{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.8656716417910446,
  "eval_steps": 2,
  "global_step": 24,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12,
      "learning_rate": 1e-05,
      "loss": 0.9774,
      "step": 1
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.953429730181653e-06,
      "loss": 0.9731,
      "step": 2
    },
    {
      "epoch": 0.24,
      "eval_loss": 0.9532256126403809,
      "eval_runtime": 45.7362,
      "eval_samples_per_second": 1.968,
      "eval_steps_per_second": 1.968,
      "step": 2
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.814586436738998e-06,
      "loss": 0.9614,
      "step": 3
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.586056507527266e-06,
      "loss": 0.999,
      "step": 4
    },
    {
      "epoch": 0.48,
      "eval_loss": 0.947267472743988,
      "eval_runtime": 45.4678,
      "eval_samples_per_second": 1.979,
      "eval_steps_per_second": 1.979,
      "step": 4
    },
    {
      "epoch": 0.6,
      "learning_rate": 9.272097022732444e-06,
      "loss": 0.9813,
      "step": 5
    },
    {
      "epoch": 0.72,
      "learning_rate": 8.8785564535221e-06,
      "loss": 0.9045,
      "step": 6
    },
    {
      "epoch": 0.72,
      "eval_loss": 0.9444782733917236,
      "eval_runtime": 45.6046,
      "eval_samples_per_second": 1.973,
      "eval_steps_per_second": 1.973,
      "step": 6
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.412765716093273e-06,
      "loss": 0.9201,
      "step": 7
    },
    {
      "epoch": 0.96,
      "learning_rate": 7.883401610574338e-06,
      "loss": 0.9041,
      "step": 8
    },
    {
      "epoch": 0.96,
      "eval_loss": 0.9425725936889648,
      "eval_runtime": 45.4409,
      "eval_samples_per_second": 1.981,
      "eval_steps_per_second": 1.981,
      "step": 8
    },
    {
      "epoch": 1.07,
      "learning_rate": 7.300325188655762e-06,
      "loss": 0.9619,
      "step": 9
    },
    {
      "epoch": 1.19,
      "learning_rate": 6.674398060854931e-06,
      "loss": 0.9645,
      "step": 10
    },
    {
      "epoch": 1.19,
      "eval_loss": 0.9409478306770325,
      "eval_runtime": 45.4535,
      "eval_samples_per_second": 1.98,
      "eval_steps_per_second": 1.98,
      "step": 10
    },
    {
      "epoch": 1.31,
      "learning_rate": 6.0172800652631706e-06,
      "loss": 0.9642,
      "step": 11
    },
    {
      "epoch": 1.43,
      "learning_rate": 5.341212066823356e-06,
      "loss": 0.8638,
      "step": 12
    },
    {
      "epoch": 1.43,
      "eval_loss": 0.9403448104858398,
      "eval_runtime": 45.3553,
      "eval_samples_per_second": 1.984,
      "eval_steps_per_second": 1.984,
      "step": 12
    },
    {
      "epoch": 1.55,
      "learning_rate": 4.6587879331766465e-06,
      "loss": 0.9883,
      "step": 13
    },
    {
      "epoch": 1.67,
      "learning_rate": 3.982719934736832e-06,
      "loss": 0.8556,
      "step": 14
    },
    {
      "epoch": 1.67,
      "eval_loss": 0.9399924278259277,
      "eval_runtime": 45.4921,
      "eval_samples_per_second": 1.978,
      "eval_steps_per_second": 1.978,
      "step": 14
    },
    {
      "epoch": 1.79,
      "learning_rate": 3.3256019391450696e-06,
      "loss": 0.9509,
      "step": 15
    },
    {
      "epoch": 1.91,
      "learning_rate": 2.6996748113442397e-06,
      "loss": 0.8926,
      "step": 16
    },
    {
      "epoch": 1.91,
      "eval_loss": 0.9397569298744202,
      "eval_runtime": 45.3853,
      "eval_samples_per_second": 1.983,
      "eval_steps_per_second": 1.983,
      "step": 16
    },
    {
      "epoch": 2.03,
      "learning_rate": 2.1165983894256647e-06,
      "loss": 0.984,
      "step": 17
    },
    {
      "epoch": 2.15,
      "learning_rate": 1.5872342839067305e-06,
      "loss": 0.9494,
      "step": 18
    },
    {
      "epoch": 2.15,
      "eval_loss": 0.9394004940986633,
      "eval_runtime": 45.4356,
      "eval_samples_per_second": 1.981,
      "eval_steps_per_second": 1.981,
      "step": 18
    },
    {
      "epoch": 2.27,
      "learning_rate": 1.1214435464779006e-06,
      "loss": 0.9571,
      "step": 19
    },
    {
      "epoch": 2.39,
      "learning_rate": 7.279029772675572e-07,
      "loss": 0.9486,
      "step": 20
    },
    {
      "epoch": 2.39,
      "eval_loss": 0.9396945238113403,
      "eval_runtime": 45.5,
      "eval_samples_per_second": 1.978,
      "eval_steps_per_second": 1.978,
      "step": 20
    },
    {
      "epoch": 2.51,
      "learning_rate": 4.139434924727359e-07,
      "loss": 0.8492,
      "step": 21
    },
    {
      "epoch": 2.63,
      "learning_rate": 1.8541356326100436e-07,
      "loss": 0.9434,
      "step": 22
    },
    {
      "epoch": 2.63,
      "eval_loss": 0.9391700029373169,
      "eval_runtime": 45.3894,
      "eval_samples_per_second": 1.983,
      "eval_steps_per_second": 1.983,
      "step": 22
    },
    {
      "epoch": 2.75,
      "learning_rate": 4.657026981834623e-08,
      "loss": 0.9995,
      "step": 23
    },
    {
      "epoch": 2.87,
      "learning_rate": 0.0,
      "loss": 0.9475,
      "step": 24
    },
    {
      "epoch": 2.87,
      "eval_loss": 0.9397835731506348,
      "eval_runtime": 45.4245,
      "eval_samples_per_second": 1.981,
      "eval_steps_per_second": 1.981,
      "step": 24
    },
    {
      "epoch": 2.87,
      "step": 24,
      "total_flos": 6.417806722282291e+16,
      "train_loss": 0.9433953066666921,
      "train_runtime": 1841.1427,
      "train_samples_per_second": 0.647,
      "train_steps_per_second": 0.013
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 24,
  "num_train_epochs": 3,
  "save_steps": 3,
  "total_flos": 6.417806722282291e+16,
  "trial_name": null,
  "trial_params": null
}