{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.84251968503937,
  "global_step": 17500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.28,
      "learning_rate": 1.943757030371204e-05,
      "loss": 2.7683,
      "step": 500
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.8875140607424075e-05,
      "loss": 2.7344,
      "step": 1000
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.831271091113611e-05,
      "loss": 2.7195,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.712266206741333,
      "eval_runtime": 91.7259,
      "eval_samples_per_second": 174.433,
      "eval_steps_per_second": 10.902,
      "step": 1778
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.7750281214848145e-05,
      "loss": 2.6796,
      "step": 2000
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.7187851518560182e-05,
      "loss": 2.6488,
      "step": 2500
    },
    {
      "epoch": 1.69,
      "learning_rate": 1.6625421822272215e-05,
      "loss": 2.6372,
      "step": 3000
    },
    {
      "epoch": 1.97,
      "learning_rate": 1.6062992125984255e-05,
      "loss": 2.6415,
      "step": 3500
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.652708053588867,
      "eval_runtime": 91.7276,
      "eval_samples_per_second": 174.43,
      "eval_steps_per_second": 10.902,
      "step": 3556
    },
    {
      "epoch": 2.25,
      "learning_rate": 1.550056242969629e-05,
      "loss": 2.5601,
      "step": 4000
    },
    {
      "epoch": 2.53,
      "learning_rate": 1.4938132733408325e-05,
      "loss": 2.5407,
      "step": 4500
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.437570303712036e-05,
      "loss": 2.5308,
      "step": 5000
    },
    {
      "epoch": 3.0,
      "eval_loss": 2.6422665119171143,
      "eval_runtime": 91.7218,
      "eval_samples_per_second": 174.441,
      "eval_steps_per_second": 10.903,
      "step": 5334
    },
    {
      "epoch": 3.09,
      "learning_rate": 1.3813273340832395e-05,
      "loss": 2.4873,
      "step": 5500
    },
    {
      "epoch": 3.37,
      "learning_rate": 1.3250843644544434e-05,
      "loss": 2.3694,
      "step": 6000
    },
    {
      "epoch": 3.66,
      "learning_rate": 1.2688413948256469e-05,
      "loss": 2.3697,
      "step": 6500
    },
    {
      "epoch": 3.94,
      "learning_rate": 1.2125984251968505e-05,
      "loss": 2.3628,
      "step": 7000
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.685481548309326,
      "eval_runtime": 91.7025,
      "eval_samples_per_second": 174.477,
      "eval_steps_per_second": 10.905,
      "step": 7112
    },
    {
      "epoch": 4.22,
      "learning_rate": 1.156355455568054e-05,
      "loss": 2.1747,
      "step": 7500
    },
    {
      "epoch": 4.5,
      "learning_rate": 1.1001124859392579e-05,
      "loss": 2.1037,
      "step": 8000
    },
    {
      "epoch": 4.78,
      "learning_rate": 1.0438695163104614e-05,
      "loss": 2.1293,
      "step": 8500
    },
    {
      "epoch": 5.0,
      "eval_loss": 2.817112445831299,
      "eval_runtime": 91.7137,
      "eval_samples_per_second": 174.456,
      "eval_steps_per_second": 10.904,
      "step": 8890
    },
    {
      "epoch": 5.06,
      "learning_rate": 9.876265466816649e-06,
      "loss": 2.0307,
      "step": 9000
    },
    {
      "epoch": 5.34,
      "learning_rate": 9.313835770528684e-06,
      "loss": 1.7546,
      "step": 9500
    },
    {
      "epoch": 5.62,
      "learning_rate": 8.75140607424072e-06,
      "loss": 1.752,
      "step": 10000
    },
    {
      "epoch": 5.91,
      "learning_rate": 8.188976377952757e-06,
      "loss": 1.7463,
      "step": 10500
    },
    {
      "epoch": 6.0,
      "eval_loss": 3.0516531467437744,
      "eval_runtime": 91.7177,
      "eval_samples_per_second": 174.448,
      "eval_steps_per_second": 10.903,
      "step": 10668
    },
    {
      "epoch": 6.19,
      "learning_rate": 7.626546681664793e-06,
      "loss": 1.4932,
      "step": 11000
    },
    {
      "epoch": 6.47,
      "learning_rate": 7.064116985376829e-06,
      "loss": 1.3659,
      "step": 11500
    },
    {
      "epoch": 6.75,
      "learning_rate": 6.501687289088865e-06,
      "loss": 1.3485,
      "step": 12000
    },
    {
      "epoch": 7.0,
      "eval_loss": 3.350566864013672,
      "eval_runtime": 91.704,
      "eval_samples_per_second": 174.474,
      "eval_steps_per_second": 10.905,
      "step": 12446
    },
    {
      "epoch": 7.03,
      "learning_rate": 5.9392575928009e-06,
      "loss": 1.3203,
      "step": 12500
    },
    {
      "epoch": 7.31,
      "learning_rate": 5.376827896512936e-06,
      "loss": 1.0176,
      "step": 13000
    },
    {
      "epoch": 7.59,
      "learning_rate": 4.814398200224972e-06,
      "loss": 0.9999,
      "step": 13500
    },
    {
      "epoch": 7.87,
      "learning_rate": 4.251968503937008e-06,
      "loss": 0.9889,
      "step": 14000
    },
    {
      "epoch": 8.0,
      "eval_loss": 3.6804659366607666,
      "eval_runtime": 91.7068,
      "eval_samples_per_second": 174.469,
      "eval_steps_per_second": 10.904,
      "step": 14224
    },
    {
      "epoch": 8.16,
      "learning_rate": 3.689538807649044e-06,
      "loss": 0.8533,
      "step": 14500
    },
    {
      "epoch": 8.44,
      "learning_rate": 3.12710911136108e-06,
      "loss": 0.7315,
      "step": 15000
    },
    {
      "epoch": 8.72,
      "learning_rate": 2.5646794150731156e-06,
      "loss": 0.7194,
      "step": 15500
    },
    {
      "epoch": 9.0,
      "learning_rate": 2.002249718785152e-06,
      "loss": 0.7069,
      "step": 16000
    },
    {
      "epoch": 9.0,
      "eval_loss": 3.918808698654175,
      "eval_runtime": 91.7922,
      "eval_samples_per_second": 174.307,
      "eval_steps_per_second": 10.894,
      "step": 16002
    },
    {
      "epoch": 9.28,
      "learning_rate": 1.4398200224971882e-06,
      "loss": 0.5375,
      "step": 16500
    },
    {
      "epoch": 9.56,
      "learning_rate": 8.77390326209224e-07,
      "loss": 0.533,
      "step": 17000
    },
    {
      "epoch": 9.84,
      "learning_rate": 3.149606299212599e-07,
      "loss": 0.5276,
      "step": 17500
    }
  ],
  "max_steps": 17780,
  "num_train_epochs": 10,
  "total_flos": 9948470041336704.0,
  "trial_name": null,
  "trial_params": null
}