{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.6665325513378,
  "eval_steps": 5000,
  "global_step": 70000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12,
      "learning_rate": 9.958730551772523e-05,
      "loss": 1.4035,
      "step": 1000
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.917461103545046e-05,
      "loss": 1.332,
      "step": 2000
    },
    {
      "epoch": 0.37,
      "learning_rate": 9.87619165531757e-05,
      "loss": 1.2981,
      "step": 3000
    },
    {
      "epoch": 0.5,
      "learning_rate": 9.834922207090091e-05,
      "loss": 1.272,
      "step": 4000
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.793652758862615e-05,
      "loss": 1.2501,
      "step": 5000
    },
    {
      "epoch": 0.62,
      "eval_loss": 1.1940463781356812,
      "eval_runtime": 120.3265,
      "eval_samples_per_second": 134.692,
      "eval_steps_per_second": 16.838,
      "step": 5000
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.752383310635137e-05,
      "loss": 1.2328,
      "step": 6000
    },
    {
      "epoch": 0.87,
      "learning_rate": 9.71111386240766e-05,
      "loss": 1.208,
      "step": 7000
    },
    {
      "epoch": 0.99,
      "learning_rate": 9.669844414180183e-05,
      "loss": 1.1886,
      "step": 8000
    },
    {
      "epoch": 1.11,
      "learning_rate": 9.628574965952706e-05,
      "loss": 1.182,
      "step": 9000
    },
    {
      "epoch": 1.24,
      "learning_rate": 9.587305517725228e-05,
      "loss": 1.1761,
      "step": 10000
    },
    {
      "epoch": 1.24,
      "eval_loss": 1.1365982294082642,
      "eval_runtime": 124.7536,
      "eval_samples_per_second": 129.912,
      "eval_steps_per_second": 16.24,
      "step": 10000
    },
    {
      "epoch": 1.36,
      "learning_rate": 9.546036069497752e-05,
      "loss": 1.1594,
      "step": 11000
    },
    {
      "epoch": 1.49,
      "learning_rate": 9.504766621270275e-05,
      "loss": 1.1449,
      "step": 12000
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.463497173042796e-05,
      "loss": 1.1387,
      "step": 13000
    },
    {
      "epoch": 1.73,
      "learning_rate": 9.422227724815321e-05,
      "loss": 1.1318,
      "step": 14000
    },
    {
      "epoch": 1.86,
      "learning_rate": 9.380958276587842e-05,
      "loss": 1.1222,
      "step": 15000
    },
    {
      "epoch": 1.86,
      "eval_loss": 1.103798270225525,
      "eval_runtime": 119.9985,
      "eval_samples_per_second": 135.06,
      "eval_steps_per_second": 16.884,
      "step": 15000
    },
    {
      "epoch": 1.98,
      "learning_rate": 9.339688828360365e-05,
      "loss": 1.1164,
      "step": 16000
    },
    {
      "epoch": 2.1,
      "learning_rate": 9.298419380132888e-05,
      "loss": 1.1073,
      "step": 17000
    },
    {
      "epoch": 2.23,
      "learning_rate": 9.257149931905411e-05,
      "loss": 1.1026,
      "step": 18000
    },
    {
      "epoch": 2.35,
      "learning_rate": 9.215880483677933e-05,
      "loss": 1.0957,
      "step": 19000
    },
    {
      "epoch": 2.48,
      "learning_rate": 9.174611035450457e-05,
      "loss": 1.0859,
      "step": 20000
    },
    {
      "epoch": 2.48,
      "eval_loss": 1.0691076517105103,
      "eval_runtime": 120.0736,
      "eval_samples_per_second": 134.976,
      "eval_steps_per_second": 16.873,
      "step": 20000
    },
    {
      "epoch": 2.6,
      "learning_rate": 9.133341587222979e-05,
      "loss": 1.0856,
      "step": 21000
    },
    {
      "epoch": 2.72,
      "learning_rate": 9.092072138995502e-05,
      "loss": 1.0769,
      "step": 22000
    },
    {
      "epoch": 2.85,
      "learning_rate": 9.050802690768025e-05,
      "loss": 1.0768,
      "step": 23000
    },
    {
      "epoch": 2.97,
      "learning_rate": 9.009533242540548e-05,
      "loss": 1.0671,
      "step": 24000
    },
    {
      "epoch": 3.09,
