{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 10000,
  "global_step": 249229,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 4.989969064595212e-05,
      "loss": 7.3065,
      "step": 500
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.9799381291904234e-05,
      "loss": 6.5187,
      "step": 1000
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.9699071937856354e-05,
      "loss": 6.1861,
      "step": 1500
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.9598762583808466e-05,
      "loss": 5.9496,
      "step": 2000
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.9498453229760586e-05,
      "loss": 5.7513,
      "step": 2500
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.93981438757127e-05,
      "loss": 5.5881,
      "step": 3000
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.929783452166482e-05,
      "loss": 5.4459,
      "step": 3500
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.919752516761693e-05,
      "loss": 5.3082,
      "step": 4000
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.909721581356905e-05,
      "loss": 5.1931,
      "step": 4500
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.899690645952116e-05,
      "loss": 5.0948,
      "step": 5000
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.889659710547328e-05,
      "loss": 5.0116,
      "step": 5500
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.8796287751425394e-05,
      "loss": 4.9386,
      "step": 6000
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.8695978397377514e-05,
      "loss": 4.8745,
      "step": 6500
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.859566904332963e-05,
      "loss": 4.8234,
      "step": 7000
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.849535968928175e-05,
      "loss": 4.7714,
      "step": 7500
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.8395050335233865e-05,
      "loss": 4.7328,
      "step": 8000
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.8294740981185984e-05,
      "loss": 4.6888,
      "step": 8500
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.81944316271381e-05,
      "loss": 4.6569,
      "step": 9000
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.8094122273090216e-05,
      "loss": 4.6302,
      "step": 9500
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.799381291904233e-05,
      "loss": 4.586,
      "step": 10000
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.2821468712144023,
      "eval_loss": 4.497698783874512,
      "eval_runtime": 4655.4137,
      "eval_samples_per_second": 90.14,
      "eval_steps_per_second": 1.408,
      "step": 10000
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.789350356499445e-05,
      "loss": 4.5693,
      "step": 10500
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.779319421094656e-05,
      "loss": 4.5425,
      "step": 11000
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.769288485689868e-05,
      "loss": 4.5203,
      "step": 11500
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.759257550285079e-05,
      "loss": 4.4994,
      "step": 12000
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.749226614880291e-05,
      "loss": 4.4772,
      "step": 12500
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.7391956794755025e-05,
      "loss": 4.4553,
      "step": 13000
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.7291647440707144e-05,
      "loss": 4.4413,
      "step": 13500
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.719133808665926e-05,
      "loss": 4.4206,
      "step": 14000
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.7091028732611376e-05,
      "loss": 4.4077,
      "step": 14500
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.699071937856349e-05,
      "loss": 4.3925,
      "step": 15000
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.689041002451561e-05,
      "loss": 4.3732,
      "step": 15500
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.679010067046772e-05,
      "loss": 4.3648,
      "step": 16000
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.668979131641984e-05,
      "loss": 4.3451,
      "step": 16500
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.658948196237195e-05,
      "loss": 4.3398,
      "step": 17000
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.648917260832407e-05,
      "loss": 4.3228,
      "step": 17500
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.6388863254276185e-05,
      "loss": 4.317,
      "step": 18000
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.6288553900228304e-05,
      "loss": 4.3033,
      "step": 18500
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.618824454618042e-05,
      "loss": 4.2927,
      "step": 19000
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.6087935192132536e-05,
      "loss": 4.2853,
      "step": 19500
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.5987625838084656e-05,
      "loss": 4.2706,
      "step": 20000
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.3058443984869302,
      "eval_loss": 4.192770004272461,
      "eval_runtime": 4658.7505,
      "eval_samples_per_second": 90.075,
      "eval_steps_per_second": 1.407,
      "step": 20000
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.5887316484036775e-05,
      "loss": 4.2613,
      "step": 20500
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.578700712998889e-05,
      "loss": 4.255,
      "step": 21000
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.568669777594101e-05,
      "loss": 4.2422,
      "step": 21500
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.558638842189312e-05,
      "loss": 4.2364,
      "step": 22000
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.548607906784524e-05,
      "loss": 4.2239,
      "step": 22500
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.538576971379735e-05,
      "loss": 4.2258,
      "step": 23000
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.528546035974947e-05,
      "loss": 4.2067,
      "step": 23500
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.5185151005701584e-05,
      "loss": 4.2101,
      "step": 24000
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.50848416516537e-05,
      "loss": 4.1927,
      "step": 24500
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.498453229760582e-05,
      "loss": 4.1868,
      "step": 25000
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.4884222943557935e-05,
      "loss": 4.1779,
      "step": 25500
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.4783913589510054e-05,
      "loss": 4.1711,
      "step": 26000
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.468360423546217e-05,
      "loss": 4.166,
      "step": 26500
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.4583294881414286e-05,
      "loss": 4.1639,
      "step": 27000
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.44829855273664e-05,
      "loss": 4.1586,
      "step": 27500
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.438267617331852e-05,
      "loss": 4.1508,
      "step": 28000
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.428236681927063e-05,
      "loss": 4.1471,
      "step": 28500
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.418205746522275e-05,
      "loss": 4.1396,
      "step": 29000
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.408174811117486e-05,
      "loss": 4.1354,
      "step": 29500
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.398143875712698e-05,
      "loss": 4.1189,
      "step": 30000
    },
    {
      "epoch": 0.12,
      "eval_accuracy": 0.31791789096235645,
      "eval_loss": 4.0469441413879395,
      "eval_runtime": 4672.6798,
      "eval_samples_per_second": 89.807,
      "eval_steps_per_second": 1.403,
      "step": 30000
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.3881129403079095e-05,
      "loss": 4.1212,
      "step": 30500
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.3780820049031214e-05,
      "loss": 4.1141,
      "step": 31000
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.368051069498333e-05,
      "loss": 4.1145,
      "step": 31500
