{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 75292,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 4.9669951654890296e-05,
      "loss": 1.5966,
      "step": 500
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.933791106624874e-05,
      "loss": 1.3931,
      "step": 1000
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.900587047760719e-05,
      "loss": 1.3138,
      "step": 1500
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.8673829888965626e-05,
      "loss": 1.2591,
      "step": 2000
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.834178930032408e-05,
      "loss": 1.2239,
      "step": 2500
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.800974871168252e-05,
      "loss": 1.1972,
      "step": 3000
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.767770812304096e-05,
      "loss": 1.1682,
      "step": 3500
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.734566753439941e-05,
      "loss": 1.1505,
      "step": 4000
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.701362694575785e-05,
      "loss": 1.1314,
      "step": 4500
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.66815863571163e-05,
      "loss": 1.1163,
      "step": 5000
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.634954576847474e-05,
      "loss": 1.1029,
      "step": 5500
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.6017505179833184e-05,
      "loss": 1.0916,
      "step": 6000
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.568546459119163e-05,
      "loss": 1.0832,
      "step": 6500
    },
    {
      "epoch": 0.37,
      "learning_rate": 4.535342400255007e-05,
      "loss": 1.0731,
      "step": 7000
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.502138341390852e-05,
      "loss": 1.0622,
      "step": 7500
    },
    {
      "epoch": 0.43,
      "learning_rate": 4.468934282526696e-05,
      "loss": 1.0573,
      "step": 8000
    },
    {
      "epoch": 0.45,
      "learning_rate": 4.435796631780269e-05,
      "loss": 1.049,
      "step": 8500
    },
    {
      "epoch": 0.48,
      "learning_rate": 4.402592572916113e-05,
      "loss": 1.042,
      "step": 9000
    },
    {
      "epoch": 0.5,
      "learning_rate": 4.369388514051958e-05,
      "loss": 1.0354,
      "step": 9500
    },
    {
      "epoch": 0.53,
      "learning_rate": 4.3361844551878024e-05,
      "loss": 1.0275,
      "step": 10000
    },
    {
      "epoch": 0.56,
      "learning_rate": 4.302980396323647e-05,
      "loss": 1.0232,
      "step": 10500
    },
    {
      "epoch": 0.58,
      "learning_rate": 4.2697763374594915e-05,
      "loss": 1.019,
      "step": 11000
    },
    {
      "epoch": 0.61,
      "learning_rate": 4.2365722785953354e-05,
      "loss": 1.0141,
      "step": 11500
    },
    {
      "epoch": 0.64,
      "learning_rate": 4.20336821973118e-05,
      "loss": 1.0089,
      "step": 12000
    },
    {
      "epoch": 0.66,
      "learning_rate": 4.1701641608670245e-05,
      "loss": 1.0028,
      "step": 12500
    },
    {
      "epoch": 0.69,
      "learning_rate": 4.136960102002869e-05,
      "loss": 0.9984,
      "step": 13000
    },
    {
      "epoch": 0.72,
      "learning_rate": 4.103822451256442e-05,
      "loss": 0.9936,
      "step": 13500
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.0706183923922865e-05,
      "loss": 0.9914,
      "step": 14000
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.0374143335281303e-05,
      "loss": 0.9851,
      "step": 14500
    },
    {
      "epoch": 0.8,
      "learning_rate": 4.0042102746639756e-05,
      "loss": 0.9828,
      "step": 15000
    },
    {
      "epoch": 0.82,
      "learning_rate": 3.9710062157998195e-05,
      "loss": 0.9813,
