{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.06798339773865751,
  "eval_steps": 25,
  "global_step": 475,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 0.6931123733520508,
      "learning_rate": 0.0001988606301829365,
      "loss": 3.3633,
      "step": 25
    },
    {
      "epoch": 0.0,
      "eval_loss": 2.791308879852295,
      "eval_runtime": 1766.1893,
      "eval_samples_per_second": 4.028,
      "eval_steps_per_second": 0.504,
      "step": 25
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.8134792447090149,
      "learning_rate": 0.00019527921915838827,
      "loss": 2.8231,
      "step": 50
    },
    {
      "epoch": 0.01,
      "eval_loss": 2.6491129398345947,
      "eval_runtime": 1766.0434,
      "eval_samples_per_second": 4.028,
      "eval_steps_per_second": 0.504,
      "step": 50
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.7844312191009521,
      "learning_rate": 0.00018934232439329787,
      "loss": 2.7094,
      "step": 75
    },
    {
      "epoch": 0.01,
      "eval_loss": 2.595423698425293,
      "eval_runtime": 1770.4727,
      "eval_samples_per_second": 4.018,
      "eval_steps_per_second": 0.503,
      "step": 75
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.7818688154220581,
      "learning_rate": 0.00018119671723205706,
      "loss": 2.6759,
      "step": 100
    },
    {
      "epoch": 0.01,
      "eval_loss": 2.5623528957366943,
      "eval_runtime": 1764.724,
      "eval_samples_per_second": 4.031,
      "eval_steps_per_second": 0.504,
      "step": 100
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.9374421238899231,
      "learning_rate": 0.00017104377259560488,
      "loss": 2.6361,
      "step": 125
    },
    {
      "epoch": 0.02,
      "eval_loss": 2.5327322483062744,
      "eval_runtime": 1766.346,
      "eval_samples_per_second": 4.028,
      "eval_steps_per_second": 0.504,
      "step": 125
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.8157618641853333,
      "learning_rate": 0.00015913449060989774,
      "loss": 2.6192,
      "step": 150
    },
    {
      "epoch": 0.02,
      "eval_loss": 2.5096042156219482,
      "eval_runtime": 1765.5704,
      "eval_samples_per_second": 4.029,
      "eval_steps_per_second": 0.504,
      "step": 150
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.8904098272323608,
      "learning_rate": 0.00014576329140480925,
      "loss": 2.5547,
      "step": 175
    },
    {
      "epoch": 0.03,
      "eval_loss": 2.508963108062744,
      "eval_runtime": 1767.6823,
      "eval_samples_per_second": 4.024,
      "eval_steps_per_second": 0.503,
      "step": 175
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.8415258526802063,
      "learning_rate": 0.00013126073648784688,
      "loss": 2.5803,
      "step": 200
    },
    {
      "epoch": 0.03,
      "eval_loss": 2.484498977661133,
      "eval_runtime": 1764.6855,
      "eval_samples_per_second": 4.031,
      "eval_steps_per_second": 0.504,
      "step": 200
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.8273302912712097,
      "learning_rate": 0.00011598535663418884,
      "loss": 2.5275,
      "step": 225
    },
    {
      "epoch": 0.03,
      "eval_loss": 2.4777684211730957,
      "eval_runtime": 1764.516,
      "eval_samples_per_second": 4.032,
      "eval_steps_per_second": 0.504,
      "step": 225
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.8461028933525085,
      "learning_rate": 0.000100314788323161,
      "loss": 2.5504,
      "step": 250
    },
    {
      "epoch": 0.04,
      "eval_loss": 2.4680240154266357,
      "eval_runtime": 1764.412,
      "eval_samples_per_second": 4.032,
      "eval_steps_per_second": 0.504,
      "step": 250
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.9055793881416321,
      "learning_rate": 8.46364378453242e-05,
      "loss": 2.5447,
      "step": 275
    },
    {
      "epoch": 0.04,
      "eval_loss": 2.4592506885528564,
      "eval_runtime": 1763.6878,
      "eval_samples_per_second": 4.034,
      "eval_steps_per_second": 0.505,
      "step": 275
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.0102213621139526,
      "learning_rate": 6.933790388122256e-05,
      "loss": 2.5308,
      "step": 300
    },
    {
      "epoch": 0.04,
      "eval_loss": 2.457402467727661,
      "eval_runtime": 1763.3112,
      "eval_samples_per_second": 4.034,
      "eval_steps_per_second": 0.505,
      "step": 300
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.9688016176223755,
      "learning_rate": 5.479739532388526e-05,
      "loss": 2.5243,
      "step": 325
    },
    {
      "epoch": 0.05,
      "eval_loss": 2.4439914226531982,
      "eval_runtime": 1763.1627,
      "eval_samples_per_second": 4.035,
      "eval_steps_per_second": 0.505,
      "step": 325
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.9817106127738953,
      "learning_rate": 4.137438123475662e-05,
      "loss": 2.4967,
      "step": 350
    },
    {
      "epoch": 0.05,
      "eval_loss": 2.4434900283813477,
      "eval_runtime": 1762.4256,
      "eval_samples_per_second": 4.036,
      "eval_steps_per_second": 0.505,
      "step": 350
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.9285837411880493,
      "learning_rate": 2.9400704083950335e-05,
      "loss": 2.4979,
      "step": 375
    },
    {
      "epoch": 0.05,
      "eval_loss": 2.4380979537963867,
      "eval_runtime": 1761.6148,
      "eval_samples_per_second": 4.038,
      "eval_steps_per_second": 0.505,
      "step": 375
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.8453341722488403,
      "learning_rate": 1.917237597245065e-05,
      "loss": 2.5183,
      "step": 400
    },
    {
      "epoch": 0.06,
      "eval_loss": 2.437168598175049,
      "eval_runtime": 1763.4013,
      "eval_samples_per_second": 4.034,
      "eval_steps_per_second": 0.505,
      "step": 400
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.091811180114746,
      "learning_rate": 1.0942260649272229e-05,
      "loss": 2.4867,
      "step": 425
    },
    {
      "epoch": 0.06,
      "eval_loss": 2.4306600093841553,
      "eval_runtime": 1759.0574,
      "eval_samples_per_second": 4.044,
      "eval_steps_per_second": 0.506,
      "step": 425
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.9597497582435608,
      "learning_rate": 4.91382223804836e-06,
      "loss": 2.474,
      "step": 450
    },
    {
      "epoch": 0.06,
      "eval_loss": 2.4304585456848145,
      "eval_runtime": 1763.2619,
      "eval_samples_per_second": 4.035,
      "eval_steps_per_second": 0.505,
      "step": 450
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.9459497928619385,
      "learning_rate": 1.2360952164325311e-06,
      "loss": 2.4979,
      "step": 475
    },
    {
      "epoch": 0.07,
      "eval_loss": 2.4304075241088867,
      "eval_runtime": 1762.5181,
      "eval_samples_per_second": 4.036,
      "eval_steps_per_second": 0.505,
      "step": 475
    }
  ],
  "logging_steps": 25,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "total_flos": 8045660258064000.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}