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.3580201340935446e-05,
      "loss": 4.0995,
      "step": 32000
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.347989198688756e-05,
      "loss": 4.1028,
      "step": 32500
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.337958263283968e-05,
      "loss": 4.0968,
      "step": 33000
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.32792732787918e-05,
      "loss": 4.0937,
      "step": 33500
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.317896392474392e-05,
      "loss": 4.0851,
      "step": 34000
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.307865457069603e-05,
      "loss": 4.0801,
      "step": 34500
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.297834521664815e-05,
      "loss": 4.0803,
      "step": 35000
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.287803586260026e-05,
      "loss": 4.0716,
      "step": 35500
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.277772650855238e-05,
      "loss": 4.0719,
      "step": 36000
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.2677417154504494e-05,
      "loss": 4.0665,
      "step": 36500
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.257710780045661e-05,
      "loss": 4.0619,
      "step": 37000
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.2476798446408726e-05,
      "loss": 4.0561,
      "step": 37500
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.2376489092360845e-05,
      "loss": 4.0577,
      "step": 38000
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.227617973831296e-05,
      "loss": 4.0542,
      "step": 38500
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.217587038426508e-05,
      "loss": 4.0338,
      "step": 39000
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.207556103021719e-05,
      "loss": 4.0415,
      "step": 39500
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.197525167616931e-05,
      "loss": 4.0314,
      "step": 40000
    },
    {
      "epoch": 0.16,
      "eval_accuracy": 0.3253092461027638,
      "eval_loss": 3.96097469329834,
      "eval_runtime": 4672.6679,
      "eval_samples_per_second": 89.807,
      "eval_steps_per_second": 1.403,
      "step": 40000
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.187494232212142e-05,
      "loss": 4.0337,
      "step": 40500
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.177463296807354e-05,
      "loss": 4.023,
      "step": 41000
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.1674323614025654e-05,
      "loss": 4.0296,
      "step": 41500
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.157401425997777e-05,
      "loss": 4.0172,
      "step": 42000
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.1473704905929886e-05,
      "loss": 4.0191,
      "step": 42500
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.1373395551882005e-05,
      "loss": 4.0218,
      "step": 43000
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.127308619783412e-05,
      "loss": 4.0075,
      "step": 43500
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.117277684378624e-05,
      "loss": 4.0105,
      "step": 44000
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.1072467489738356e-05,
      "loss": 4.0073,
      "step": 44500
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.097215813569047e-05,
      "loss": 3.9932,
      "step": 45000
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.087184878164259e-05,
      "loss": 3.9959,
      "step": 45500
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.07715394275947e-05,
      "loss": 3.9903,
      "step": 46000
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.067123007354682e-05,
      "loss": 3.9957,
      "step": 46500
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.057092071949894e-05,
      "loss": 3.9921,
      "step": 47000
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.047061136545105e-05,
      "loss": 3.9841,
      "step": 47500
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.037030201140317e-05,
      "loss": 3.987,
      "step": 48000
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.0269992657355284e-05,
      "loss": 3.9793,
      "step": 48500
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.0169683303307404e-05,
      "loss": 3.9824,
      "step": 49000
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.006937394925952e-05,
      "loss": 3.9745,
      "step": 49500
    },
    {
      "epoch": 0.2,
      "learning_rate": 3.9969064595211636e-05,
      "loss": 3.9704,
      "step": 50000
    },
    {
      "epoch": 0.2,
      "eval_accuracy": 0.33107534589784043,
      "eval_loss": 3.897735834121704,
      "eval_runtime": 4799.5493,
      "eval_samples_per_second": 87.433,
      "eval_steps_per_second": 1.366,
      "step": 50000
    },
    {
      "epoch": 0.2,
      "learning_rate": 3.9868755241163755e-05,
      "loss": 3.9677,
      "step": 50500
    },
    {
      "epoch": 0.2,
      "learning_rate": 3.976844588711587e-05,
      "loss": 3.9727,
      "step": 51000
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.966813653306799e-05,
      "loss": 3.9714,
      "step": 51500
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.95678271790201e-05,
      "loss": 3.9605,
      "step": 52000
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.946751782497222e-05,
      "loss": 3.9584,
      "step": 52500
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.936720847092433e-05,
      "loss": 3.9582,
      "step": 53000
    },
    {
      "epoch": 0.21,
      "learning_rate": 3.926689911687645e-05,
      "loss": 3.9571,
      "step": 53500
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.9166589762828564e-05,
      "loss": 3.9517,
      "step": 54000
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.906628040878068e-05,
      "loss": 3.9453,
      "step": 54500
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.8965971054732796e-05,
      "loss": 3.952,
      "step": 55000
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.8865661700684915e-05,
      "loss": 3.9473,
      "step": 55500
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.876535234663703e-05,
      "loss": 3.9465,
      "step": 56000
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.866504299258915e-05,
      "loss": 3.9409,
      "step": 56500
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.856473363854126e-05,
      "loss": 3.9323,
      "step": 57000
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.846442428449338e-05,
      "loss": 3.9435,
      "step": 57500
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.836411493044549e-05,
      "loss": 3.9361,
      "step": 58000
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.826380557639761e-05,
      "loss": 3.9332,
      "step": 58500
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.8163496222349724e-05,
      "loss": 3.9293,
      "step": 59000
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.806318686830184e-05,
      "loss": 3.9302,
      "step": 59500
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.7962877514253956e-05,
      "loss": 3.923,
      "step": 60000
    },
    {
      "epoch": 0.24,
      "eval_accuracy": 0.33527885650874006,
      "eval_loss": 3.848573923110962,
      "eval_runtime": 4791.4499,
      "eval_samples_per_second": 87.581,
      "eval_steps_per_second": 1.368,
      "step": 60000
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.786256816020608e-05,
      "loss": 3.9301,
      "step": 60500
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.7762258806158194e-05,
      "loss": 3.9248,
      "step": 61000
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.7661949452110314e-05,
      "loss": 3.915,
      "step": 61500
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.7561640098062426e-05,
      "loss": 3.9202,
      "step": 62000
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.7461330744014546e-05,
      "loss": 3.9194,
      "step": 62500
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.736102138996666e-05,
      "loss": 3.9193,
      "step": 63000
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.726071203591878e-05,