      "step": 15500
    },
    {
      "epoch": 0.85,
      "learning_rate": 3.937868565053392e-05,
      "loss": 0.976,
      "step": 16000
    },
    {
      "epoch": 0.88,
      "learning_rate": 3.904664506189237e-05,
      "loss": 0.9739,
      "step": 16500
    },
    {
      "epoch": 0.9,
      "learning_rate": 3.871460447325081e-05,
      "loss": 0.9719,
      "step": 17000
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.838256388460926e-05,
      "loss": 0.9664,
      "step": 17500
    },
    {
      "epoch": 0.96,
      "learning_rate": 3.805118737714499e-05,
      "loss": 0.9642,
      "step": 18000
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.7719146788503427e-05,
      "loss": 0.9603,
      "step": 18500
    },
    {
      "epoch": 1.01,
      "learning_rate": 3.738710619986187e-05,
      "loss": 0.9575,
      "step": 19000
    },
    {
      "epoch": 1.04,
      "learning_rate": 3.705506561122032e-05,
      "loss": 0.9539,
      "step": 19500
    },
    {
      "epoch": 1.06,
      "learning_rate": 3.672302502257876e-05,
      "loss": 0.9525,
      "step": 20000
    },
    {
      "epoch": 1.09,
      "learning_rate": 3.639164851511449e-05,
      "loss": 0.9518,
      "step": 20500
    },
    {
      "epoch": 1.12,
      "learning_rate": 3.605960792647293e-05,
      "loss": 0.9469,
      "step": 21000
    },
    {
      "epoch": 1.14,
      "learning_rate": 3.572756733783138e-05,
      "loss": 0.9466,
      "step": 21500
    },
    {
      "epoch": 1.17,
      "learning_rate": 3.5396190830367104e-05,
      "loss": 0.9435,
      "step": 22000
    },
    {
      "epoch": 1.2,
      "learning_rate": 3.506481432290283e-05,
      "loss": 0.9424,
      "step": 22500
    },
    {
      "epoch": 1.22,
      "learning_rate": 3.473277373426128e-05,
      "loss": 0.9373,
      "step": 23000
    },
    {
      "epoch": 1.25,
      "learning_rate": 3.4400733145619723e-05,
      "loss": 0.9374,
      "step": 23500
    },
    {
      "epoch": 1.28,
      "learning_rate": 3.406869255697817e-05,
      "loss": 0.934,
      "step": 24000
    },
    {
      "epoch": 1.3,
      "learning_rate": 3.3736651968336615e-05,
      "loss": 0.9335,
      "step": 24500
    },
    {
      "epoch": 1.33,
      "learning_rate": 3.3404611379695053e-05,
      "loss": 0.9324,
      "step": 25000
    },
    {
      "epoch": 1.35,
      "learning_rate": 3.30725707910535e-05,
      "loss": 0.9284,
      "step": 25500
    },
    {
      "epoch": 1.38,
      "learning_rate": 3.2740530202411945e-05,
      "loss": 0.9269,
      "step": 26000
    },
    {
      "epoch": 1.41,
      "learning_rate": 3.240848961377039e-05,
      "loss": 0.9257,
      "step": 26500
    },
    {
      "epoch": 1.43,
      "learning_rate": 3.2076449025128836e-05,
      "loss": 0.9203,
      "step": 27000
    },
    {
      "epoch": 1.46,
      "learning_rate": 3.1744408436487275e-05,
      "loss": 0.9233,
      "step": 27500
    },
    {
      "epoch": 1.49,
      "learning_rate": 3.141236784784572e-05,
      "loss": 0.9208,
      "step": 28000
    },
    {
      "epoch": 1.51,
      "learning_rate": 3.1080327259204166e-05,
      "loss": 0.919,
      "step": 28500
    },
    {
      "epoch": 1.54,
      "learning_rate": 3.074828667056261e-05,
      "loss": 0.9174,
      "step": 29000
    },
    {
      "epoch": 1.57,
      "learning_rate": 3.0416246081921057e-05,
      "loss": 0.9167,
      "step": 29500
    },
    {
      "epoch": 1.59,
      "learning_rate": 3.00842054932795e-05,
      "loss": 0.9138,
      "step": 30000
    },
    {
      "epoch": 1.62,
      "learning_rate": 2.9752164904637948e-05,
      "loss": 0.9133,
      "step": 30500
    },
    {
      "epoch": 1.65,
      "learning_rate": 2.942078839717367e-05,
