{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 39924,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 4.9373810239454966e-05,
      "loss": 3.8228,
      "step": 500
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.874762047890993e-05,
      "loss": 2.8535,
      "step": 1000
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.8121430718364894e-05,
      "loss": 2.4924,
      "step": 1500
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.749524095781986e-05,
      "loss": 2.2832,
      "step": 2000
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.686905119727483e-05,
      "loss": 2.1875,
      "step": 2500
    },
    {
      "epoch": 0.23,
      "learning_rate": 4.624286143672979e-05,
      "loss": 2.067,
      "step": 3000
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.5616671676184756e-05,
      "loss": 1.9847,
      "step": 3500
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.499048191563971e-05,
      "loss": 1.9308,
      "step": 4000
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.436429215509468e-05,
      "loss": 1.8689,
      "step": 4500
    },
    {
      "epoch": 0.38,
      "learning_rate": 4.373810239454965e-05,
      "loss": 1.8071,
      "step": 5000
    },
    {
      "epoch": 0.41,
      "learning_rate": 4.311191263400461e-05,
      "loss": 1.7735,
      "step": 5500
    },
    {
      "epoch": 0.45,
      "learning_rate": 4.2485722873459575e-05,
      "loss": 1.7528,
      "step": 6000
    },
    {
      "epoch": 0.49,
      "learning_rate": 4.185953311291454e-05,
      "loss": 1.7191,
      "step": 6500
    },
    {
      "epoch": 0.53,
      "learning_rate": 4.12333433523695e-05,
      "loss": 1.7013,
      "step": 7000
    },
    {
      "epoch": 0.56,
      "learning_rate": 4.060715359182447e-05,
      "loss": 1.6758,
      "step": 7500
    },
    {
      "epoch": 0.6,
      "learning_rate": 3.9980963831279437e-05,
      "loss": 1.6408,
      "step": 8000
    },
    {
      "epoch": 0.64,
      "learning_rate": 3.9354774070734394e-05,
      "loss": 1.6212,
      "step": 8500
    },
    {
      "epoch": 0.68,
      "learning_rate": 3.8728584310189364e-05,
      "loss": 1.5981,
      "step": 9000
    },
    {
      "epoch": 0.71,
      "learning_rate": 3.810239454964433e-05,
      "loss": 1.6128,
      "step": 9500
    },
    {
      "epoch": 0.75,
      "learning_rate": 3.747620478909929e-05,
      "loss": 1.5632,
      "step": 10000
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.6850015028554255e-05,
      "loss": 1.5367,
      "step": 10500
    },
    {
      "epoch": 0.83,
      "learning_rate": 3.622382526800922e-05,
      "loss": 1.5306,
      "step": 11000
    },
    {
      "epoch": 0.86,
      "learning_rate": 3.559763550746418e-05,
      "loss": 1.5173,
      "step": 11500
    },
    {
      "epoch": 0.9,
      "learning_rate": 3.497144574691915e-05,
      "loss": 1.5114,
      "step": 12000
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.434525598637412e-05,
      "loss": 1.4808,
      "step": 12500
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.3719066225829074e-05,
      "loss": 1.4615,
      "step": 13000
    },
    {
      "epoch": 1.01,
      "learning_rate": 3.309287646528404e-05,
      "loss": 1.4125,
      "step": 13500
    },
    {
      "epoch": 1.05,
      "learning_rate": 3.246668670473901e-05,
      "loss": 1.2897,
      "step": 14000
    },
    {
      "epoch": 1.09,
      "learning_rate": 3.184049694419397e-05,
      "loss": 1.3042,
      "step": 14500
    },
    {
      "epoch": 1.13,
      "learning_rate": 3.1214307183648936e-05,
      "loss": 1.2825,
      "step": 15000
    },
    {
      "epoch": 1.16,
      "learning_rate": 3.05881174231039e-05,
      "loss": 1.2593,
      "step": 15500
    },
    {
      "epoch": 1.2,
      "learning_rate": 2.9961927662558864e-05,
      "loss": 1.2918,
      "step": 16000
    },
    {
      "epoch": 1.24,
      "learning_rate": 2.9335737902013828e-05,
      "loss": 1.284,
      "step": 16500
    },
    {
      "epoch": 1.28,
      "learning_rate": 2.8709548141468788e-05,
      "loss": 1.2604,
      "step": 17000
    },
    {
      "epoch": 1.31,
      "learning_rate": 2.808335838092376e-05,
      "loss": 1.2638,
      "step": 17500
    },
    {
      "epoch": 1.35,
      "learning_rate": 2.745716862037872e-05,
      "loss": 1.2803,
      "step": 18000
    },
    {
      "epoch": 1.39,
      "learning_rate": 2.6830978859833683e-05,
      "loss": 1.2515,
      "step": 18500
    },
    {
      "epoch": 1.43,
      "learning_rate": 2.620478909928865e-05,
      "loss": 1.2441,
      "step": 19000
    },
    {
      "epoch": 1.47,
      "learning_rate": 2.5578599338743614e-05,
      "loss": 1.2529,
      "step": 19500
    },
    {
      "epoch": 1.5,
      "learning_rate": 2.495240957819858e-05,
      "loss": 1.2428,
      "step": 20000
    },
    {
      "epoch": 1.54,
      "learning_rate": 2.432621981765354e-05,
      "loss": 1.2433,
      "step": 20500
    },
    {
      "epoch": 1.58,