      "loss": 3.9076,
      "step": 63500
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.716040268187089e-05,
      "loss": 3.9139,
      "step": 64000
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.706009332782301e-05,
      "loss": 3.9101,
      "step": 64500
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.695978397377512e-05,
      "loss": 3.9046,
      "step": 65000
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.685947461972724e-05,
      "loss": 3.9086,
      "step": 65500
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.6759165265679354e-05,
      "loss": 3.9011,
      "step": 66000
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.6658855911631474e-05,
      "loss": 3.901,
      "step": 66500
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.6558546557583586e-05,
      "loss": 3.9051,
      "step": 67000
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.6458237203535706e-05,
      "loss": 3.8936,
      "step": 67500
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.6357927849487825e-05,
      "loss": 3.8905,
      "step": 68000
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.625761849543994e-05,
      "loss": 3.8844,
      "step": 68500
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.615730914139206e-05,
      "loss": 3.8948,
      "step": 69000
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.605699978734417e-05,
      "loss": 3.8902,
      "step": 69500
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.595669043329629e-05,
      "loss": 3.888,
      "step": 70000
    },
    {
      "epoch": 0.28,
      "eval_accuracy": 0.33901861842582626,
      "eval_loss": 3.808432102203369,
      "eval_runtime": 4676.4302,
      "eval_samples_per_second": 89.735,
      "eval_steps_per_second": 1.402,
      "step": 70000
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.58563810792484e-05,
      "loss": 3.8876,
      "step": 70500
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.575607172520052e-05,
      "loss": 3.8866,
      "step": 71000
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.5655762371152634e-05,
      "loss": 3.8802,
      "step": 71500
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.555545301710475e-05,
      "loss": 3.8877,
      "step": 72000
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.5455143663056866e-05,
      "loss": 3.881,
      "step": 72500
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.5354834309008985e-05,
      "loss": 3.8813,
      "step": 73000
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.52545249549611e-05,
      "loss": 3.8704,
      "step": 73500
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.5154215600913224e-05,
      "loss": 3.8745,
      "step": 74000
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.5053906246865336e-05,
      "loss": 3.8739,
      "step": 74500
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.4953596892817456e-05,
      "loss": 3.8779,
      "step": 75000
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.485328753876957e-05,
      "loss": 3.8753,
      "step": 75500
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.475297818472169e-05,
      "loss": 3.8621,
      "step": 76000
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.46526688306738e-05,
      "loss": 3.8634,
      "step": 76500
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.455235947662592e-05,
      "loss": 3.8675,
      "step": 77000
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.445205012257803e-05,
      "loss": 3.8672,
      "step": 77500
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.435174076853015e-05,
      "loss": 3.8639,
      "step": 78000
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.4251431414482264e-05,
      "loss": 3.8547,
      "step": 78500
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.4151122060434384e-05,
      "loss": 3.8573,
      "step": 79000
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.4050812706386496e-05,
      "loss": 3.8516,
      "step": 79500
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.3950503352338616e-05,
      "loss": 3.8529,
      "step": 80000
    },
    {
      "epoch": 0.32,
      "eval_accuracy": 0.3423139332553484,
      "eval_loss": 3.7777137756347656,
      "eval_runtime": 4683.7605,
      "eval_samples_per_second": 89.594,
      "eval_steps_per_second": 1.4,
      "step": 80000
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.385019399829073e-05,
      "loss": 3.8512,
      "step": 80500
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.374988464424285e-05,
      "loss": 3.8508,
      "step": 81000
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.364957529019496e-05,
      "loss": 3.8484,
      "step": 81500
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.354926593614708e-05,
      "loss": 3.8489,
      "step": 82000
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.344895658209919e-05,
      "loss": 3.8471,
      "step": 82500
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.334864722805131e-05,
      "loss": 3.8505,
      "step": 83000
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.3248337874003424e-05,
      "loss": 3.839,
      "step": 83500
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.3148028519955544e-05,
      "loss": 3.8493,
      "step": 84000
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.3047719165907656e-05,
      "loss": 3.839,
      "step": 84500
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.2947409811859776e-05,
      "loss": 3.8447,
      "step": 85000
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.284710045781189e-05,
      "loss": 3.8401,
      "step": 85500
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.274679110376401e-05,
      "loss": 3.8343,
      "step": 86000
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.264648174971612e-05,
      "loss": 3.8379,
      "step": 86500
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.2546172395668246e-05,
      "loss": 3.8378,
      "step": 87000
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.244586304162036e-05,
      "loss": 3.8355,
      "step": 87500
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.234555368757248e-05,
      "loss": 3.8363,
      "step": 88000
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.224524433352459e-05,
      "loss": 3.8349,
      "step": 88500
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.214493497947671e-05,
      "loss": 3.8364,
      "step": 89000
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.204462562542882e-05,
      "loss": 3.8271,
      "step": 89500
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.194431627138094e-05,
      "loss": 3.832,
      "step": 90000
    },
    {
      "epoch": 0.36,
      "eval_accuracy": 0.3446369162758726,
      "eval_loss": 3.752626419067383,
      "eval_runtime": 4676.2979,
      "eval_samples_per_second": 89.737,
      "eval_steps_per_second": 1.402,
      "step": 90000
    },
    {
      "epoch": 0.36,
      "learning_rate": 3.1844006917333055e-05,
      "loss": 3.8362,
      "step": 90500
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.1743697563285174e-05,
      "loss": 3.8261,
      "step": 91000
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.164338820923729e-05,
      "loss": 3.8246,
      "step": 91500
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.1543078855189406e-05,
      "loss": 3.8274,
      "step": 92000
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.1442769501141526e-05,
      "loss": 3.8221,
      "step": 92500
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.134246014709364e-05,
      "loss": 3.823,
      "step": 93000
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.124215079304576e-05,
      "loss": 3.8257,
      "step": 93500
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.114184143899787e-05,
      "loss": 3.8206,
      "step": 94000
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.104153208494999e-05,
      "loss": 3.8241,
      "step": 94500
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.09412227309021e-05,
      "loss": 3.8211,
      "step": 95000
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.084091337685422e-05,
