{
  "best_metric": 3.715158224105835,
  "best_model_checkpoint": "runs/checkpoint-1000",
  "epoch": 4.587155963302752,
  "eval_steps": 100,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09,
      "learning_rate": 6.666666666666667e-06,
      "loss": 6.9799,
      "step": 20
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 6.7268,
      "step": 40
    },
    {
      "epoch": 0.28,
      "learning_rate": 2e-05,
      "loss": 6.4767,
      "step": 60
    },
    {
      "epoch": 0.37,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 6.3317,
      "step": 80
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 6.2151,
      "step": 100
    },
    {
      "epoch": 0.46,
      "eval_accuracy": 0.010721727291912753,
      "eval_loss": 6.165957927703857,
      "eval_runtime": 954.4032,
      "eval_samples_per_second": 3.09,
      "eval_steps_per_second": 0.065,
      "step": 100
    },
    {
      "epoch": 0.55,
      "learning_rate": 4e-05,
      "loss": 6.13,
      "step": 120
    },
    {
      "epoch": 0.64,
      "learning_rate": 4.666666666666667e-05,
      "loss": 6.0514,
      "step": 140
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.333333333333333e-05,
      "loss": 5.9384,
      "step": 160
    },
    {
      "epoch": 0.83,
      "learning_rate": 6e-05,
      "loss": 5.679,
      "step": 180
    },
    {
      "epoch": 0.92,
      "learning_rate": 6.666666666666667e-05,
      "loss": 5.2673,
      "step": 200
    },
    {
      "epoch": 0.92,
      "eval_accuracy": 0.0003641545871791163,
      "eval_loss": 4.998520851135254,
      "eval_runtime": 3395.6647,
      "eval_samples_per_second": 0.868,
      "eval_steps_per_second": 0.018,
      "step": 200
    },
    {
      "epoch": 1.01,
      "learning_rate": 7.333333333333333e-05,
      "loss": 4.8258,
      "step": 220
    },
    {
      "epoch": 1.1,
      "learning_rate": 8e-05,
      "loss": 4.6404,
      "step": 240
    },
    {
      "epoch": 1.19,
      "learning_rate": 8.666666666666667e-05,
      "loss": 4.5153,
      "step": 260
    },
    {
      "epoch": 1.28,
      "learning_rate": 9.333333333333334e-05,
      "loss": 4.4108,
      "step": 280
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0001,
      "loss": 4.3584,
      "step": 300
    },
    {
      "epoch": 1.38,
      "eval_accuracy": 2.3712391723291295e-06,
      "eval_loss": 4.306852340698242,
      "eval_runtime": 3914.0936,
      "eval_samples_per_second": 0.753,
      "eval_steps_per_second": 0.016,
      "step": 300
    },
    {
      "epoch": 1.47,
      "learning_rate": 9.979871469976196e-05,
      "loss": 4.2798,
      "step": 320
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.919647942993148e-05,
      "loss": 4.2084,
      "step": 340
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.819814303479267e-05,
      "loss": 4.173,
      "step": 360
    },
    {
      "epoch": 1.74,
      "learning_rate": 9.681174353198687e-05,
      "loss": 4.1324,
      "step": 380
    },
    {
      "epoch": 1.83,
      "learning_rate": 9.504844339512095e-05,
      "loss": 4.0826,
      "step": 400
    },
    {
      "epoch": 1.83,
      "eval_accuracy": 8.299337103151952e-05,
      "eval_loss": 4.064357280731201,
      "eval_runtime": 625.6747,
      "eval_samples_per_second": 4.713,
      "eval_steps_per_second": 0.099,
      "step": 400
    },
    {
      "epoch": 1.93,
      "learning_rate": 9.292243968009331e-05,
      "loss": 4.0714,
      "step": 420
    },
    {
      "epoch": 2.02,
      "learning_rate": 9.045084971874738e-05,
      "loss": 4.0267,
      "step": 440
    },
    {
      "epoch": 2.11,
      "learning_rate": 8.765357330018056e-05,
      "loss": 3.9907,
      "step": 460
    },
    {
      "epoch": 2.2,
      "learning_rate": 8.455313244934324e-05,
      "loss": 3.9733,
      "step": 480
    },
    {
      "epoch": 2.29,
      "learning_rate": 8.117449009293668e-05,
      "loss": 3.9506,
      "step": 500
    },
    {
      "epoch": 2.29,
      "eval_accuracy": 0.000247963867734989,
      "eval_loss": 3.9375269412994385,
      "eval_runtime": 646.8101,
      "eval_samples_per_second": 4.559,
      "eval_steps_per_second": 0.096,
      "step": 500
    },
    {
      "epoch": 2.39,
      "learning_rate": 7.754484907260513e-05,
      "loss": 3.9255,