      "learning_rate": 8.968263794313071e-05,
      "loss": 1.0703,
      "step": 25000
    },
    {
      "epoch": 3.09,
      "eval_loss": 1.0342341661453247,
      "eval_runtime": 142.3338,
      "eval_samples_per_second": 113.866,
      "eval_steps_per_second": 14.234,
      "step": 25000
    },
    {
      "epoch": 3.22,
      "learning_rate": 8.926994346085594e-05,
      "loss": 1.0535,
      "step": 26000
    },
    {
      "epoch": 3.34,
      "learning_rate": 8.885724897858117e-05,
      "loss": 1.0537,
      "step": 27000
    },
    {
      "epoch": 3.47,
      "learning_rate": 8.844455449630638e-05,
      "loss": 1.0502,
      "step": 28000
    },
    {
      "epoch": 3.59,
      "learning_rate": 8.803186001403163e-05,
      "loss": 1.0401,
      "step": 29000
    },
    {
      "epoch": 3.71,
      "learning_rate": 8.761916553175684e-05,
      "loss": 1.0461,
      "step": 30000
    },
    {
      "epoch": 3.71,
      "eval_loss": 1.0171171426773071,
      "eval_runtime": 155.2187,
      "eval_samples_per_second": 104.414,
      "eval_steps_per_second": 13.053,
      "step": 30000
    },
    {
      "epoch": 3.84,
      "learning_rate": 8.720647104948207e-05,
      "loss": 1.0362,
      "step": 31000
    },
    {
      "epoch": 3.96,
      "learning_rate": 8.67937765672073e-05,
      "loss": 1.0332,
      "step": 32000
    },
    {
      "epoch": 4.09,
      "learning_rate": 8.638108208493253e-05,
      "loss": 1.0293,
      "step": 33000
    },
    {
      "epoch": 4.21,
      "learning_rate": 8.596838760265775e-05,
      "loss": 1.0286,
      "step": 34000
    },
    {
      "epoch": 4.33,
      "learning_rate": 8.555569312038299e-05,
      "loss": 1.025,
      "step": 35000
    },
    {
      "epoch": 4.33,
      "eval_loss": 0.9936763048171997,
      "eval_runtime": 119.72,
      "eval_samples_per_second": 135.374,
      "eval_steps_per_second": 16.923,
      "step": 35000
    },
    {
      "epoch": 4.46,
      "learning_rate": 8.514299863810822e-05,
      "loss": 1.0184,
      "step": 36000
    },
    {
      "epoch": 4.58,
      "learning_rate": 8.473030415583344e-05,
      "loss": 1.0136,
      "step": 37000
    },
    {
      "epoch": 4.7,
      "learning_rate": 8.431760967355867e-05,
      "loss": 1.0103,
      "step": 38000
    },
    {
      "epoch": 4.83,
      "learning_rate": 8.39049151912839e-05,
      "loss": 1.0081,
      "step": 39000
    },
    {
      "epoch": 4.95,
      "learning_rate": 8.349222070900913e-05,
      "loss": 1.0063,
      "step": 40000
    },
    {
      "epoch": 4.95,
      "eval_loss": 0.9914126396179199,
      "eval_runtime": 176.7148,
      "eval_samples_per_second": 91.713,
      "eval_steps_per_second": 11.465,
      "step": 40000
    },
    {
      "epoch": 5.08,
      "learning_rate": 8.307952622673436e-05,
      "loss": 1.003,
      "step": 41000
    },
    {
      "epoch": 5.2,
      "learning_rate": 8.266683174445959e-05,
      "loss": 0.9947,
      "step": 42000
    },
    {
      "epoch": 5.32,
      "learning_rate": 8.22541372621848e-05,
      "loss": 0.999,
      "step": 43000
    },
    {
      "epoch": 5.45,
      "learning_rate": 8.184144277991003e-05,
      "loss": 0.9987,
      "step": 44000
    },
    {
      "epoch": 5.57,
      "learning_rate": 8.142874829763526e-05,
      "loss": 0.988,
      "step": 45000
    },
    {
      "epoch": 5.57,
      "eval_loss": 0.9876496195793152,
      "eval_runtime": 149.4553,
      "eval_samples_per_second": 108.44,
      "eval_steps_per_second": 13.556,
      "step": 45000
    },
    {
      "epoch": 5.69,
      "learning_rate": 8.101605381536049e-05,
      "loss": 0.9899,
      "step": 46000
    },
    {
      "epoch": 5.82,
      "learning_rate": 8.060335933308572e-05,
      "loss": 0.9876,
      "step": 47000
    },
    {
      "epoch": 5.94,
      "learning_rate": 8.019066485081095e-05,
      "loss": 0.9889,
      "step": 48000
