{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 8240,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12135922330097088,
      "grad_norm": 1.147416591644287,
      "learning_rate": 9.696601941747573e-05,
      "loss": 1.1102,
      "step": 250
    },
    {
      "epoch": 0.24271844660194175,
      "grad_norm": 1.0398495197296143,
      "learning_rate": 9.393203883495146e-05,
      "loss": 0.5891,
      "step": 500
    },
    {
      "epoch": 0.3640776699029126,
      "grad_norm": 0.7667867541313171,
      "learning_rate": 9.089805825242718e-05,
      "loss": 0.5176,
      "step": 750
    },
    {
      "epoch": 0.4854368932038835,
      "grad_norm": 0.7023970484733582,
      "learning_rate": 8.786407766990292e-05,
      "loss": 0.4917,
      "step": 1000
    },
    {
      "epoch": 0.6067961165048543,
      "grad_norm": 0.6781546473503113,
      "learning_rate": 8.483009708737865e-05,
      "loss": 0.4822,
      "step": 1250
    },
    {
      "epoch": 0.7281553398058253,
      "grad_norm": 0.7623434066772461,
      "learning_rate": 8.179611650485438e-05,
      "loss": 0.4655,
      "step": 1500
    },
    {
      "epoch": 0.8495145631067961,
      "grad_norm": 0.8142368793487549,
      "learning_rate": 7.87621359223301e-05,
      "loss": 0.45,
      "step": 1750
    },
    {
      "epoch": 0.970873786407767,
      "grad_norm": 0.8743399381637573,
      "learning_rate": 7.572815533980583e-05,
      "loss": 0.4304,
      "step": 2000
    },
    {
      "epoch": 1.0922330097087378,
      "grad_norm": 0.9348050355911255,
      "learning_rate": 7.269417475728155e-05,
      "loss": 0.3979,
      "step": 2250
    },
    {
      "epoch": 1.2135922330097086,
      "grad_norm": 1.0624396800994873,
      "learning_rate": 6.966019417475728e-05,
      "loss": 0.3695,
      "step": 2500
    },
    {
      "epoch": 1.3349514563106797,
      "grad_norm": 1.0327861309051514,
      "learning_rate": 6.662621359223301e-05,
      "loss": 0.3452,
      "step": 2750
    },
    {
      "epoch": 1.4563106796116505,
      "grad_norm": 1.0362504720687866,
      "learning_rate": 6.359223300970875e-05,
      "loss": 0.3194,
      "step": 3000
    },
    {
      "epoch": 1.5776699029126213,
      "grad_norm": 1.1497058868408203,
      "learning_rate": 6.055825242718447e-05,
      "loss": 0.3,
      "step": 3250
    },
    {
      "epoch": 1.6990291262135924,
      "grad_norm": 1.1127945184707642,
      "learning_rate": 5.752427184466019e-05,
      "loss": 0.2829,
      "step": 3500
    },
    {
      "epoch": 1.820388349514563,
      "grad_norm": 1.0874605178833008,
      "learning_rate": 5.4490291262135926e-05,
      "loss": 0.2714,
      "step": 3750
    },
    {
      "epoch": 1.941747572815534,
      "grad_norm": 0.965722382068634,
      "learning_rate": 5.145631067961165e-05,
      "loss": 0.2596,
      "step": 4000
    },
    {
      "epoch": 2.063106796116505,
      "grad_norm": 1.0977766513824463,
      "learning_rate": 4.8422330097087385e-05,
      "loss": 0.2352,
      "step": 4250
    },
    {
      "epoch": 2.1844660194174756,
      "grad_norm": 1.022316813468933,
      "learning_rate": 4.538834951456311e-05,
      "loss": 0.2209,
      "step": 4500
    },
    {
      "epoch": 2.3058252427184467,
      "grad_norm": 0.982537031173706,
      "learning_rate": 4.235436893203884e-05,
      "loss": 0.2171,
      "step": 4750
    },
    {
      "epoch": 2.4271844660194173,
      "grad_norm": 1.093375563621521,
      "learning_rate": 3.9320388349514564e-05,
      "loss": 0.2129,
      "step": 5000
    },
    {
      "epoch": 2.5485436893203883,
      "grad_norm": 0.9269146919250488,
      "learning_rate": 3.62864077669903e-05,
      "loss": 0.2088,
      "step": 5250
    },
    {
      "epoch": 2.6699029126213594,
      "grad_norm": 1.0317578315734863,
      "learning_rate": 3.325242718446602e-05,
      "loss": 0.2032,
      "step": 5500
    },
    {
      "epoch": 2.79126213592233,
      "grad_norm": 0.9504765868186951,
      "learning_rate": 3.0218446601941746e-05,
      "loss": 0.2013,
      "step": 5750
    },
    {
      "epoch": 2.912621359223301,
      "grad_norm": 0.9305168390274048,
      "learning_rate": 2.7184466019417475e-05,
      "loss": 0.1963,
      "step": 6000
    },
    {
      "epoch": 3.033980582524272,
      "grad_norm": 0.830916166305542,
      "learning_rate": 2.4150485436893205e-05,
      "loss": 0.1879,
      "step": 6250
    },
    {
      "epoch": 3.1553398058252426,
      "grad_norm": 0.8673647046089172,
      "learning_rate": 2.111650485436893e-05,
      "loss": 0.1753,
      "step": 6500
    },
    {
      "epoch": 3.2766990291262137,
      "grad_norm": 0.8825583457946777,
      "learning_rate": 1.808252427184466e-05,
      "loss": 0.1739,
      "step": 6750
    },
    {
      "epoch": 3.3980582524271843,
      "grad_norm": 1.0287333726882935,
      "learning_rate": 1.5048543689320387e-05,
      "loss": 0.1729,
      "step": 7000
    },
    {
      "epoch": 3.5194174757281553,
      "grad_norm": 0.809319794178009,
      "learning_rate": 1.2014563106796117e-05,
      "loss": 0.1718,
      "step": 7250
    },
    {
      "epoch": 3.6407766990291264,
      "grad_norm": 0.8941086530685425,
      "learning_rate": 8.980582524271845e-06,
      "loss": 0.1702,
      "step": 7500
    },
    {
      "epoch": 3.762135922330097,
      "grad_norm": 0.9008864760398865,
      "learning_rate": 5.946601941747574e-06,
      "loss": 0.1688,
      "step": 7750
    },
    {
      "epoch": 3.883495145631068,
      "grad_norm": 0.8730018138885498,
      "learning_rate": 2.912621359223301e-06,
      "loss": 0.1689,
      "step": 8000
    },
    {
      "epoch": 4.0,
      "step": 8240,
      "total_flos": 2.943708398017659e+18,
      "train_loss": 0.31335733145186045,
      "train_runtime": 45318.0558,
      "train_samples_per_second": 11.637,
      "train_steps_per_second": 0.182
    }
  ],
  "logging_steps": 250,
  "max_steps": 8240,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.943708398017659e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}