{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.62962962962963,
  "eval_steps": 500,
  "global_step": 130,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07407407407407407,
      "grad_norm": 178.0,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 48.7098,
      "step": 1
    },
    {
      "epoch": 0.37037037037037035,
      "grad_norm": 141.0,
      "learning_rate": 7.692307692307693e-05,
      "loss": 49.2488,
      "step": 5
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 17.75,
      "learning_rate": 0.00015384615384615385,
      "loss": 31.311,
      "step": 10
    },
    {
      "epoch": 0.9629629629629629,
      "eval_loss": 11.023000717163086,
      "eval_runtime": 0.2736,
      "eval_samples_per_second": 36.555,
      "eval_steps_per_second": 3.655,
      "step": 13
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 10.0,
      "learning_rate": 0.00019985583705641418,
      "loss": 23.1246,
      "step": 15
    },
    {
      "epoch": 1.4814814814814814,
      "grad_norm": 4.46875,
      "learning_rate": 0.00019823877374156647,
      "loss": 20.1941,
      "step": 20
    },
    {
      "epoch": 1.8518518518518519,
      "grad_norm": 6.0625,
      "learning_rate": 0.00019485364419471454,
      "loss": 19.0651,
      "step": 25
    },
    {
      "epoch": 2.0,
      "eval_loss": 7.434217929840088,
      "eval_runtime": 0.2352,
      "eval_samples_per_second": 42.524,
      "eval_steps_per_second": 4.252,
      "step": 27
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 11.5,
      "learning_rate": 0.0001897613727639014,
      "loss": 17.6692,
      "step": 30
    },
    {
      "epoch": 2.5925925925925926,
      "grad_norm": 18.625,
      "learning_rate": 0.00018305360832480117,
      "loss": 15.2716,
      "step": 35
    },
    {
      "epoch": 2.962962962962963,
      "grad_norm": 25.75,
      "learning_rate": 0.00017485107481711012,
      "loss": 11.34,
      "step": 40
    },
    {
      "epoch": 2.962962962962963,
      "eval_loss": 6.81179666519165,
      "eval_runtime": 0.2444,
      "eval_samples_per_second": 40.915,
      "eval_steps_per_second": 4.091,
      "step": 40
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 26.125,
      "learning_rate": 0.0001653013984983585,
      "loss": 6.3302,
      "step": 45
    },
    {
      "epoch": 3.7037037037037037,
      "grad_norm": 7.375,
      "learning_rate": 0.00015457645101945046,
      "loss": 3.0136,
      "step": 50
    },
    {
      "epoch": 4.0,
      "eval_loss": 3.630751848220825,
      "eval_runtime": 0.2376,
      "eval_samples_per_second": 42.085,
      "eval_steps_per_second": 4.208,
      "step": 54
    },
    {
      "epoch": 4.074074074074074,
      "grad_norm": 5.59375,
      "learning_rate": 0.00014286925614030542,
      "loss": 2.4006,
      "step": 55
    },
    {
      "epoch": 4.444444444444445,
      "grad_norm": 2.859375,
      "learning_rate": 0.0001303905157574247,
      "loss": 2.0441,
      "step": 60
    },
    {
      "epoch": 4.814814814814815,
      "grad_norm": 1.046875,
      "learning_rate": 0.00011736481776669306,
      "loss": 1.7786,
      "step": 65
    },
    {
      "epoch": 4.962962962962963,
      "eval_loss": 3.097276210784912,
      "eval_runtime": 0.2571,
      "eval_samples_per_second": 38.893,
      "eval_steps_per_second": 3.889,
      "step": 67
    },
    {
      "epoch": 5.185185185185185,
      "grad_norm": 0.703125,
      "learning_rate": 0.00010402659401094152,
      "loss": 1.6515,
      "step": 70
    },
    {
      "epoch": 5.555555555555555,
      "grad_norm": 0.69921875,
      "learning_rate": 9.061590105968208e-05,
      "loss": 1.5441,
      "step": 75
    },
    {
      "epoch": 5.925925925925926,
      "grad_norm": 1.5,
      "learning_rate": 7.73740997570278e-05,
      "loss": 1.4865,
      "step": 80
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.9240853786468506,
      "eval_runtime": 0.2376,
      "eval_samples_per_second": 42.082,
      "eval_steps_per_second": 4.208,
      "step": 81
    },
    {
      "epoch": 6.296296296296296,
      "grad_norm": 1.2421875,
      "learning_rate": 6.453951129574644e-05,
      "loss": 1.448,
      "step": 85
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 0.5,
      "learning_rate": 5.234312799786921e-05,
      "loss": 1.4036,
      "step": 90
    },
    {
      "epoch": 6.962962962962963,
      "eval_loss": 2.864516019821167,
      "eval_runtime": 0.263,
      "eval_samples_per_second": 38.017,
      "eval_steps_per_second": 3.802,
      "step": 94
    },
    {
      "epoch": 7.037037037037037,
      "grad_norm": 0.484375,
      "learning_rate": 4.100445599768774e-05,
      "loss": 1.3829,
      "step": 95
    },
    {
      "epoch": 7.407407407407407,
      "grad_norm": 0.4765625,
      "learning_rate": 3.072756464904006e-05,
      "loss": 1.3704,
      "step": 100
    },
    {
      "epoch": 7.777777777777778,
      "grad_norm": 0.51953125,
      "learning_rate": 2.1697413758237784e-05,
      "loss": 1.3424,
      "step": 105
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.851011276245117,
      "eval_runtime": 0.2392,
      "eval_samples_per_second": 41.815,
      "eval_steps_per_second": 4.181,
      "step": 108
    },
    {
      "epoch": 8.148148148148149,
      "grad_norm": 0.447265625,
      "learning_rate": 1.4076524743778319e-05,
      "loss": 1.3438,
      "step": 110
    },
    {
      "epoch": 8.518518518518519,
      "grad_norm": 0.451171875,
      "learning_rate": 8.002055634117578e-06,
      "loss": 1.3448,
      "step": 115
    },
    {
      "epoch": 8.88888888888889,
      "grad_norm": 0.54296875,
      "learning_rate": 3.5833325466437694e-06,
      "loss": 1.3298,
      "step": 120
    },
    {
      "epoch": 8.962962962962964,
      "eval_loss": 2.8409743309020996,
      "eval_runtime": 0.2622,
      "eval_samples_per_second": 38.132,
      "eval_steps_per_second": 3.813,
      "step": 121
    },
    {
      "epoch": 9.25925925925926,
      "grad_norm": 0.451171875,
      "learning_rate": 8.998820754091531e-07,
      "loss": 1.3428,
      "step": 125
    },
    {
      "epoch": 9.62962962962963,
      "grad_norm": 0.40625,
      "learning_rate": 0.0,
      "loss": 1.3245,
      "step": 130
    },
    {
      "epoch": 9.62962962962963,
      "eval_loss": 2.839648723602295,
      "eval_runtime": 0.2376,
      "eval_samples_per_second": 42.081,
      "eval_steps_per_second": 4.208,
      "step": 130
    },
    {
      "epoch": 9.62962962962963,
      "step": 130,
      "total_flos": 3.9639312421093376e+17,
      "train_loss": 8.499957752227782,
      "train_runtime": 318.2463,
      "train_samples_per_second": 26.238,
      "train_steps_per_second": 0.408
    }
  ],
  "logging_steps": 5,
  "max_steps": 130,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 100,
  "total_flos": 3.9639312421093376e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}