{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.2650056625141564,
  "eval_steps": 20,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09060022650056625,
      "grad_norm": 4.824601173400879,
      "learning_rate": 1.9393939393939395e-05,
      "loss": 6.3414,
      "step": 20
    },
    {
      "epoch": 0.09060022650056625,
      "eval_loss": 5.462944984436035,
      "eval_runtime": 169.7392,
      "eval_samples_per_second": 2.327,
      "eval_steps_per_second": 0.583,
      "step": 20
    },
    {
      "epoch": 0.1812004530011325,
      "grad_norm": 3.5273354053497314,
      "learning_rate": 1.8585858585858588e-05,
      "loss": 4.9204,
      "step": 40
    },
    {
      "epoch": 0.1812004530011325,
      "eval_loss": 4.40663480758667,
      "eval_runtime": 169.7659,
      "eval_samples_per_second": 2.327,
      "eval_steps_per_second": 0.583,
      "step": 40
    },
    {
      "epoch": 0.2718006795016987,
      "grad_norm": 2.5483546257019043,
      "learning_rate": 1.7777777777777777e-05,
      "loss": 4.0241,
      "step": 60
    },
    {
      "epoch": 0.2718006795016987,
      "eval_loss": 3.6545450687408447,
      "eval_runtime": 169.7082,
      "eval_samples_per_second": 2.328,
      "eval_steps_per_second": 0.583,
      "step": 60
    },
    {
      "epoch": 0.362400906002265,
      "grad_norm": 1.4762500524520874,
      "learning_rate": 1.6969696969696972e-05,
      "loss": 3.4114,
      "step": 80
    },
    {
      "epoch": 0.362400906002265,
      "eval_loss": 3.1997323036193848,
      "eval_runtime": 169.459,
      "eval_samples_per_second": 2.331,
      "eval_steps_per_second": 0.584,
      "step": 80
    },
    {
      "epoch": 0.45300113250283125,
      "grad_norm": 1.2465909719467163,
      "learning_rate": 1.616161616161616e-05,
      "loss": 3.0527,
      "step": 100
    },
    {
      "epoch": 0.45300113250283125,
      "eval_loss": 2.9332973957061768,
      "eval_runtime": 169.4702,
      "eval_samples_per_second": 2.331,
      "eval_steps_per_second": 0.584,
      "step": 100
    },
    {
      "epoch": 0.5436013590033975,
      "grad_norm": 1.2837079763412476,
      "learning_rate": 1.5353535353535354e-05,
      "loss": 2.8401,
      "step": 120
    },
    {
      "epoch": 0.5436013590033975,
      "eval_loss": 2.765261173248291,
      "eval_runtime": 169.5654,
      "eval_samples_per_second": 2.329,
      "eval_steps_per_second": 0.584,
      "step": 120
    },
    {
      "epoch": 0.6342015855039638,
      "grad_norm": 1.069353699684143,
      "learning_rate": 1.4545454545454546e-05,
      "loss": 2.7202,
      "step": 140
    },
    {
      "epoch": 0.6342015855039638,
      "eval_loss": 2.654095411300659,
      "eval_runtime": 169.5632,
      "eval_samples_per_second": 2.33,
      "eval_steps_per_second": 0.584,
      "step": 140
    },
    {
      "epoch": 0.72480181200453,
      "grad_norm": 1.0665814876556396,
      "learning_rate": 1.3737373737373739e-05,
      "loss": 2.605,
      "step": 160
    },
    {
      "epoch": 0.72480181200453,
      "eval_loss": 2.576014995574951,
      "eval_runtime": 169.9075,
      "eval_samples_per_second": 2.325,
      "eval_steps_per_second": 0.583,
      "step": 160
    },
    {
      "epoch": 0.8154020385050963,
      "grad_norm": 1.076709508895874,
      "learning_rate": 1.2929292929292931e-05,
      "loss": 2.5533,
      "step": 180
    },
    {
      "epoch": 0.8154020385050963,
      "eval_loss": 2.519667148590088,
      "eval_runtime": 169.6071,
      "eval_samples_per_second": 2.329,
      "eval_steps_per_second": 0.584,
      "step": 180
    },
    {
      "epoch": 0.9060022650056625,
      "grad_norm": 1.0686030387878418,
      "learning_rate": 1.2121212121212122e-05,
      "loss": 2.5004,
      "step": 200
    },
    {
      "epoch": 0.9060022650056625,
      "eval_loss": 2.4773340225219727,
      "eval_runtime": 169.6567,
      "eval_samples_per_second": 2.328,
      "eval_steps_per_second": 0.584,
      "step": 200
    },
    {
      "epoch": 0.9966024915062288,
      "grad_norm": 1.1253015995025635,
      "learning_rate": 1.1313131313131314e-05,
      "loss": 2.4613,
      "step": 220
    },
    {
      "epoch": 0.9966024915062288,
      "eval_loss": 2.444694995880127,
      "eval_runtime": 169.7657,
      "eval_samples_per_second": 2.327,
      "eval_steps_per_second": 0.583,
      "step": 220
    },
    {
      "epoch": 1.087202718006795,
      "grad_norm": 1.1171083450317383,
      "learning_rate": 1.0505050505050507e-05,
      "loss": 2.4456,
      "step": 240
    },
    {
      "epoch": 1.087202718006795,
      "eval_loss": 2.4184916019439697,
      "eval_runtime": 169.7027,
      "eval_samples_per_second": 2.328,
      "eval_steps_per_second": 0.583,
      "step": 240
    },
    {
      "epoch": 1.1778029445073612,
      "grad_norm": 1.1789259910583496,
      "learning_rate": 9.696969696969698e-06,
      "loss": 2.4151,
      "step": 260
    },
    {
      "epoch": 1.1778029445073612,
