{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.009844336430197501,
  "eval_steps": 5,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0004922168215098751,
      "grad_norm": 0.9545145630836487,
      "learning_rate": 1e-05,
      "loss": 0.7725,
      "step": 1
    },
    {
      "epoch": 0.0004922168215098751,
      "eval_loss": 0.8794116973876953,
      "eval_runtime": 188.7707,
      "eval_samples_per_second": 18.128,
      "eval_steps_per_second": 9.064,
      "step": 1
    },
    {
      "epoch": 0.0009844336430197501,
      "grad_norm": 0.8624743223190308,
      "learning_rate": 2e-05,
      "loss": 0.7692,
      "step": 2
    },
    {
      "epoch": 0.0014766504645296252,
      "grad_norm": 0.7749966382980347,
      "learning_rate": 3e-05,
      "loss": 0.7575,
      "step": 3
    },
    {
      "epoch": 0.0019688672860395003,
      "grad_norm": 0.7187788486480713,
      "learning_rate": 4e-05,
      "loss": 0.6767,
      "step": 4
    },
    {
      "epoch": 0.0024610841075493753,
      "grad_norm": 0.7192044854164124,
      "learning_rate": 5e-05,
      "loss": 0.6997,
      "step": 5
    },
    {
      "epoch": 0.0024610841075493753,
      "eval_loss": 0.7504404783248901,
      "eval_runtime": 102.4307,
      "eval_samples_per_second": 33.408,
      "eval_steps_per_second": 16.704,
      "step": 5
    },
    {
      "epoch": 0.0029533009290592504,
      "grad_norm": 0.7936996817588806,
      "learning_rate": 6e-05,
      "loss": 0.7108,
      "step": 6
    },
    {
      "epoch": 0.003445517750569126,
      "grad_norm": 0.7949678301811218,
      "learning_rate": 7e-05,
      "loss": 0.6656,
      "step": 7
    },
    {
      "epoch": 0.0039377345720790005,
      "grad_norm": 0.6918225288391113,
      "learning_rate": 8e-05,
      "loss": 0.4623,
      "step": 8
    },
    {
      "epoch": 0.004429951393588876,
      "grad_norm": 0.9470645189285278,
      "learning_rate": 9e-05,
      "loss": 0.4808,
      "step": 9
    },
    {
      "epoch": 0.004922168215098751,
      "grad_norm": 1.2205185890197754,
      "learning_rate": 0.0001,
      "loss": 0.4456,
      "step": 10
    },
    {
      "epoch": 0.004922168215098751,
      "eval_loss": 0.28611862659454346,
      "eval_runtime": 102.5815,
      "eval_samples_per_second": 33.359,
      "eval_steps_per_second": 16.679,
      "step": 10
    },
    {
      "epoch": 0.005414385036608626,
      "grad_norm": 0.9904372692108154,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.2681,
      "step": 11
    },
    {
      "epoch": 0.005906601858118501,
      "grad_norm": 1.1141595840454102,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.1601,
      "step": 12
    },
    {
      "epoch": 0.006398818679628377,
      "grad_norm": 0.5964771509170532,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.097,
      "step": 13
    },
    {
      "epoch": 0.006891035501138252,
      "grad_norm": 0.49781107902526855,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.0517,
      "step": 14
    },
    {
      "epoch": 0.007383252322648127,
      "grad_norm": 0.4129631221294403,
      "learning_rate": 5e-05,
      "loss": 0.0329,
      "step": 15
    },
    {
      "epoch": 0.007383252322648127,
      "eval_loss": 0.018943211063742638,
      "eval_runtime": 102.3977,
      "eval_samples_per_second": 33.419,
      "eval_steps_per_second": 16.709,
      "step": 15
    },
    {
      "epoch": 0.007875469144158001,
      "grad_norm": 0.32676512002944946,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.0197,
      "step": 16
    },
    {
      "epoch": 0.008367685965667877,
      "grad_norm": 0.2452167570590973,
      "learning_rate": 2.061073738537635e-05,
      "loss": 0.0112,
      "step": 17
    },
    {
      "epoch": 0.008859902787177751,
      "grad_norm": 0.17988842725753784,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.0089,
      "step": 18
    },
    {
      "epoch": 0.009352119608687627,
      "grad_norm": 0.2163618505001068,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 0.0091,
      "step": 19
    },
    {
      "epoch": 0.009844336430197501,
      "grad_norm": 0.17688779532909393,
      "learning_rate": 0.0,
      "loss": 0.0077,
      "step": 20
    },
    {
      "epoch": 0.009844336430197501,
      "eval_loss": 0.008986509405076504,
      "eval_runtime": 102.424,
      "eval_samples_per_second": 33.41,
      "eval_steps_per_second": 16.705,
      "step": 20
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.844940337525555e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}