      "step": 520
    },
    {
      "epoch": 2.48,
      "learning_rate": 7.369343312364993e-05,
      "loss": 3.9064,
      "step": 540
    },
    {
      "epoch": 2.57,
      "learning_rate": 6.965125158269619e-05,
      "loss": 3.8806,
      "step": 560
    },
    {
      "epoch": 2.66,
      "learning_rate": 6.545084971874738e-05,
      "loss": 3.89,
      "step": 580
    },
    {
      "epoch": 2.75,
      "learning_rate": 6.112604669781572e-05,
      "loss": 3.8576,
      "step": 600
    },
    {
      "epoch": 2.75,
      "eval_accuracy": 0.0002405114017648117,
      "eval_loss": 3.8471720218658447,
      "eval_runtime": 641.884,
      "eval_samples_per_second": 4.594,
      "eval_steps_per_second": 0.097,
      "step": 600
    },
    {
      "epoch": 2.84,
      "learning_rate": 5.6711663290882776e-05,
      "loss": 3.8314,
      "step": 620
    },
    {
      "epoch": 2.94,
      "learning_rate": 5.2243241517525754e-05,
      "loss": 3.8356,
      "step": 640
    },
    {
      "epoch": 3.03,
      "learning_rate": 4.775675848247427e-05,
      "loss": 3.8056,
      "step": 660
    },
    {
      "epoch": 3.12,
      "learning_rate": 4.328833670911724e-05,
      "loss": 3.7761,
      "step": 680
    },
    {
      "epoch": 3.21,
      "learning_rate": 3.887395330218429e-05,
      "loss": 3.7937,
      "step": 700
    },
    {
      "epoch": 3.21,
      "eval_accuracy": 0.00025236759762645736,
      "eval_loss": 3.7817437648773193,
      "eval_runtime": 1546.8275,
      "eval_samples_per_second": 1.906,
      "eval_steps_per_second": 0.04,
      "step": 700
    },
    {
      "epoch": 3.3,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 3.7696,
      "step": 720
    },
    {
      "epoch": 3.39,
      "learning_rate": 3.0348748417303823e-05,
      "loss": 3.7522,
      "step": 740
    },
    {
      "epoch": 3.49,
      "learning_rate": 2.630656687635007e-05,
      "loss": 3.743,
      "step": 760
    },
    {
      "epoch": 3.58,
      "learning_rate": 2.245515092739488e-05,
      "loss": 3.7389,
      "step": 780
    },
    {
      "epoch": 3.67,
      "learning_rate": 1.8825509907063327e-05,
      "loss": 3.7287,
      "step": 800
    },
    {
      "epoch": 3.67,
      "eval_accuracy": 0.00031435856456020456,
      "eval_loss": 3.7385406494140625,
      "eval_runtime": 664.5264,
      "eval_samples_per_second": 4.438,
      "eval_steps_per_second": 0.093,
      "step": 800
    },
    {
      "epoch": 3.76,
      "learning_rate": 1.544686755065677e-05,
      "loss": 3.7229,
      "step": 820
    },
    {
      "epoch": 3.85,
      "learning_rate": 1.2346426699819458e-05,
      "loss": 3.7138,
      "step": 840
    },
    {
      "epoch": 3.94,
      "learning_rate": 9.549150281252633e-06,
      "loss": 3.7314,
      "step": 860
    },
    {
      "epoch": 4.04,
      "learning_rate": 7.077560319906695e-06,
      "loss": 3.7059,
      "step": 880
    },
    {
      "epoch": 4.13,
      "learning_rate": 4.951556604879048e-06,
      "loss": 3.7205,
      "step": 900
    },
    {
      "epoch": 4.13,
      "eval_accuracy": 0.0003251985150622806,
      "eval_loss": 3.7189857959747314,
      "eval_runtime": 713.2391,
      "eval_samples_per_second": 4.135,
      "eval_steps_per_second": 0.087,
      "step": 900
    },
    {
      "epoch": 4.22,
      "learning_rate": 3.18825646801314e-06,
      "loss": 3.71,
      "step": 920
    },
    {
      "epoch": 4.31,
      "learning_rate": 1.8018569652073381e-06,
      "loss": 3.6848,
      "step": 940
    },
    {
      "epoch": 4.4,
      "learning_rate": 8.035205700685167e-07,
      "loss": 3.6929,
      "step": 960
    },
    {
      "epoch": 4.5,
      "learning_rate": 2.012853002380466e-07,
      "loss": 3.7022,
      "step": 980
    },
    {
      "epoch": 4.59,
      "learning_rate": 0.0,
      "loss": 3.6858,
      "step": 1000
    },
    {
      "epoch": 4.59,
      "eval_accuracy": 0.00034552342225367314,
      "eval_loss": 3.715158224105835,
      "eval_runtime": 606.7664,
      "eval_samples_per_second": 4.86,
      "eval_steps_per_second": 0.102,
      "step": 1000
    }
  ],
  "logging_steps": 20,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 1000,
  "total_flos": 9439661425655808.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}