      "loss": 3.8144,
      "step": 95500
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.0740604022806334e-05,
      "loss": 3.8205,
      "step": 96000
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.0640294668758454e-05,
      "loss": 3.8178,
      "step": 96500
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.0539985314710566e-05,
      "loss": 3.8113,
      "step": 97000
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.0439675960662682e-05,
      "loss": 3.8081,
      "step": 97500
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.03393666066148e-05,
      "loss": 3.8163,
      "step": 98000
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.0239057252566914e-05,
      "loss": 3.8163,
      "step": 98500
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.013874789851903e-05,
      "loss": 3.808,
      "step": 99000
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.003843854447115e-05,
      "loss": 3.8116,
      "step": 99500
    },
    {
      "epoch": 0.4,
      "learning_rate": 2.9938129190423266e-05,
      "loss": 3.8102,
      "step": 100000
    },
    {
      "epoch": 0.4,
      "eval_accuracy": 0.3469910693173023,
      "eval_loss": 3.7277164459228516,
      "eval_runtime": 4664.5454,
      "eval_samples_per_second": 89.964,
      "eval_steps_per_second": 1.406,
      "step": 100000
    },
    {
      "epoch": 0.4,
      "learning_rate": 2.9837819836375385e-05,
      "loss": 3.8106,
      "step": 100500
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.97375104823275e-05,
      "loss": 3.8051,
      "step": 101000
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.9637201128279617e-05,
      "loss": 3.8013,
      "step": 101500
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.9536891774231733e-05,
      "loss": 3.8023,
      "step": 102000
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.943658242018385e-05,
      "loss": 3.8013,
      "step": 102500
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.9336273066135965e-05,
      "loss": 3.7951,
      "step": 103000
    },
    {
      "epoch": 0.42,
      "learning_rate": 2.923596371208808e-05,
      "loss": 3.806,
      "step": 103500
    },
    {
      "epoch": 0.42,
      "learning_rate": 2.9135654358040197e-05,
      "loss": 3.8028,
      "step": 104000
    },
    {
      "epoch": 0.42,
      "learning_rate": 2.9035345003992316e-05,
      "loss": 3.8002,
      "step": 104500
    },
    {
      "epoch": 0.42,
      "learning_rate": 2.8935035649944432e-05,
      "loss": 3.7978,
      "step": 105000
    },
    {
      "epoch": 0.42,
      "learning_rate": 2.883472629589655e-05,
      "loss": 3.7911,
      "step": 105500
    },
    {
      "epoch": 0.43,
      "learning_rate": 2.8734416941848664e-05,
      "loss": 3.792,
      "step": 106000
    },
    {
      "epoch": 0.43,
      "learning_rate": 2.863410758780078e-05,
      "loss": 3.7943,
      "step": 106500
    },
    {
      "epoch": 0.43,
      "learning_rate": 2.8533798233752896e-05,
      "loss": 3.788,
      "step": 107000
    },
    {
      "epoch": 0.43,
      "learning_rate": 2.8433488879705012e-05,
      "loss": 3.7955,
      "step": 107500
    },
    {
      "epoch": 0.43,
      "learning_rate": 2.833317952565713e-05,
      "loss": 3.7807,
      "step": 108000
    },
    {
      "epoch": 0.44,
      "learning_rate": 2.8232870171609244e-05,
      "loss": 3.7892,
      "step": 108500
    },
    {
      "epoch": 0.44,
      "learning_rate": 2.813256081756136e-05,
      "loss": 3.7892,
      "step": 109000
    },
    {
      "epoch": 0.44,
      "learning_rate": 2.8032251463513476e-05,
      "loss": 3.7916,
      "step": 109500
    },
    {
      "epoch": 0.44,
      "learning_rate": 2.7931942109465592e-05,
      "loss": 3.7876,
      "step": 110000
    },
    {
      "epoch": 0.44,
      "eval_accuracy": 0.34899459281783596,
      "eval_loss": 3.7072978019714355,
      "eval_runtime": 4653.1421,
      "eval_samples_per_second": 90.184,
      "eval_steps_per_second": 1.409,
      "step": 110000
    },
    {
      "epoch": 0.44,
      "learning_rate": 2.783163275541771e-05,
      "loss": 3.7845,
      "step": 110500
    },
    {
      "epoch": 0.45,
      "learning_rate": 2.7731323401369824e-05,
      "loss": 3.7859,
      "step": 111000
    },
    {
      "epoch": 0.45,
      "learning_rate": 2.763101404732194e-05,
      "loss": 3.7828,
      "step": 111500
    },
    {
      "epoch": 0.45,
      "learning_rate": 2.7530704693274056e-05,
      "loss": 3.7866,
      "step": 112000
    },
    {
      "epoch": 0.45,
      "learning_rate": 2.7430395339226172e-05,
      "loss": 3.7789,
      "step": 112500
    },
    {
      "epoch": 0.45,
      "learning_rate": 2.733008598517829e-05,
      "loss": 3.7793,
      "step": 113000
    },
    {
      "epoch": 0.46,
      "learning_rate": 2.7229776631130404e-05,
      "loss": 3.7869,
      "step": 113500
    },
    {
      "epoch": 0.46,
      "learning_rate": 2.7129467277082527e-05,
      "loss": 3.7817,
      "step": 114000
    },
    {
      "epoch": 0.46,
      "learning_rate": 2.7029157923034643e-05,
      "loss": 3.7724,
      "step": 114500
    },
    {
      "epoch": 0.46,
      "learning_rate": 2.692884856898676e-05,
      "loss": 3.7779,
      "step": 115000
    },
    {
      "epoch": 0.46,
      "learning_rate": 2.6828539214938875e-05,
      "loss": 3.7778,
      "step": 115500
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.672822986089099e-05,
      "loss": 3.784,
      "step": 116000
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.6627920506843107e-05,
      "loss": 3.7777,
      "step": 116500
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.6527611152795223e-05,
      "loss": 3.7763,
      "step": 117000
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.642730179874734e-05,
      "loss": 3.7693,
      "step": 117500
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.6326992444699455e-05,
      "loss": 3.7743,
      "step": 118000
    },
    {
      "epoch": 0.48,
      "learning_rate": 2.622668309065157e-05,
      "loss": 3.771,
      "step": 118500
    },
    {
      "epoch": 0.48,
      "learning_rate": 2.6126373736603687e-05,
      "loss": 3.7796,
      "step": 119000
    },
    {
      "epoch": 0.48,
      "learning_rate": 2.6026064382555803e-05,
      "loss": 3.765,
      "step": 119500
    },
    {
      "epoch": 0.48,
      "learning_rate": 2.592575502850792e-05,
      "loss": 3.7686,
      "step": 120000
    },
    {
      "epoch": 0.48,
      "eval_accuracy": 0.35060705962607897,
      "eval_loss": 3.6921615600585938,
      "eval_runtime": 4753.6745,
      "eval_samples_per_second": 88.277,
      "eval_steps_per_second": 1.379,
      "step": 120000
    },
    {
      "epoch": 0.48,
      "learning_rate": 2.5825445674460035e-05,
      "loss": 3.7761,
      "step": 120500
    },
    {
      "epoch": 0.49,
      "learning_rate": 2.572513632041215e-05,
      "loss": 3.772,
      "step": 121000
    },
    {
      "epoch": 0.49,
      "learning_rate": 2.5624826966364267e-05,
      "loss": 3.7714,
      "step": 121500
    },
    {
      "epoch": 0.49,
      "learning_rate": 2.5524517612316383e-05,
      "loss": 3.7655,
      "step": 122000
    },
    {
      "epoch": 0.49,
      "learning_rate": 2.54242082582685e-05,
      "loss": 3.7657,
      "step": 122500
    },
    {
      "epoch": 0.49,
      "learning_rate": 2.5323898904220615e-05,
      "loss": 3.7732,
      "step": 123000
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.522358955017273e-05,
      "loss": 3.7634,
      "step": 123500
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.512328019612485e-05,
      "loss": 3.7609,
      "step": 124000
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.5022970842076966e-05,
      "loss": 3.7604,
      "step": 124500
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.4922661488029082e-05,
      "loss": 3.7602,
      "step": 125000
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.48223521339812e-05,
      "loss": 3.7664,
      "step": 125500
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.4722042779933314e-05,
      "loss": 3.7629,