      "learning_rate": 2.370003005710851e-05,
      "loss": 1.2489,
      "step": 21000
    },
    {
      "epoch": 1.62,
      "learning_rate": 2.307384029656347e-05,
      "loss": 1.2225,
      "step": 21500
    },
    {
      "epoch": 1.65,
      "learning_rate": 2.2447650536018436e-05,
      "loss": 1.2023,
      "step": 22000
    },
    {
      "epoch": 1.69,
      "learning_rate": 2.18214607754734e-05,
      "loss": 1.2342,
      "step": 22500
    },
    {
      "epoch": 1.73,
      "learning_rate": 2.1195271014928364e-05,
      "loss": 1.2031,
      "step": 23000
    },
    {
      "epoch": 1.77,
      "learning_rate": 2.056908125438333e-05,
      "loss": 1.1875,
      "step": 23500
    },
    {
      "epoch": 1.8,
      "learning_rate": 1.9942891493838295e-05,
      "loss": 1.2218,
      "step": 24000
    },
    {
      "epoch": 1.84,
      "learning_rate": 1.931670173329326e-05,
      "loss": 1.171,
      "step": 24500
    },
    {
      "epoch": 1.88,
      "learning_rate": 1.8690511972748222e-05,
      "loss": 1.1975,
      "step": 25000
    },
    {
      "epoch": 1.92,
      "learning_rate": 1.8064322212203186e-05,
      "loss": 1.1822,
      "step": 25500
    },
    {
      "epoch": 1.95,
      "learning_rate": 1.743813245165815e-05,
      "loss": 1.1852,
      "step": 26000
    },
    {
      "epoch": 1.99,
      "learning_rate": 1.6811942691113117e-05,
      "loss": 1.1766,
      "step": 26500
    },
    {
      "epoch": 2.03,
      "learning_rate": 1.618575293056808e-05,
      "loss": 1.0406,
      "step": 27000
    },
    {
      "epoch": 2.07,
      "learning_rate": 1.5559563170023044e-05,
      "loss": 0.9998,
      "step": 27500
    },
    {
      "epoch": 2.1,
      "learning_rate": 1.4933373409478008e-05,
      "loss": 1.0107,
      "step": 28000
    },
    {
      "epoch": 2.14,
      "learning_rate": 1.4307183648932974e-05,
      "loss": 1.0144,
      "step": 28500
    },
    {
      "epoch": 2.18,
      "learning_rate": 1.368099388838794e-05,
      "loss": 0.9972,
      "step": 29000
    },
    {
      "epoch": 2.22,
      "learning_rate": 1.3054804127842901e-05,
      "loss": 1.0242,
      "step": 29500
    },
    {
      "epoch": 2.25,
      "learning_rate": 1.2428614367297867e-05,
      "loss": 1.0047,
      "step": 30000
    },
    {
      "epoch": 2.29,
      "learning_rate": 1.180242460675283e-05,
      "loss": 1.0112,
      "step": 30500
    },
    {
      "epoch": 2.33,
      "learning_rate": 1.1176234846207794e-05,
      "loss": 0.9964,
      "step": 31000
    },
    {
      "epoch": 2.37,
      "learning_rate": 1.055004508566276e-05,
      "loss": 0.9952,
      "step": 31500
    },
    {
      "epoch": 2.4,
      "learning_rate": 9.923855325117724e-06,
      "loss": 1.0004,
      "step": 32000
    },
    {
      "epoch": 2.44,
      "learning_rate": 9.297665564572689e-06,
      "loss": 1.0177,
      "step": 32500
    },
    {
      "epoch": 2.48,
      "learning_rate": 8.671475804027653e-06,
      "loss": 0.9929,
      "step": 33000
    },
    {
      "epoch": 2.52,
      "learning_rate": 8.045286043482618e-06,
      "loss": 0.9847,
      "step": 33500
    },
    {
      "epoch": 2.55,
      "learning_rate": 7.419096282937582e-06,
      "loss": 0.994,
      "step": 34000
    },
    {
      "epoch": 2.59,
      "learning_rate": 6.792906522392546e-06,
      "loss": 0.9835,
      "step": 34500
    },
    {
      "epoch": 2.63,
      "learning_rate": 6.1667167618475106e-06,
      "loss": 0.9755,
      "step": 35000
    },
    {
      "epoch": 2.67,
      "learning_rate": 5.540527001302475e-06,
      "loss": 0.9851,
      "step": 35500
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.914337240757439e-06,
      "loss": 0.9688,
      "step": 36000
    },
    {
      "epoch": 2.74,
      "learning_rate": 4.288147480212404e-06,
      "loss": 0.9624,
      "step": 36500
    },
    {
      "epoch": 2.78,
      "learning_rate": 3.661957719667368e-06,
      "loss": 0.9728,
      "step": 37000
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.035767959122333e-06,
      "loss": 0.9714,
      "step": 37500
    },
    {
      "epoch": 2.86,
      "learning_rate": 2.409578198577297e-06,
      "loss": 0.9691,
      "step": 38000
    },
    {
      "epoch": 2.89,
      "learning_rate": 1.7833884380322613e-06,
      "loss": 0.9749,
      "step": 38500
    },
    {
      "epoch": 2.93,
      "learning_rate": 1.157198677487226e-06,
      "loss": 0.979,
      "step": 39000
    },
    {
      "epoch": 2.97,
      "learning_rate": 5.310089169421901e-07,
      "loss": 0.9538,
      "step": 39500
    },
    {
      "epoch": 3.0,
      "step": 39924,
      "total_flos": 3.493213683491635e+16,
      "train_loss": 1.367635575770233,
      "train_runtime": 30297.0784,
      "train_samples_per_second": 10.542,
      "train_steps_per_second": 1.318
    }
  ],
  "logging_steps": 500,
  "max_steps": 39924,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 10000,
  "total_flos": 3.493213683491635e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}