{
  "best_metric": 0.8851989507675171,
  "best_model_checkpoint": "outputs/checkpoint-185",
  "epoch": 9.997300944669366,
  "eval_steps": 500,
  "global_step": 463,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4318488529014845,
      "grad_norm": 0.5840024948120117,
      "learning_rate": 4e-05,
      "loss": 1.9547,
      "step": 20
    },
    {
      "epoch": 0.863697705802969,
      "grad_norm": 0.4871944487094879,
      "learning_rate": 8e-05,
      "loss": 1.8203,
      "step": 40
    },
    {
      "epoch": 0.9932523616734144,
      "eval_loss": 1.5045615434646606,
      "eval_runtime": 13.3054,
      "eval_samples_per_second": 27.959,
      "eval_steps_per_second": 3.532,
      "step": 46
    },
    {
      "epoch": 1.2955465587044535,
      "grad_norm": 0.6788378953933716,
      "learning_rate": 0.00012,
      "loss": 1.6913,
      "step": 60
    },
    {
      "epoch": 1.7273954116059378,
      "grad_norm": 0.8430602550506592,
      "learning_rate": 0.00016,
      "loss": 1.4654,
      "step": 80
    },
    {
      "epoch": 1.9865047233468287,
      "eval_loss": 1.10453200340271,
      "eval_runtime": 13.2625,
      "eval_samples_per_second": 28.049,
      "eval_steps_per_second": 3.544,
      "step": 92
    },
    {
      "epoch": 2.1592442645074224,
      "grad_norm": 1.1823046207427979,
      "learning_rate": 0.0002,
      "loss": 1.3454,
      "step": 100
    },
    {
      "epoch": 2.591093117408907,
      "grad_norm": 1.2263983488082886,
      "learning_rate": 0.00019957591414302984,
      "loss": 1.1674,
      "step": 120
    },
    {
      "epoch": 2.979757085020243,
      "eval_loss": 0.9334564208984375,
      "eval_runtime": 13.257,
      "eval_samples_per_second": 28.061,
      "eval_steps_per_second": 3.545,
      "step": 138
    },
    {
      "epoch": 3.0229419703103915,
      "grad_norm": 1.068957805633545,
      "learning_rate": 0.00019830725354840089,
      "loss": 1.1417,
      "step": 140
    },
    {
      "epoch": 3.454790823211876,
      "grad_norm": 1.226943850517273,
      "learning_rate": 0.00019620477863642276,
      "loss": 0.9287,
      "step": 160
    },
    {
      "epoch": 3.8866396761133606,
      "grad_norm": 0.9932326674461365,
      "learning_rate": 0.00019328632200459156,
      "loss": 0.9289,
      "step": 180
    },
    {
      "epoch": 3.9946018893387314,
      "eval_loss": 0.8851989507675171,
      "eval_runtime": 13.2564,
      "eval_samples_per_second": 28.062,
      "eval_steps_per_second": 3.545,
      "step": 185
    },
    {
      "epoch": 4.318488529014845,
      "grad_norm": 1.303941011428833,
      "learning_rate": 0.00018957663717654208,
      "loss": 0.7678,
      "step": 200
    },
    {
      "epoch": 4.75033738191633,
      "grad_norm": 1.4492546319961548,
      "learning_rate": 0.0001851071886496621,
      "loss": 0.7871,
      "step": 220
    },
    {
      "epoch": 4.987854251012146,
      "eval_loss": 0.8959786891937256,
      "eval_runtime": 13.2594,
      "eval_samples_per_second": 28.055,
      "eval_steps_per_second": 3.545,
      "step": 231
    },
    {
      "epoch": 5.182186234817814,
      "grad_norm": 1.1834120750427246,
      "learning_rate": 0.0001799158850221259,
      "loss": 0.7191,
      "step": 240
    },
    {
      "epoch": 5.614035087719298,
      "grad_norm": 1.1505727767944336,
      "learning_rate": 0.00017404675746288687,
      "loss": 0.6127,
      "step": 260
    },
    {
      "epoch": 5.98110661268556,
      "eval_loss": 0.9636960625648499,
      "eval_runtime": 13.2504,
      "eval_samples_per_second": 28.075,
      "eval_steps_per_second": 3.547,
      "step": 277
    },
    {
      "epoch": 6.045883940620783,
      "grad_norm": 0.9700413346290588,
      "learning_rate": 0.00016754958625175758,
      "loss": 0.5765,
      "step": 280
    },
    {
      "epoch": 6.477732793522267,
      "grad_norm": 1.064491868019104,
      "learning_rate": 0.00016047947855715714,
      "loss": 0.4534,
      "step": 300
    },
    {
      "epoch": 6.909581646423752,
      "grad_norm": 1.1159394979476929,
      "learning_rate": 0.00015289640103269625,
      "loss": 0.5077,
      "step": 320
    },
    {
      "epoch": 6.995951417004049,
      "eval_loss": 1.0732767581939697,
      "eval_runtime": 13.2691,
      "eval_samples_per_second": 28.035,
      "eval_steps_per_second": 3.542,
      "step": 324
    },
    {
      "epoch": 7.341430499325236,
      "grad_norm": 1.0435690879821777,
      "learning_rate": 0.00014486467119698357,
      "loss": 0.3868,
      "step": 340
    },
    {
      "epoch": 7.77327935222672,
      "grad_norm": 1.2058143615722656,
      "learning_rate": 0.0001364524119106257,
      "loss": 0.4055,
      "step": 360
    },
    {
      "epoch": 7.989203778677463,
      "eval_loss": 1.1590379476547241,
      "eval_runtime": 13.2669,
      "eval_samples_per_second": 28.04,
      "eval_steps_per_second": 3.543,
      "step": 370
    },
    {
      "epoch": 8.205128205128204,
      "grad_norm": 1.1939080953598022,
      "learning_rate": 0.00012773097357739288,
      "loss": 0.3305,
      "step": 380
    },
    {
      "epoch": 8.63697705802969,
      "grad_norm": 1.2853295803070068,
      "learning_rate": 0.00011877432897027637,
      "loss": 0.2833,
      "step": 400
    },
    {
      "epoch": 8.982456140350877,
      "eval_loss": 1.2504498958587646,
      "eval_runtime": 13.2621,
      "eval_samples_per_second": 28.05,
      "eval_steps_per_second": 3.544,
      "step": 416
    },
    {
      "epoch": 9.068825910931174,
      "grad_norm": 1.0106446743011475,
      "learning_rate": 0.00010965844581535178,
      "loss": 0.2861,
      "step": 420
    },
    {
      "epoch": 9.50067476383266,
      "grad_norm": 0.9899333715438843,
      "learning_rate": 0.00010046064245501518,
      "loss": 0.2101,
      "step": 440
    },
    {
      "epoch": 9.932523616734143,
      "grad_norm": 0.9647331833839417,
      "learning_rate": 9.125893205567273e-05,
      "loss": 0.2175,
      "step": 460
    },
    {
      "epoch": 9.997300944669366,
      "eval_loss": 1.3450562953948975,
      "eval_runtime": 13.2627,
      "eval_samples_per_second": 28.049,
      "eval_steps_per_second": 3.544,
      "step": 463
    }
  ],
  "logging_steps": 20,
  "max_steps": 782,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 17,
  "save_steps": 500,
  "total_flos": 3.5665061878738944e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}