      "step": 126000
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.4621733425885434e-05,
      "loss": 3.7592,
      "step": 126500
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.452142407183755e-05,
      "loss": 3.7627,
      "step": 127000
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.4421114717789666e-05,
      "loss": 3.7622,
      "step": 127500
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.4320805363741782e-05,
      "loss": 3.7569,
      "step": 128000
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.4220496009693898e-05,
      "loss": 3.7576,
      "step": 128500
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.4120186655646014e-05,
      "loss": 3.7599,
      "step": 129000
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.401987730159813e-05,
      "loss": 3.764,
      "step": 129500
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.3919567947550246e-05,
      "loss": 3.7585,
      "step": 130000
    },
    {
      "epoch": 0.52,
      "eval_accuracy": 0.3521617310984962,
      "eval_loss": 3.6750001907348633,
      "eval_runtime": 4751.082,
      "eval_samples_per_second": 88.325,
      "eval_steps_per_second": 1.38,
      "step": 130000
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.3819258593502365e-05,
      "loss": 3.754,
      "step": 130500
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.371894923945448e-05,
      "loss": 3.751,
      "step": 131000
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.3618639885406597e-05,
      "loss": 3.7583,
      "step": 131500
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.3518330531358713e-05,
      "loss": 3.7475,
      "step": 132000
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.341802117731083e-05,
      "loss": 3.7542,
      "step": 132500
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.3317711823262945e-05,
      "loss": 3.7468,
      "step": 133000
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.321740246921506e-05,
      "loss": 3.746,
      "step": 133500
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.3117093115167177e-05,
      "loss": 3.7527,
      "step": 134000
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.3016783761119293e-05,
      "loss": 3.748,
      "step": 134500
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.291647440707141e-05,
      "loss": 3.7517,
      "step": 135000
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.2816165053023525e-05,
      "loss": 3.7453,
      "step": 135500
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.271585569897564e-05,
      "loss": 3.751,
      "step": 136000
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.2615546344927757e-05,
      "loss": 3.7542,
      "step": 136500
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.2515236990879873e-05,
      "loss": 3.7535,
      "step": 137000
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.2414927636831992e-05,
      "loss": 3.7499,
      "step": 137500
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.231461828278411e-05,
      "loss": 3.7421,
      "step": 138000
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.2214308928736224e-05,
      "loss": 3.7428,
      "step": 138500
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.211399957468834e-05,
      "loss": 3.7429,
      "step": 139000
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.2013690220640456e-05,
      "loss": 3.7469,
      "step": 139500
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.1913380866592572e-05,
      "loss": 3.7459,
      "step": 140000
    },
    {
      "epoch": 0.56,
      "eval_accuracy": 0.35345456367995787,
      "eval_loss": 3.6619808673858643,
      "eval_runtime": 4644.9079,
      "eval_samples_per_second": 90.344,
      "eval_steps_per_second": 1.412,
      "step": 140000
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.181307151254469e-05,
      "loss": 3.7399,
      "step": 140500
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.1712762158496804e-05,
      "loss": 3.7392,
      "step": 141000
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.161245280444892e-05,
      "loss": 3.7433,
      "step": 141500
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.1512143450401036e-05,
      "loss": 3.7379,
      "step": 142000
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.1411834096353152e-05,
      "loss": 3.7382,
      "step": 142500
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.131152474230527e-05,
      "loss": 3.7347,
      "step": 143000
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.1211215388257384e-05,
      "loss": 3.7424,
      "step": 143500
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.1110906034209504e-05,
      "loss": 3.7371,
      "step": 144000
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.101059668016162e-05,
      "loss": 3.7472,
      "step": 144500
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.0910287326113736e-05,
      "loss": 3.7358,
      "step": 145000
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.080997797206585e-05,
      "loss": 3.7351,
      "step": 145500
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.0709668618017968e-05,
      "loss": 3.7308,
      "step": 146000
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.0609359263970084e-05,
      "loss": 3.7387,
      "step": 146500
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.05090499099222e-05,
      "loss": 3.7255,
      "step": 147000
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.0408740555874316e-05,
      "loss": 3.7346,
      "step": 147500
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.030843120182643e-05,
      "loss": 3.7399,
      "step": 148000
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.020812184777855e-05,
      "loss": 3.7363,
      "step": 148500
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.0107812493730667e-05,
      "loss": 3.7349,
      "step": 149000
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.0007503139682783e-05,
      "loss": 3.7266,
      "step": 149500
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.99071937856349e-05,
      "loss": 3.7378,
      "step": 150000
    },
    {
      "epoch": 0.6,
      "eval_accuracy": 0.3545389268009225,
      "eval_loss": 3.650148630142212,
      "eval_runtime": 4647.2835,
      "eval_samples_per_second": 90.298,
      "eval_steps_per_second": 1.411,
      "step": 150000
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.9806884431587015e-05,
      "loss": 3.7285,
      "step": 150500
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.9706575077539134e-05,
      "loss": 3.7343,
      "step": 151000
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.960626572349125e-05,
      "loss": 3.7287,
      "step": 151500
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.9505956369443366e-05,
      "loss": 3.728,
      "step": 152000
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.9405647015395482e-05,
      "loss": 3.7368,
      "step": 152500
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.93053376613476e-05,
      "loss": 3.7284,
      "step": 153000
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.9205028307299714e-05,
      "loss": 3.7245,
      "step": 153500
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.910471895325183e-05,
      "loss": 3.7256,
      "step": 154000
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.9004409599203946e-05,
      "loss": 3.7238,
      "step": 154500
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.8904100245156062e-05,
      "loss": 3.7256,
      "step": 155000
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.880379089110818e-05,
      "loss": 3.7265,
      "step": 155500
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8703481537060294e-05,
      "loss": 3.725,
      "step": 156000
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.860317218301241e-05,
      "loss": 3.7152,
      "step": 156500
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8502862828964526e-05,
      "loss": 3.7176,
      "step": 157000
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8402553474916646e-05,
      "loss": 3.7206,
      "step": 157500
