3.5276401042938232,
      "learning_rate": 7.537691814803521e-05,
      "loss": 1.8807,
      "step": 129
    },
    {
      "epoch": 1.0709939148073022,
      "grad_norm": 4.932689189910889,
      "learning_rate": 7.424048101231686e-05,
      "loss": 1.6877,
      "step": 132
    },
    {
      "epoch": 1.0953346855983772,
      "grad_norm": 4.79948091506958,
      "learning_rate": 7.308743066175172e-05,
      "loss": 1.675,
      "step": 135
    },
    {
      "epoch": 1.1196754563894524,
      "grad_norm": 5.320326328277588,
      "learning_rate": 7.191855733945387e-05,
      "loss": 1.8877,
      "step": 138
    },
    {
      "epoch": 1.1440162271805274,
      "grad_norm": 5.259227752685547,
      "learning_rate": 7.073466213281196e-05,
      "loss": 1.7584,
      "step": 141
    },
    {
      "epoch": 1.1683569979716024,
      "grad_norm": 3.7490923404693604,
      "learning_rate": 6.953655642446368e-05,
      "loss": 1.6734,
      "step": 144
    },
    {
      "epoch": 1.1926977687626774,
      "grad_norm": 5.446073532104492,
      "learning_rate": 6.832506133621487e-05,
      "loss": 1.9009,
      "step": 147
    },
    {
      "epoch": 1.2170385395537526,
      "grad_norm": 4.892331600189209,
      "learning_rate": 6.710100716628344e-05,
      "loss": 1.7449,
      "step": 150
    },
    {
      "epoch": 1.2413793103448276,
      "grad_norm": 5.024781227111816,
      "learning_rate": 6.586523282025462e-05,
      "loss": 1.7636,
      "step": 153
    },
    {
      "epoch": 1.2576064908722109,
      "eval_loss": 0.5101395845413208,
      "eval_runtime": 13.4502,
      "eval_samples_per_second": 15.464,
      "eval_steps_per_second": 1.933,
      "step": 155
    },
    {
      "epoch": 1.2657200811359026,
      "grad_norm": 4.381926536560059,
      "learning_rate": 6.461858523613684e-05,
      "loss": 1.8001,
      "step": 156
    },
    {
      "epoch": 1.2900608519269776,
      "grad_norm": 4.504693031311035,
      "learning_rate": 6.336191880391284e-05,
      "loss": 1.7454,
      "step": 159
    },
    {
      "epoch": 1.3144016227180528,
      "grad_norm": 5.055481910705566,
      "learning_rate": 6.209609477998338e-05,
      "loss": 1.7643,
      "step": 162
    },
    {
      "epoch": 1.3387423935091278,
      "grad_norm": 4.872294902801514,
      "learning_rate": 6.0821980696905146e-05,
      "loss": 1.6705,
      "step": 165
    },
    {
      "epoch": 1.3630831643002028,
      "grad_norm": 4.806439399719238,
      "learning_rate": 5.9540449768827246e-05,
      "loss": 1.7039,
      "step": 168
    },
    {
      "epoch": 1.3874239350912778,
      "grad_norm": 5.403498649597168,
      "learning_rate": 5.8252380293033884e-05,
      "loss": 1.7613,
      "step": 171
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 3.657991647720337,
      "learning_rate": 5.695865504800327e-05,
      "loss": 1.8176,
      "step": 174
    },
    {
      "epoch": 1.436105476673428,
      "grad_norm": 4.227349281311035,
      "learning_rate": 5.566016068839535e-05,
      "loss": 1.715,
      "step": 177
    },
    {
      "epoch": 1.460446247464503,
      "grad_norm": 3.618288278579712,
      "learning_rate": 5.435778713738292e-05,
      "loss": 1.6001,
      "step": 180
    },
    {
      "epoch": 1.484787018255578,
      "grad_norm": 6.720012187957764,
      "learning_rate": 5.3052426976742855e-05,
      "loss": 1.7047,
      "step": 183
    },
    {
      "epoch": 1.5091277890466532,
      "grad_norm": 6.494766712188721,
      "learning_rate": 5.174497483512506e-05,
      "loss": 1.7607,
      "step": 186
    },
    {
      "epoch": 1.5091277890466532,
      "eval_loss": 0.5061110258102417,
      "eval_runtime": 13.4249,
      "eval_samples_per_second": 15.494,
      "eval_steps_per_second": 1.937,
      "step": 186
    },
    {
      "epoch": 1.5334685598377282,
      "grad_norm": 15.1577730178833,
      "learning_rate": 5.04363267749187e-05,
      "loss": 1.6155,
      "step": 189
    },