      "loss": 0.9124,
      "step": 31000
    },
    {
      "epoch": 1.67,
      "learning_rate": 2.908874780853212e-05,
      "loss": 0.9107,
      "step": 31500
    },
    {
      "epoch": 1.7,
      "learning_rate": 2.875670721989056e-05,
      "loss": 0.9079,
      "step": 32000
    },
    {
      "epoch": 1.73,
      "learning_rate": 2.842533071242629e-05,
      "loss": 0.9069,
      "step": 32500
    },
    {
      "epoch": 1.75,
      "learning_rate": 2.809329012378473e-05,
      "loss": 0.9064,
      "step": 33000
    },
    {
      "epoch": 1.78,
      "learning_rate": 2.7761249535143173e-05,
      "loss": 0.9052,
      "step": 33500
    },
    {
      "epoch": 1.81,
      "learning_rate": 2.7429208946501622e-05,
      "loss": 0.9028,
      "step": 34000
    },
    {
      "epoch": 1.83,
      "learning_rate": 2.7097168357860064e-05,
      "loss": 0.9024,
      "step": 34500
    },
    {
      "epoch": 1.86,
      "learning_rate": 2.6765127769218513e-05,
      "loss": 0.9009,
      "step": 35000
    },
    {
      "epoch": 1.89,
      "learning_rate": 2.6433087180576956e-05,
      "loss": 0.9006,
      "step": 35500
    },
    {
      "epoch": 1.91,
      "learning_rate": 2.61010465919354e-05,
      "loss": 0.8988,
      "step": 36000
    },
    {
      "epoch": 1.94,
      "learning_rate": 2.5769006003293843e-05,
      "loss": 0.898,
      "step": 36500
    },
    {
      "epoch": 1.97,
      "learning_rate": 2.5437629495829575e-05,
      "loss": 0.8953,
      "step": 37000
    },
    {
      "epoch": 1.99,
      "learning_rate": 2.5105588907188017e-05,
      "loss": 0.8956,
      "step": 37500
    },
    {
      "epoch": 2.02,
      "learning_rate": 2.477354831854646e-05,
      "loss": 0.8945,
      "step": 38000
    },
    {
      "epoch": 2.05,
      "learning_rate": 2.4441507729904905e-05,
      "loss": 0.8926,
      "step": 38500
    },
    {
      "epoch": 2.07,
      "learning_rate": 2.410946714126335e-05,
      "loss": 0.8892,
      "step": 39000
    },
    {
      "epoch": 2.1,
      "learning_rate": 2.3777426552621793e-05,
      "loss": 0.8905,
      "step": 39500
    },
    {
      "epoch": 2.13,
      "learning_rate": 2.3445385963980238e-05,
      "loss": 0.8877,
      "step": 40000
    },
    {
      "epoch": 2.15,
      "learning_rate": 2.3113345375338684e-05,
      "loss": 0.8876,
      "step": 40500
    },
    {
      "epoch": 2.18,
      "learning_rate": 2.2782632949051695e-05,
      "loss": 0.8872,
      "step": 41000
    },
    {
      "epoch": 2.2,
      "learning_rate": 2.245059236041014e-05,
      "loss": 0.8855,
      "step": 41500
    },
    {
      "epoch": 2.23,
      "learning_rate": 2.2118551771768582e-05,
      "loss": 0.8839,
      "step": 42000
    },
    {
      "epoch": 2.26,
      "learning_rate": 2.1786511183127025e-05,
      "loss": 0.8853,
      "step": 42500
    },
    {
      "epoch": 2.28,
      "learning_rate": 2.145447059448547e-05,
      "loss": 0.8834,
      "step": 43000
    },
    {
      "epoch": 2.31,
      "learning_rate": 2.1122430005843916e-05,
      "loss": 0.8823,
      "step": 43500
    },
    {
      "epoch": 2.34,
      "learning_rate": 2.0791053498379644e-05,
      "loss": 0.8832,
      "step": 44000
    },
    {
      "epoch": 2.36,
      "learning_rate": 2.0459012909738086e-05,
      "loss": 0.8812,
      "step": 44500
    },
    {
      "epoch": 2.39,
      "learning_rate": 2.0126972321096532e-05,
      "loss": 0.8805,
      "step": 45000
    },
    {
      "epoch": 2.42,
      "learning_rate": 1.9794931732454977e-05,
      "loss": 0.8795,
      "step": 45500
    },
    {
      "epoch": 2.44,
      "learning_rate": 1.9463555224990705e-05,
      "loss": 0.8784,
      "step": 46000
    },
    {