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8302244120868762e-05,
      "loss": 3.7272,
      "step": 158000
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.8201934766820878e-05,
      "loss": 3.7181,
      "step": 158500
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.8101625412772994e-05,
      "loss": 3.7257,
      "step": 159000
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.800131605872511e-05,
      "loss": 3.7215,
      "step": 159500
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7901006704677226e-05,
      "loss": 3.7181,
      "step": 160000
    },
    {
      "epoch": 0.64,
      "eval_accuracy": 0.3558949659698775,
      "eval_loss": 3.638543128967285,
      "eval_runtime": 4644.9417,
      "eval_samples_per_second": 90.343,
      "eval_steps_per_second": 1.412,
      "step": 160000
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.780069735062934e-05,
      "loss": 3.719,
      "step": 160500
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.7700387996581458e-05,
      "loss": 3.7194,
      "step": 161000
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.7600078642533574e-05,
      "loss": 3.7197,
      "step": 161500
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.749976928848569e-05,
      "loss": 3.7198,
      "step": 162000
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.7399459934437806e-05,
      "loss": 3.7223,
      "step": 162500
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.729915058038992e-05,
      "loss": 3.7201,
      "step": 163000
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.7198841226342038e-05,
      "loss": 3.7158,
      "step": 163500
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.7098531872294157e-05,
      "loss": 3.7124,
      "step": 164000
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.6998222518246273e-05,
      "loss": 3.7162,
      "step": 164500
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.689791316419839e-05,
      "loss": 3.7146,
      "step": 165000
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.6797603810150505e-05,
      "loss": 3.7111,
      "step": 165500
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.669729445610262e-05,
      "loss": 3.7226,
      "step": 166000
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.6596985102054737e-05,
      "loss": 3.7134,
      "step": 166500
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.6496675748006853e-05,
      "loss": 3.7138,
      "step": 167000
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.639636639395897e-05,
      "loss": 3.7155,
      "step": 167500
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.6296057039911085e-05,
      "loss": 3.7161,
      "step": 168000
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.61957476858632e-05,
      "loss": 3.7155,
      "step": 168500
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.6095438331815317e-05,
      "loss": 3.7118,
      "step": 169000
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5995128977767433e-05,
      "loss": 3.714,
      "step": 169500
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5894819623719552e-05,
      "loss": 3.7139,
      "step": 170000
    },
    {
      "epoch": 0.68,
      "eval_accuracy": 0.35682303406635435,
      "eval_loss": 3.629270315170288,
      "eval_runtime": 4647.4689,
      "eval_samples_per_second": 90.294,
      "eval_steps_per_second": 1.411,
      "step": 170000
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.579451026967167e-05,
      "loss": 3.7157,
      "step": 170500
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5694200915623784e-05,
      "loss": 3.7122,
      "step": 171000
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.55938915615759e-05,
      "loss": 3.7123,
      "step": 171500
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5493582207528016e-05,
      "loss": 3.7171,
      "step": 172000
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5393272853480136e-05,
      "loss": 3.7049,
      "step": 172500
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5292963499432252e-05,
      "loss": 3.7011,
      "step": 173000
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5192654145384366e-05,
      "loss": 3.71,
      "step": 173500
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5092344791336482e-05,
      "loss": 3.7042,
      "step": 174000
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.4992035437288598e-05,
      "loss": 3.7035,
      "step": 174500
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.4891726083240714e-05,
      "loss": 3.7043,
      "step": 175000
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.479141672919283e-05,
      "loss": 3.7054,
      "step": 175500
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.4691107375144948e-05,
      "loss": 3.7039,
      "step": 176000
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.4590798021097064e-05,
      "loss": 3.7069,
      "step": 176500
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.449048866704918e-05,
      "loss": 3.7039,
      "step": 177000
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.4390179313001297e-05,
      "loss": 3.7046,
      "step": 177500
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.4289869958953413e-05,
      "loss": 3.7101,
      "step": 178000
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.418956060490553e-05,
      "loss": 3.7113,
      "step": 178500
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.4089251250857647e-05,
      "loss": 3.7057,
      "step": 179000
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.3988941896809763e-05,
      "loss": 3.7077,
      "step": 179500
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.3888632542761879e-05,
      "loss": 3.6958,
      "step": 180000
    },
    {
      "epoch": 0.72,
      "eval_accuracy": 0.35775385321243053,
      "eval_loss": 3.6201136112213135,
      "eval_runtime": 4642.689,
      "eval_samples_per_second": 90.387,
      "eval_steps_per_second": 1.412,
      "step": 180000
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.3788323188713995e-05,
      "loss": 3.7004,
      "step": 180500
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.3688013834666111e-05,
      "loss": 3.7042,
      "step": 181000
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.3587704480618227e-05,
      "loss": 3.6989,
      "step": 181500
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.3487395126570343e-05,
      "loss": 3.7118,
      "step": 182000
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.3387085772522459e-05,
      "loss": 3.6947,
      "step": 182500
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.3286776418474575e-05,
      "loss": 3.6928,
      "step": 183000
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.3186467064426691e-05,
      "loss": 3.6959,
      "step": 183500
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.308615771037881e-05,
      "loss": 3.7036,
      "step": 184000
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.2985848356330926e-05,
      "loss": 3.7079,
      "step": 184500
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.2885539002283042e-05,
      "loss": 3.7025,
      "step": 185000
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.2785229648235158e-05,
      "loss": 3.7001,
      "step": 185500
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.2684920294187274e-05,
      "loss": 3.6922,
      "step": 186000
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.258461094013939e-05,
      "loss": 3.6997,
      "step": 186500
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.2484301586091506e-05,
      "loss": 3.6995,
      "step": 187000
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.2383992232043622e-05,
      "loss": 3.6927,
      "step": 187500
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.2283682877995738e-05,
      "loss": 3.6924,
      "step": 188000
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.2183373523947856e-05,
      "loss": 3.6931,
      "step": 188500
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.2083064169899972e-05,
      "loss": 3.6964,