      "eval_loss": 2.397007465362549,
      "eval_runtime": 169.6356,
      "eval_samples_per_second": 2.329,
      "eval_steps_per_second": 0.584,
      "step": 260
    },
    {
      "epoch": 1.2684031710079275,
      "grad_norm": 1.1507657766342163,
      "learning_rate": 8.888888888888888e-06,
      "loss": 2.3943,
      "step": 280
    },
    {
      "epoch": 1.2684031710079275,
      "eval_loss": 2.3794679641723633,
      "eval_runtime": 169.6152,
      "eval_samples_per_second": 2.329,
      "eval_steps_per_second": 0.584,
      "step": 280
    },
    {
      "epoch": 1.3590033975084936,
      "grad_norm": 1.1052231788635254,
      "learning_rate": 8.08080808080808e-06,
      "loss": 2.3621,
      "step": 300
    },
    {
      "epoch": 1.3590033975084936,
      "eval_loss": 2.3650312423706055,
      "eval_runtime": 169.6247,
      "eval_samples_per_second": 2.329,
      "eval_steps_per_second": 0.584,
      "step": 300
    },
    {
      "epoch": 1.44960362400906,
      "grad_norm": 1.156396508216858,
      "learning_rate": 7.272727272727273e-06,
      "loss": 2.3475,
      "step": 320
    },
    {
      "epoch": 1.44960362400906,
      "eval_loss": 2.3532702922821045,
      "eval_runtime": 169.6216,
      "eval_samples_per_second": 2.329,
      "eval_steps_per_second": 0.584,
      "step": 320
    },
    {
      "epoch": 1.5402038505096263,
      "grad_norm": 1.1324844360351562,
      "learning_rate": 6.464646464646466e-06,
      "loss": 2.339,
      "step": 340
    },
    {
      "epoch": 1.5402038505096263,
      "eval_loss": 2.342968702316284,
      "eval_runtime": 169.6943,
      "eval_samples_per_second": 2.328,
      "eval_steps_per_second": 0.583,
      "step": 340
    },
    {
      "epoch": 1.6308040770101924,
      "grad_norm": 1.2476097345352173,
      "learning_rate": 5.656565656565657e-06,
      "loss": 2.3247,
      "step": 360
    },
    {
      "epoch": 1.6308040770101924,
      "eval_loss": 2.334242343902588,
      "eval_runtime": 169.8944,
      "eval_samples_per_second": 2.325,
      "eval_steps_per_second": 0.583,
      "step": 360
    },
    {
      "epoch": 1.721404303510759,
      "grad_norm": 1.1416445970535278,
      "learning_rate": 4.848484848484849e-06,
      "loss": 2.3335,
      "step": 380
    },
    {
      "epoch": 1.721404303510759,
      "eval_loss": 2.326986789703369,
      "eval_runtime": 169.8042,
      "eval_samples_per_second": 2.326,
      "eval_steps_per_second": 0.583,
      "step": 380
    },
    {
      "epoch": 1.812004530011325,
      "grad_norm": 1.1464892625808716,
      "learning_rate": 4.04040404040404e-06,
      "loss": 2.3007,
      "step": 400
    },
    {
      "epoch": 1.812004530011325,
      "eval_loss": 2.321030378341675,
      "eval_runtime": 169.8884,
      "eval_samples_per_second": 2.325,
      "eval_steps_per_second": 0.583,
      "step": 400
    },
    {
      "epoch": 1.9026047565118913,
      "grad_norm": 1.1699483394622803,
      "learning_rate": 3.232323232323233e-06,
      "loss": 2.3095,
      "step": 420
    },
    {
      "epoch": 1.9026047565118913,
      "eval_loss": 2.3161513805389404,
      "eval_runtime": 169.7467,
      "eval_samples_per_second": 2.327,
      "eval_steps_per_second": 0.583,
      "step": 420
    },
    {
      "epoch": 1.9932049830124576,
      "grad_norm": 1.1978620290756226,
      "learning_rate": 2.4242424242424244e-06,
      "loss": 2.3093,
      "step": 440
    },
    {
      "epoch": 1.9932049830124576,
      "eval_loss": 2.312627077102661,
      "eval_runtime": 170.0514,
      "eval_samples_per_second": 2.323,
      "eval_steps_per_second": 0.582,
      "step": 440
    },
    {
      "epoch": 2.0838052095130237,
      "grad_norm": 1.23793363571167,
      "learning_rate": 1.6161616161616164e-06,
      "loss": 2.327,
      "step": 460
    },
    {
      "epoch": 2.0838052095130237,
      "eval_loss": 2.309922933578491,
      "eval_runtime": 169.6935,
      "eval_samples_per_second": 2.328,
      "eval_steps_per_second": 0.583,
      "step": 460
    },
    {
      "epoch": 2.17440543601359,
      "grad_norm": 1.2800756692886353,
      "learning_rate": 8.080808080808082e-07,
      "loss": 2.3005,
      "step": 480
    },
    {
      "epoch": 2.17440543601359,
      "eval_loss": 2.3085172176361084,
      "eval_runtime": 169.7305,
      "eval_samples_per_second": 2.327,
      "eval_steps_per_second": 0.583,
      "step": 480
    },
    {
      "epoch": 2.2650056625141564,
      "grad_norm": 1.1457535028457642,
      "learning_rate": 0.0,
      "loss": 2.2871,
      "step": 500
    },
    {
      "epoch": 2.2650056625141564,
      "eval_loss": 2.3081250190734863,
      "eval_runtime": 169.7694,
      "eval_samples_per_second": 2.327,
      "eval_steps_per_second": 0.583,
      "step": 500
    }
  ],
  "logging_steps": 20,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 20,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.75343643596161e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}