      "epoch": 2.47,
      "learning_rate": 1.913217871752643e-05,
      "loss": 0.8777,
      "step": 46500
    },
    {
      "epoch": 2.5,
      "learning_rate": 1.8800138128884876e-05,
      "loss": 0.876,
      "step": 47000
    },
    {
      "epoch": 2.52,
      "learning_rate": 1.846809754024332e-05,
      "loss": 0.877,
      "step": 47500
    },
    {
      "epoch": 2.55,
      "learning_rate": 1.8136056951601764e-05,
      "loss": 0.876,
      "step": 48000
    },
    {
      "epoch": 2.58,
      "learning_rate": 1.780401636296021e-05,
      "loss": 0.8765,
      "step": 48500
    },
    {
      "epoch": 2.6,
      "learning_rate": 1.7472639855495937e-05,
      "loss": 0.8752,
      "step": 49000
    },
    {
      "epoch": 2.63,
      "learning_rate": 1.7140599266854383e-05,
      "loss": 0.873,
      "step": 49500
    },
    {
      "epoch": 2.66,
      "learning_rate": 1.6808558678212825e-05,
      "loss": 0.8713,
      "step": 50000
    },
    {
      "epoch": 2.68,
      "learning_rate": 1.6476518089571267e-05,
      "loss": 0.8736,
      "step": 50500
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.6144477500929713e-05,
      "loss": 0.8715,
      "step": 51000
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.581243691228816e-05,
      "loss": 0.8708,
      "step": 51500
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.5481060404823887e-05,
      "loss": 0.8694,
      "step": 52000
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.514901981618233e-05,
      "loss": 0.8698,
      "step": 52500
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.4816979227540776e-05,
      "loss": 0.8686,
      "step": 53000
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.448493863889922e-05,
      "loss": 0.8696,
      "step": 53500
    },
    {
      "epoch": 2.87,
      "learning_rate": 1.4152898050257666e-05,
      "loss": 0.8685,
      "step": 54000
    },
    {
      "epoch": 2.9,
      "learning_rate": 1.3820857461616108e-05,
      "loss": 0.8664,
      "step": 54500
    },
    {
      "epoch": 2.92,
      "learning_rate": 1.3489480954151836e-05,
      "loss": 0.8666,
      "step": 55000
    },
    {
      "epoch": 2.95,
      "learning_rate": 1.3157440365510282e-05,
      "loss": 0.8664,
      "step": 55500
    },
    {
      "epoch": 2.98,
      "learning_rate": 1.2825399776868724e-05,
      "loss": 0.8687,
      "step": 56000
    },
    {
      "epoch": 3.0,
      "learning_rate": 1.249335918822717e-05,
      "loss": 0.8643,
      "step": 56500
    },
    {
      "epoch": 3.03,
      "learning_rate": 1.2161318599585613e-05,
      "loss": 0.8626,
      "step": 57000
    },
    {
      "epoch": 3.05,
      "learning_rate": 1.1829278010944059e-05,
      "loss": 0.8628,
      "step": 57500
    },
    {
      "epoch": 3.08,
      "learning_rate": 1.1497237422302503e-05,
      "loss": 0.8622,
      "step": 58000
    },
    {
      "epoch": 3.11,
      "learning_rate": 1.1165196833660947e-05,
      "loss": 0.8618,
      "step": 58500
    },
    {
      "epoch": 3.13,
      "learning_rate": 1.0833156245019392e-05,
      "loss": 0.862,
      "step": 59000
    },
    {
      "epoch": 3.16,
      "learning_rate": 1.0501115656377836e-05,
      "loss": 0.8605,
      "step": 59500
    },
    {
      "epoch": 3.19,
      "learning_rate": 1.016907506773628e-05,
      "loss": 0.8595,
      "step": 60000
    },
    {
      "epoch": 3.21,
      "learning_rate": 9.837698560272008e-06,
      "loss": 0.8604,
      "step": 60500
    },
    {
      "epoch": 3.24,
      "learning_rate": 9.505657971630452e-06,
      "loss": 0.8595,
      "step": 61000
    },
    {
      "epoch": 3.27,
      "learning_rate": 9.173617382988898e-06,
      "loss": 0.8602,