      "step": 189000
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.198275481585209e-05,
      "loss": 3.695,
      "step": 189500
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.1882445461804206e-05,
      "loss": 3.6872,
      "step": 190000
    },
    {
      "epoch": 0.76,
      "eval_accuracy": 0.3585257078142553,
      "eval_loss": 3.612175226211548,
      "eval_runtime": 4645.5279,
      "eval_samples_per_second": 90.332,
      "eval_steps_per_second": 1.411,
      "step": 190000
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.1782136107756322e-05,
      "loss": 3.688,
      "step": 190500
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.1681826753708438e-05,
      "loss": 3.7005,
      "step": 191000
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.1581517399660554e-05,
      "loss": 3.6877,
      "step": 191500
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.148120804561267e-05,
      "loss": 3.7002,
      "step": 192000
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.1380898691564787e-05,
      "loss": 3.6913,
      "step": 192500
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.1280589337516903e-05,
      "loss": 3.6948,
      "step": 193000
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.118027998346902e-05,
      "loss": 3.6953,
      "step": 193500
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.1079970629421135e-05,
      "loss": 3.693,
      "step": 194000
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.0979661275373251e-05,
      "loss": 3.6806,
      "step": 194500
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.0879351921325367e-05,
      "loss": 3.6927,
      "step": 195000
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.0779042567277483e-05,
      "loss": 3.6979,
      "step": 195500
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.0678733213229601e-05,
      "loss": 3.6916,
      "step": 196000
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.0578423859181717e-05,
      "loss": 3.6958,
      "step": 196500
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.0478114505133833e-05,
      "loss": 3.6912,
      "step": 197000
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.0377805151085949e-05,
      "loss": 3.6865,
      "step": 197500
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.0277495797038065e-05,
      "loss": 3.687,
      "step": 198000
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.0177186442990181e-05,
      "loss": 3.6877,
      "step": 198500
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.0076877088942299e-05,
      "loss": 3.6895,
      "step": 199000
    },
    {
      "epoch": 0.8,
      "learning_rate": 9.976567734894415e-06,
      "loss": 3.6805,
      "step": 199500
    },
    {
      "epoch": 0.8,
      "learning_rate": 9.87625838084653e-06,
      "loss": 3.6888,
      "step": 200000
    },
    {
      "epoch": 0.8,
      "eval_accuracy": 0.3592407500971306,
      "eval_loss": 3.6060221195220947,
      "eval_runtime": 4651.0982,
      "eval_samples_per_second": 90.224,
      "eval_steps_per_second": 1.41,
      "step": 200000
    },
    {
      "epoch": 0.8,
      "learning_rate": 9.775949026798648e-06,
      "loss": 3.6844,
      "step": 200500
    },
    {
      "epoch": 0.81,
      "learning_rate": 9.675639672750764e-06,
      "loss": 3.6824,
      "step": 201000
    },
    {
      "epoch": 0.81,
      "learning_rate": 9.57533031870288e-06,
      "loss": 3.6902,
      "step": 201500
    },
    {
      "epoch": 0.81,
      "learning_rate": 9.475020964654996e-06,
      "loss": 3.6883,
      "step": 202000
    },
    {
      "epoch": 0.81,
      "learning_rate": 9.374711610607114e-06,
      "loss": 3.6904,
      "step": 202500
    },
    {
      "epoch": 0.81,
      "learning_rate": 9.27440225655923e-06,
      "loss": 3.6861,
      "step": 203000
    },
    {
      "epoch": 0.82,
      "learning_rate": 9.174092902511346e-06,
      "loss": 3.6838,
      "step": 203500
    },
    {
      "epoch": 0.82,
      "learning_rate": 9.073783548463462e-06,
      "loss": 3.6837,
      "step": 204000
    },
    {
      "epoch": 0.82,
      "learning_rate": 8.973474194415578e-06,
      "loss": 3.6876,
      "step": 204500
    },
    {
      "epoch": 0.82,
      "learning_rate": 8.873164840367694e-06,
      "loss": 3.6857,
      "step": 205000
    },
    {
      "epoch": 0.82,
      "learning_rate": 8.77285548631981e-06,
      "loss": 3.6821,
      "step": 205500
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.672546132271928e-06,
      "loss": 3.6787,
      "step": 206000
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.572236778224044e-06,
      "loss": 3.6878,
      "step": 206500
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.47192742417616e-06,
      "loss": 3.6759,
      "step": 207000
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.371618070128276e-06,
      "loss": 3.6827,
      "step": 207500
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.271308716080392e-06,
      "loss": 3.679,
      "step": 208000
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.170999362032508e-06,
      "loss": 3.6827,
      "step": 208500
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.070690007984625e-06,
      "loss": 3.6825,
      "step": 209000
    },
    {
      "epoch": 0.84,
      "learning_rate": 7.970380653936741e-06,
      "loss": 3.6823,
      "step": 209500
    },
    {
      "epoch": 0.84,
      "learning_rate": 7.870071299888857e-06,
      "loss": 3.6765,
      "step": 210000
    },
    {
      "epoch": 0.84,
      "eval_accuracy": 0.35991400717449046,
      "eval_loss": 3.6000776290893555,
      "eval_runtime": 4655.0214,
      "eval_samples_per_second": 90.148,
      "eval_steps_per_second": 1.409,
      "step": 210000
    },
    {
      "epoch": 0.84,
      "learning_rate": 7.769761945840973e-06,
      "loss": 3.6822,
      "step": 210500
    },
    {
      "epoch": 0.85,
      "learning_rate": 7.66945259179309e-06,
      "loss": 3.6778,
      "step": 211000
    },
    {
      "epoch": 0.85,
      "learning_rate": 7.569143237745206e-06,
      "loss": 3.6786,
      "step": 211500
    },
    {
      "epoch": 0.85,
      "learning_rate": 7.468833883697322e-06,
      "loss": 3.6745,
      "step": 212000
    },
    {
      "epoch": 0.85,
      "learning_rate": 7.36852452964944e-06,
      "loss": 3.6767,
      "step": 212500
    },
    {
      "epoch": 0.85,
      "learning_rate": 7.268215175601556e-06,
      "loss": 3.6785,
      "step": 213000
    },
    {
      "epoch": 0.86,
      "learning_rate": 7.167905821553672e-06,
      "loss": 3.685,
      "step": 213500
    },
    {
      "epoch": 0.86,
      "learning_rate": 7.067596467505788e-06,
      "loss": 3.6694,
      "step": 214000
    },
    {
      "epoch": 0.86,
      "learning_rate": 6.967287113457905e-06,
      "loss": 3.6816,
      "step": 214500
    },
    {
      "epoch": 0.86,
      "learning_rate": 6.866977759410021e-06,
      "loss": 3.6779,
      "step": 215000
    },
    {
      "epoch": 0.86,
      "learning_rate": 6.766668405362137e-06,
      "loss": 3.6772,
      "step": 215500
    },
    {
      "epoch": 0.87,
      "learning_rate": 6.666359051314254e-06,
      "loss": 3.673,
      "step": 216000
    },
    {
      "epoch": 0.87,
      "learning_rate": 6.56604969726637e-06,
      "loss": 3.6848,
      "step": 216500
    },
    {
      "epoch": 0.87,
      "learning_rate": 6.465740343218486e-06,
      "loss": 3.6748,
      "step": 217000
    },
    {
      "epoch": 0.87,
      "learning_rate": 6.365430989170602e-06,
      "loss": 3.6738,
      "step": 217500
    },
    {
      "epoch": 0.87,
      "learning_rate": 6.265121635122718e-06,
      "loss": 3.6802,
      "step": 218000
    },
    {
      "epoch": 0.88,
      "learning_rate": 6.164812281074835e-06,
      "loss": 3.6818,
      "step": 218500
    },
    {
      "epoch": 0.88,
      "learning_rate": 6.064502927026951e-06,
      "loss": 3.6722,
      "step": 219000
    },
    {
      "epoch": 0.88,
      "learning_rate": 5.964193572979067e-06,
      "loss": 3.6728,
      "step": 219500
    },
    {
      "epoch": 0.88,
      "learning_rate": 5.863884218931184e-06,
