{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.47619047619047616,
  "eval_steps": 20,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "eval_loss": 1.0689338445663452,
      "eval_runtime": 9.952,
      "eval_samples_per_second": 3.818,
      "eval_steps_per_second": 0.502,
      "step": 20
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.3734177215189873e-05,
      "loss": 1.5255,
      "step": 25
    },
    {
      "epoch": 0.05,
      "eval_loss": 0.8571718335151672,
      "eval_runtime": 9.9124,
      "eval_samples_per_second": 3.834,
      "eval_steps_per_second": 0.504,
      "step": 40
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.2151898734177217e-05,
      "loss": 0.9309,
      "step": 50
    },
    {
      "epoch": 0.07,
      "eval_loss": 0.7807316780090332,
      "eval_runtime": 9.9144,
      "eval_samples_per_second": 3.833,
      "eval_steps_per_second": 0.504,
      "step": 60
    },
    {
      "epoch": 0.09,
      "learning_rate": 2.056962025316456e-05,
      "loss": 0.7794,
      "step": 75
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.7378336191177368,
      "eval_runtime": 9.9203,
      "eval_samples_per_second": 3.831,
      "eval_steps_per_second": 0.504,
      "step": 80
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.89873417721519e-05,
      "loss": 0.7628,
      "step": 100
    },
    {
      "epoch": 0.12,
      "eval_loss": 0.7243497371673584,
      "eval_runtime": 9.9102,
      "eval_samples_per_second": 3.834,
      "eval_steps_per_second": 0.505,
      "step": 100
    },
    {
      "epoch": 0.14,
      "eval_loss": 0.7043508291244507,
      "eval_runtime": 9.9123,
      "eval_samples_per_second": 3.834,
      "eval_steps_per_second": 0.504,
      "step": 120
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.7405063291139243e-05,
      "loss": 0.7323,
      "step": 125
    },
    {
      "epoch": 0.17,
      "eval_loss": 0.6957876086235046,
      "eval_runtime": 9.9105,
      "eval_samples_per_second": 3.834,
      "eval_steps_per_second": 0.505,
      "step": 140
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.5822784810126583e-05,
      "loss": 0.706,
      "step": 150
    },
    {
      "epoch": 0.19,
      "eval_loss": 0.6854685544967651,
      "eval_runtime": 9.9176,
      "eval_samples_per_second": 3.832,
      "eval_steps_per_second": 0.504,
      "step": 160
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.4240506329113925e-05,
      "loss": 0.6974,
      "step": 175
    },
    {
      "epoch": 0.21,
      "eval_loss": 0.6812862753868103,
      "eval_runtime": 9.918,
      "eval_samples_per_second": 3.831,
      "eval_steps_per_second": 0.504,
      "step": 180
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.2658227848101267e-05,
      "loss": 0.6823,
      "step": 200
    },
    {
      "epoch": 0.24,
      "eval_loss": 0.6784127950668335,
      "eval_runtime": 9.9087,
      "eval_samples_per_second": 3.835,
      "eval_steps_per_second": 0.505,
      "step": 200
    },
    {
      "epoch": 0.26,
      "eval_loss": 0.6723252534866333,
      "eval_runtime": 9.9104,
      "eval_samples_per_second": 3.834,
      "eval_steps_per_second": 0.505,
      "step": 220
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.1075949367088608e-05,
      "loss": 0.6856,
      "step": 225
    },
    {
      "epoch": 0.29,
      "eval_loss": 0.669580340385437,
      "eval_runtime": 9.9088,
      "eval_samples_per_second": 3.835,
      "eval_steps_per_second": 0.505,
      "step": 240
    },
    {
      "epoch": 0.3,
      "learning_rate": 9.49367088607595e-06,
      "loss": 0.6857,
      "step": 250
    },
    {
      "epoch": 0.31,
      "eval_loss": 0.6672892570495605,
      "eval_runtime": 9.8865,
      "eval_samples_per_second": 3.844,
      "eval_steps_per_second": 0.506,
      "step": 260
    },
    {
      "epoch": 0.33,
      "learning_rate": 7.911392405063292e-06,
      "loss": 0.6915,
      "step": 275
    },
    {
      "epoch": 0.33,
      "eval_loss": 0.6666972637176514,
      "eval_runtime": 9.9103,
      "eval_samples_per_second": 3.834,
      "eval_steps_per_second": 0.505,
      "step": 280
    },
    {
      "epoch": 0.36,
      "learning_rate": 6.329113924050633e-06,
      "loss": 0.6685,
      "step": 300
    },
    {
      "epoch": 0.36,
      "eval_loss": 0.6638540029525757,
      "eval_runtime": 9.9153,
      "eval_samples_per_second": 3.832,
      "eval_steps_per_second": 0.504,
      "step": 300
    },
    {
      "epoch": 0.38,
      "eval_loss": 0.6611661911010742,
      "eval_runtime": 9.9319,
      "eval_samples_per_second": 3.826,
      "eval_steps_per_second": 0.503,
      "step": 320
    },
    {
      "epoch": 0.39,
      "learning_rate": 4.746835443037975e-06,
      "loss": 0.6759,
      "step": 325
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.658918559551239,
      "eval_runtime": 9.9278,
      "eval_samples_per_second": 3.828,
      "eval_steps_per_second": 0.504,
      "step": 340
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.1645569620253167e-06,
      "loss": 0.6682,
      "step": 350
    },
    {
      "epoch": 0.43,
      "eval_loss": 0.6601454615592957,
      "eval_runtime": 9.9146,
      "eval_samples_per_second": 3.833,
      "eval_steps_per_second": 0.504,
      "step": 360
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.5822784810126583e-06,
      "loss": 0.6876,
      "step": 375
    },
    {
      "epoch": 0.45,
      "eval_loss": 0.6586702466011047,
      "eval_runtime": 9.9,
      "eval_samples_per_second": 3.838,
      "eval_steps_per_second": 0.505,
      "step": 380
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0,
      "loss": 0.6721,
      "step": 400
    },
    {
      "epoch": 0.48,
      "eval_loss": 0.6582174897193909,
      "eval_runtime": 9.916,
      "eval_samples_per_second": 3.832,
      "eval_steps_per_second": 0.504,
      "step": 400
    }
  ],
  "logging_steps": 25,
  "max_steps": 400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 20,
  "total_flos": 2.51016575975424e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}