    {
      "epoch": 1.5578093306288032,
      "grad_norm": 4.470703125,
      "learning_rate": 4.912737967813583e-05,
      "loss": 1.6756,
      "step": 192
    },
    {
      "epoch": 1.5821501014198782,
      "grad_norm": 5.094776630401611,
      "learning_rate": 4.781903063173321e-05,
      "loss": 1.5824,
      "step": 195
    },
    {
      "epoch": 1.6064908722109532,
      "grad_norm": 3.8165762424468994,
      "learning_rate": 4.6512176312793736e-05,
      "loss": 1.5824,
      "step": 198
    },
    {
      "epoch": 1.6308316430020284,
      "grad_norm": 5.532138347625732,
      "learning_rate": 4.52077123739888e-05,
      "loss": 1.744,
      "step": 201
    },
    {
      "epoch": 1.6551724137931034,
      "grad_norm": 4.430169582366943,
      "learning_rate": 4.390653282974264e-05,
      "loss": 1.6626,
      "step": 204
    },
    {
      "epoch": 1.6795131845841786,
      "grad_norm": 3.9386327266693115,
      "learning_rate": 4.260952944351947e-05,
      "loss": 1.4812,
      "step": 207
    },
    {
      "epoch": 1.7038539553752536,
      "grad_norm": 4.511838436126709,
      "learning_rate": 4.131759111665349e-05,
      "loss": 1.7028,
      "step": 210
    },
    {
      "epoch": 1.7281947261663286,
      "grad_norm": 4.205485820770264,
      "learning_rate": 4.003160327914015e-05,
      "loss": 1.761,
      "step": 213
    },
    {
      "epoch": 1.7525354969574036,
      "grad_norm": 4.336887836456299,
      "learning_rate": 3.875244728280676e-05,
      "loss": 1.5719,
      "step": 216
    },
    {
      "epoch": 1.7606490872210954,
      "eval_loss": 0.49112430214881897,
      "eval_runtime": 13.4322,
      "eval_samples_per_second": 15.485,
      "eval_steps_per_second": 1.936,
      "step": 217
    },
    {
      "epoch": 1.7768762677484786,
      "grad_norm": 5.897202968597412,
      "learning_rate": 3.748099979727792e-05,
      "loss": 1.6029,
      "step": 219
    },
    {
      "epoch": 1.8012170385395536,
      "grad_norm": 4.379933834075928,
      "learning_rate": 3.6218132209150045e-05,
      "loss": 1.6497,
      "step": 222
    },
    {
      "epoch": 1.8255578093306288,
      "grad_norm": 4.874183177947998,
      "learning_rate": 3.4964710024786354e-05,
      "loss": 1.6421,
      "step": 225
    },
    {
      "epoch": 1.8498985801217038,
      "grad_norm": 4.163862705230713,
      "learning_rate": 3.372159227714218e-05,
      "loss": 1.4842,
      "step": 228
    },
    {
      "epoch": 1.874239350912779,
      "grad_norm": 4.190274238586426,
      "learning_rate": 3.248963093702663e-05,
      "loss": 1.6199,
      "step": 231
    },
    {
      "epoch": 1.898580121703854,
      "grad_norm": 4.762985706329346,
      "learning_rate": 3.12696703292044e-05,
      "loss": 1.6452,
      "step": 234
    },
    {
      "epoch": 1.922920892494929,
      "grad_norm": 5.081422805786133,
      "learning_rate": 3.006254655373769e-05,
      "loss": 1.6196,
      "step": 237
    },
    {
      "epoch": 1.947261663286004,
      "grad_norm": 4.602591037750244,
      "learning_rate": 2.886908691296504e-05,
      "loss": 1.6198,
      "step": 240
    },
    {
      "epoch": 1.971602434077079,
      "grad_norm": 4.777091026306152,
      "learning_rate": 2.7690109344509563e-05,
      "loss": 1.6883,
      "step": 243
    },
    {
      "epoch": 1.995943204868154,
      "grad_norm": 5.713743209838867,
      "learning_rate": 2.6526421860705473e-05,
      "loss": 1.5588,
      "step": 246
    },
    {
      "epoch": 2.0121703853955375,
      "eval_loss": 0.48639488220214844,
      "eval_runtime": 13.4291,
      "eval_samples_per_second": 15.489,
      "eval_steps_per_second": 1.936,
      "step": 248
    },
    {
      "epoch": 2.020283975659229,
      "grad_norm": 3.395972967147827,
      "learning_rate": 2.537882199482665e-05,
      "loss": 1.1378,
      "step": 249
    },
    {
      "epoch": 2.0446247464503045,
      "grad_norm": 4.442982196807861,
      "learning_rate": 2.4248096254497288e-05,