      "loss": 3.6734,
      "step": 220000
    },
    {
      "epoch": 0.88,
      "eval_accuracy": 0.36040756550566483,
      "eval_loss": 3.5944714546203613,
      "eval_runtime": 4642.1498,
      "eval_samples_per_second": 90.398,
      "eval_steps_per_second": 1.412,
      "step": 220000
    },
    {
      "epoch": 0.88,
      "learning_rate": 5.763574864883301e-06,
      "loss": 3.6763,
      "step": 220500
    },
    {
      "epoch": 0.89,
      "learning_rate": 5.663265510835417e-06,
      "loss": 3.6743,
      "step": 221000
    },
    {
      "epoch": 0.89,
      "learning_rate": 5.562956156787533e-06,
      "loss": 3.6718,
      "step": 221500
    },
    {
      "epoch": 0.89,
      "learning_rate": 5.46264680273965e-06,
      "loss": 3.6695,
      "step": 222000
    },
    {
      "epoch": 0.89,
      "learning_rate": 5.362337448691766e-06,
      "loss": 3.6704,
      "step": 222500
    },
    {
      "epoch": 0.89,
      "learning_rate": 5.262028094643882e-06,
      "loss": 3.6728,
      "step": 223000
    },
    {
      "epoch": 0.9,
      "learning_rate": 5.1617187405959985e-06,
      "loss": 3.6713,
      "step": 223500
    },
    {
      "epoch": 0.9,
      "learning_rate": 5.0614093865481145e-06,
      "loss": 3.664,
      "step": 224000
    },
    {
      "epoch": 0.9,
      "learning_rate": 4.9611000325002305e-06,
      "loss": 3.6659,
      "step": 224500
    },
    {
      "epoch": 0.9,
      "learning_rate": 4.860790678452347e-06,
      "loss": 3.6739,
      "step": 225000
    },
    {
      "epoch": 0.9,
      "learning_rate": 4.760481324404463e-06,
      "loss": 3.6794,
      "step": 225500
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.66017197035658e-06,
      "loss": 3.6702,
      "step": 226000
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.559862616308696e-06,
      "loss": 3.6667,
      "step": 226500
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.459553262260813e-06,
      "loss": 3.6651,
      "step": 227000
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.359243908212929e-06,
      "loss": 3.6701,
      "step": 227500
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.258934554165045e-06,
      "loss": 3.6626,
      "step": 228000
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.158625200117162e-06,
      "loss": 3.6664,
      "step": 228500
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.058315846069278e-06,
      "loss": 3.6732,
      "step": 229000
    },
    {
      "epoch": 0.92,
      "learning_rate": 3.958006492021394e-06,
      "loss": 3.6714,
      "step": 229500
    },
    {
      "epoch": 0.92,
      "learning_rate": 3.857697137973511e-06,
      "loss": 3.6669,
      "step": 230000
    },
    {
      "epoch": 0.92,
      "eval_accuracy": 0.36110471781316056,
      "eval_loss": 3.5890560150146484,
      "eval_runtime": 4652.0644,
      "eval_samples_per_second": 90.205,
      "eval_steps_per_second": 1.409,
      "step": 230000
    },
    {
      "epoch": 0.92,
      "learning_rate": 3.757387783925627e-06,
      "loss": 3.668,
      "step": 230500
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.657078429877743e-06,
      "loss": 3.6712,
      "step": 231000
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.556769075829859e-06,
      "loss": 3.6695,
      "step": 231500
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.456459721781976e-06,
      "loss": 3.6673,
      "step": 232000
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.356150367734092e-06,
      "loss": 3.6736,
      "step": 232500
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.2558410136862083e-06,
      "loss": 3.6695,
      "step": 233000
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.155531659638325e-06,
      "loss": 3.6642,
      "step": 233500
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.055222305590441e-06,
      "loss": 3.6652,
      "step": 234000
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.9549129515425576e-06,
      "loss": 3.6664,
      "step": 234500
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.8546035974946736e-06,
      "loss": 3.67,
      "step": 235000
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.75429424344679e-06,
      "loss": 3.6692,
      "step": 235500
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.6539848893989064e-06,
      "loss": 3.6647,
      "step": 236000
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.553675535351023e-06,
      "loss": 3.6749,
      "step": 236500
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.4533661813031392e-06,
      "loss": 3.672,
      "step": 237000
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.3530568272552552e-06,
      "loss": 3.6705,
      "step": 237500
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.2527474732073717e-06,
      "loss": 3.6687,
      "step": 238000
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.1524381191594877e-06,
      "loss": 3.666,
      "step": 238500
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.0521287651116045e-06,
      "loss": 3.6686,
      "step": 239000
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.951819411063721e-06,
      "loss": 3.6631,
      "step": 239500
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.851510057015837e-06,
      "loss": 3.6696,
      "step": 240000
    },
    {
      "epoch": 0.96,
      "eval_accuracy": 0.3614285682971602,
      "eval_loss": 3.58561635017395,
      "eval_runtime": 4649.4492,
      "eval_samples_per_second": 90.256,
      "eval_steps_per_second": 1.41,
      "step": 240000
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.7512007029679533e-06,
      "loss": 3.6624,
      "step": 240500
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.6508913489200695e-06,
      "loss": 3.6652,
      "step": 241000
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.550581994872186e-06,
      "loss": 3.6659,
      "step": 241500
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4502726408243022e-06,
      "loss": 3.6623,
      "step": 242000
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3499632867764186e-06,
      "loss": 3.6672,
      "step": 242500
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.2496539327285348e-06,
      "loss": 3.6639,
      "step": 243000
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.1493445786806512e-06,
      "loss": 3.6624,
      "step": 243500
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.0490352246327676e-06,
      "loss": 3.6595,
      "step": 244000
    },
    {
      "epoch": 0.98,
      "learning_rate": 9.487258705848838e-07,
      "loss": 3.6607,
      "step": 244500
    },
    {
      "epoch": 0.98,
      "learning_rate": 8.484165165370001e-07,
      "loss": 3.6529,
      "step": 245000
    },
    {
      "epoch": 0.99,
      "learning_rate": 7.481071624891164e-07,
      "loss": 3.6692,
      "step": 245500
    },
    {
      "epoch": 0.99,
      "learning_rate": 6.477978084412328e-07,
      "loss": 3.6662,
      "step": 246000
    },
    {
      "epoch": 0.99,
      "learning_rate": 5.474884543933492e-07,
      "loss": 3.6656,
      "step": 246500
    },
    {
      "epoch": 0.99,
      "learning_rate": 4.4717910034546543e-07,
      "loss": 3.6597,
      "step": 247000
    },
    {
      "epoch": 0.99,
      "learning_rate": 3.468697462975818e-07,
      "loss": 3.6637,
      "step": 247500
    },
    {
      "epoch": 1.0,
      "learning_rate": 2.4656039224969806e-07,
      "loss": 3.6581,
      "step": 248000
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.462510382018144e-07,
      "loss": 3.6684,
      "step": 248500
    },
    {
      "epoch": 1.0,
      "learning_rate": 4.594168415393073e-08,
      "loss": 3.6663,
      "step": 249000
    },
    {
      "epoch": 1.0,
      "step": 249229,
      "total_flos": 4.167764163035136e+18,
      "train_loss": 3.8837062227938306,
      "train_runtime": 230124.717,
      "train_samples_per_second": 34.656,
      "train_steps_per_second": 1.083
    }
  ],
  "logging_steps": 500,
  "max_steps": 249229,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10000,
  "total_flos": 4.167764163035136e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}