    },
    {
      "epoch": 6.07,
      "learning_rate": 7.977797036853618e-05,
      "loss": 0.9807,
      "step": 49000
    },
    {
      "epoch": 6.19,
      "learning_rate": 7.93652758862614e-05,
      "loss": 0.971,
      "step": 50000
    },
    {
      "epoch": 6.19,
      "eval_loss": 0.9691153764724731,
      "eval_runtime": 155.3931,
      "eval_samples_per_second": 104.297,
      "eval_steps_per_second": 13.038,
      "step": 50000
    },
    {
      "epoch": 6.31,
      "learning_rate": 7.895258140398664e-05,
      "loss": 0.9796,
      "step": 51000
    },
    {
      "epoch": 6.44,
      "learning_rate": 7.853988692171186e-05,
      "loss": 0.9784,
      "step": 52000
    },
    {
      "epoch": 6.56,
      "learning_rate": 7.812719243943709e-05,
      "loss": 0.9682,
      "step": 53000
    },
    {
      "epoch": 6.69,
      "learning_rate": 7.771449795716232e-05,
      "loss": 0.9609,
      "step": 54000
    },
    {
      "epoch": 6.81,
      "learning_rate": 7.730180347488755e-05,
      "loss": 0.9704,
      "step": 55000
    },
    {
      "epoch": 6.81,
      "eval_loss": 0.9535136222839355,
      "eval_runtime": 160.7447,
      "eval_samples_per_second": 100.824,
      "eval_steps_per_second": 12.604,
      "step": 55000
    },
    {
      "epoch": 6.93,
      "learning_rate": 7.688910899261278e-05,
      "loss": 0.965,
      "step": 56000
    },
    {
      "epoch": 7.06,
      "learning_rate": 7.6476414510338e-05,
      "loss": 0.9574,
      "step": 57000
    },
    {
      "epoch": 7.18,
      "learning_rate": 7.606372002806322e-05,
      "loss": 0.9565,
      "step": 58000
    },
    {
      "epoch": 7.3,
      "learning_rate": 7.565102554578845e-05,
      "loss": 0.9514,
      "step": 59000
    },
    {
      "epoch": 7.43,
      "learning_rate": 7.52383310635137e-05,
      "loss": 0.951,
      "step": 60000
    },
    {
      "epoch": 7.43,
      "eval_loss": 0.9359220266342163,
      "eval_runtime": 124.2464,
      "eval_samples_per_second": 130.442,
      "eval_steps_per_second": 16.306,
      "step": 60000
    },
    {
      "epoch": 7.55,
      "learning_rate": 7.482563658123891e-05,
      "loss": 0.9477,
      "step": 61000
    },
    {
      "epoch": 7.68,
      "learning_rate": 7.441294209896414e-05,
      "loss": 0.9581,
      "step": 62000
    },
    {
      "epoch": 7.8,
      "learning_rate": 7.400024761668937e-05,
      "loss": 0.948,
      "step": 63000
    },
    {
      "epoch": 7.92,
      "learning_rate": 7.35875531344146e-05,
      "loss": 0.9439,
      "step": 64000
    },
    {
      "epoch": 8.05,
      "learning_rate": 7.317485865213982e-05,
      "loss": 0.9447,
      "step": 65000
    },
    {
      "epoch": 8.05,
      "eval_loss": 0.9360346794128418,
      "eval_runtime": 160.0152,
      "eval_samples_per_second": 101.284,
      "eval_steps_per_second": 12.661,
      "step": 65000
    },
    {
      "epoch": 8.17,
      "learning_rate": 7.276216416986506e-05,
      "loss": 0.9377,
      "step": 66000
    },
    {
      "epoch": 8.29,
      "learning_rate": 7.234946968759027e-05,
      "loss": 0.9455,
      "step": 67000
    },
    {
      "epoch": 8.42,
      "learning_rate": 7.19367752053155e-05,
      "loss": 0.9332,
      "step": 68000
    },
    {
      "epoch": 8.54,
      "learning_rate": 7.152408072304073e-05,
      "loss": 0.9374,
      "step": 69000
    },
    {
      "epoch": 8.67,
      "learning_rate": 7.111138624076596e-05,
      "loss": 0.9512,
      "step": 70000
    },
    {
      "epoch": 8.67,
      "eval_loss": 0.930946946144104,
      "eval_runtime": 132.6824,
      "eval_samples_per_second": 122.149,
      "eval_steps_per_second": 15.27,
      "step": 70000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 242310,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "total_flos": 3.5013441751675136e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}