      "loss": 1.253,
      "step": 252
    },
    {
      "epoch": 2.0689655172413794,
      "grad_norm": 5.2550764083862305,
      "learning_rate": 2.3135019582658802e-05,
      "loss": 1.0433,
      "step": 255
    },
    {
      "epoch": 2.0933062880324544,
      "grad_norm": 4.615274429321289,
      "learning_rate": 2.2040354826462668e-05,
      "loss": 1.1078,
      "step": 258
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 5.723622798919678,
      "learning_rate": 2.0964852214453013e-05,
      "loss": 0.9585,
      "step": 261
    },
    {
      "epoch": 2.1419878296146044,
      "grad_norm": 4.3719587326049805,
      "learning_rate": 1.9909248842397584e-05,
      "loss": 0.9587,
      "step": 264
    },
    {
      "epoch": 2.1663286004056794,
      "grad_norm": 6.265243053436279,
      "learning_rate": 1.887426816811903e-05,
      "loss": 0.9681,
      "step": 267
    },
    {
      "epoch": 2.1906693711967544,
      "grad_norm": 5.796363830566406,
      "learning_rate": 1.7860619515673033e-05,
      "loss": 1.0059,
      "step": 270
    },
    {
      "epoch": 2.2150101419878294,
      "grad_norm": 5.817225456237793,
      "learning_rate": 1.6868997589213136e-05,
      "loss": 1.0253,
      "step": 273
    },
    {
      "epoch": 2.239350912778905,
      "grad_norm": 4.450856685638428,
      "learning_rate": 1.5900081996875083e-05,
      "loss": 0.7533,
      "step": 276
    },
    {
      "epoch": 2.26369168356998,
      "grad_norm": 7.340899467468262,
      "learning_rate": 1.4954536785007456e-05,
      "loss": 0.9246,
      "step": 279
    },
    {
      "epoch": 2.26369168356998,
      "eval_loss": 0.5402039885520935,
      "eval_runtime": 13.4355,
      "eval_samples_per_second": 15.481,
      "eval_steps_per_second": 1.935,
      "step": 279
    },
    {
      "epoch": 2.288032454361055,
      "grad_norm": 5.454842567443848,
      "learning_rate": 1.4033009983067452e-05,
      "loss": 1.0822,
      "step": 282
    },
    {
      "epoch": 2.31237322515213,
      "grad_norm": 5.592954635620117,
      "learning_rate": 1.3136133159493802e-05,
      "loss": 0.9626,
      "step": 285
    },
    {
      "epoch": 2.336713995943205,
      "grad_norm": 5.249399662017822,
      "learning_rate": 1.22645209888614e-05,
      "loss": 0.8161,
      "step": 288
    },
    {
      "epoch": 2.36105476673428,
      "grad_norm": 5.653781890869141,
      "learning_rate": 1.1418770830614013e-05,
      "loss": 0.9639,
      "step": 291
    },
    {
      "epoch": 2.385395537525355,
      "grad_norm": 5.4389119148254395,
      "learning_rate": 1.0599462319663905e-05,
      "loss": 0.868,
      "step": 294
    },
    {
      "epoch": 2.40973630831643,
      "grad_norm": 5.680983543395996,
      "learning_rate": 9.807156969139136e-06,
      "loss": 1.1558,
      "step": 297
    },
    {
      "epoch": 2.4340770791075053,
      "grad_norm": 8.884838104248047,
      "learning_rate": 9.042397785550405e-06,
      "loss": 0.9113,
      "step": 300
    },
    {
      "epoch": 2.4584178498985803,
      "grad_norm": 6.362790107727051,
      "learning_rate": 8.305708896641594e-06,
      "loss": 1.0533,
      "step": 303
    },
    {
      "epoch": 2.4827586206896552,
      "grad_norm": 6.722659111022949,
      "learning_rate": 7.597595192178702e-06,
      "loss": 0.945,
      "step": 306
    },
    {
      "epoch": 2.5070993914807302,
      "grad_norm": 4.19068717956543,
      "learning_rate": 6.918541977923709e-06,
      "loss": 0.8369,
      "step": 309
    },
    {
      "epoch": 2.5152129817444218,
      "eval_loss": 0.5427043437957764,
      "eval_runtime": 13.4361,
      "eval_samples_per_second": 15.481,
      "eval_steps_per_second": 1.935,
      "step": 310
    }
  ],
  "logging_steps": 3,
  "max_steps": 370,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 31,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.118715981197476e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}