      "step": 61500
    },
    {
      "epoch": 3.29,
      "learning_rate": 8.841576794347342e-06,
      "loss": 0.8583,
      "step": 62000
    },
    {
      "epoch": 3.32,
      "learning_rate": 8.510200286883068e-06,
      "loss": 0.8575,
      "step": 62500
    },
    {
      "epoch": 3.35,
      "learning_rate": 8.178159698241514e-06,
      "loss": 0.8583,
      "step": 63000
    },
    {
      "epoch": 3.37,
      "learning_rate": 7.846119109599958e-06,
      "loss": 0.8568,
      "step": 63500
    },
    {
      "epoch": 3.4,
      "learning_rate": 7.514078520958402e-06,
      "loss": 0.8555,
      "step": 64000
    },
    {
      "epoch": 3.43,
      "learning_rate": 7.182037932316847e-06,
      "loss": 0.8562,
      "step": 64500
    },
    {
      "epoch": 3.45,
      "learning_rate": 6.849997343675291e-06,
      "loss": 0.8553,
      "step": 65000
    },
    {
      "epoch": 3.48,
      "learning_rate": 6.517956755033736e-06,
      "loss": 0.8537,
      "step": 65500
    },
    {
      "epoch": 3.51,
      "learning_rate": 6.186580247569463e-06,
      "loss": 0.8556,
      "step": 66000
    },
    {
      "epoch": 3.53,
      "learning_rate": 5.854539658927908e-06,
      "loss": 0.8556,
      "step": 66500
    },
    {
      "epoch": 3.56,
      "learning_rate": 5.5224990702863525e-06,
      "loss": 0.8559,
      "step": 67000
    },
    {
      "epoch": 3.59,
      "learning_rate": 5.190458481644796e-06,
      "loss": 0.8538,
      "step": 67500
    },
    {
      "epoch": 3.61,
      "learning_rate": 4.85841789300324e-06,
      "loss": 0.8538,
      "step": 68000
    },
    {
      "epoch": 3.64,
      "learning_rate": 4.526377304361686e-06,
      "loss": 0.8523,
      "step": 68500
    },
    {
      "epoch": 3.67,
      "learning_rate": 4.19433671572013e-06,
      "loss": 0.8529,
      "step": 69000
    },
    {
      "epoch": 3.69,
      "learning_rate": 3.862296127078574e-06,
      "loss": 0.8538,
      "step": 69500
    },
    {
      "epoch": 3.72,
      "learning_rate": 3.5302555384370188e-06,
      "loss": 0.8541,
      "step": 70000
    },
    {
      "epoch": 3.75,
      "learning_rate": 3.198214949795463e-06,
      "loss": 0.8521,
      "step": 70500
    },
    {
      "epoch": 3.77,
      "learning_rate": 2.8668384423311904e-06,
      "loss": 0.8529,
      "step": 71000
    },
    {
      "epoch": 3.8,
      "learning_rate": 2.534797853689635e-06,
      "loss": 0.8522,
      "step": 71500
    },
    {
      "epoch": 3.83,
      "learning_rate": 2.20275726504808e-06,
      "loss": 0.855,
      "step": 72000
    },
    {
      "epoch": 3.85,
      "learning_rate": 1.870716676406524e-06,
      "loss": 0.8542,
      "step": 72500
    },
    {
      "epoch": 3.88,
      "learning_rate": 1.5386760877649683e-06,
      "loss": 0.8516,
      "step": 73000
    },
    {
      "epoch": 3.9,
      "learning_rate": 1.206635499123413e-06,
      "loss": 0.8521,
      "step": 73500
    },
    {
      "epoch": 3.93,
      "learning_rate": 8.745949104818574e-07,
      "loss": 0.8514,
      "step": 74000
    },
    {
      "epoch": 3.96,
      "learning_rate": 5.425543218403018e-07,
      "loss": 0.852,
      "step": 74500
    },
    {
      "epoch": 3.98,
      "learning_rate": 2.1184189555331246e-07,
      "loss": 0.8498,
      "step": 75000
    },
    {
      "epoch": 4.0,
      "step": 75292,
      "total_flos": 5.54885793210119e+18,
      "train_loss": 0.9333500321268725,
      "train_runtime": 60105.9393,
      "train_samples_per_second": 350.725,
      "train_steps_per_second": 1.253
    }
  ],
  "logging_steps": 500,
  "max_steps": 75292,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 5.54885793210119e+18,
  "trial_name": null,
  "trial_params": null
}