{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9927360774818403,
  "eval_steps": 500,
  "global_step": 309,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09685230024213075,
      "grad_norm": 1.6106698543432953,
      "learning_rate": 5e-06,
      "loss": 0.649,
      "step": 10
    },
    {
      "epoch": 0.1937046004842615,
      "grad_norm": 0.9911008535692477,
      "learning_rate": 5e-06,
      "loss": 0.5864,
      "step": 20
    },
    {
      "epoch": 0.29055690072639223,
      "grad_norm": 0.6386745349148736,
      "learning_rate": 5e-06,
      "loss": 0.5582,
      "step": 30
    },
    {
      "epoch": 0.387409200968523,
      "grad_norm": 1.0205474281151061,
      "learning_rate": 5e-06,
      "loss": 0.5464,
      "step": 40
    },
    {
      "epoch": 0.48426150121065376,
      "grad_norm": 0.741493469056709,
      "learning_rate": 5e-06,
      "loss": 0.5311,
      "step": 50
    },
    {
      "epoch": 0.5811138014527845,
      "grad_norm": 0.7786501028026512,
      "learning_rate": 5e-06,
      "loss": 0.5181,
      "step": 60
    },
    {
      "epoch": 0.6779661016949152,
      "grad_norm": 0.6711455522499474,
      "learning_rate": 5e-06,
      "loss": 0.5158,
      "step": 70
    },
    {
      "epoch": 0.774818401937046,
      "grad_norm": 0.6218873174772125,
      "learning_rate": 5e-06,
      "loss": 0.5126,
      "step": 80
    },
    {
      "epoch": 0.8716707021791767,
      "grad_norm": 0.6652769231147759,
      "learning_rate": 5e-06,
      "loss": 0.5105,
      "step": 90
    },
    {
      "epoch": 0.9685230024213075,
      "grad_norm": 1.4354098566929865,
      "learning_rate": 5e-06,
      "loss": 0.5082,
      "step": 100
    },
    {
      "epoch": 0.9975786924939467,
      "eval_loss": 0.49066221714019775,
      "eval_runtime": 69.4631,
      "eval_samples_per_second": 40.021,
      "eval_steps_per_second": 0.633,
      "step": 103
    },
    {
      "epoch": 1.0653753026634383,
      "grad_norm": 0.8075871605198771,
      "learning_rate": 5e-06,
      "loss": 0.5159,
      "step": 110
    },
    {
      "epoch": 1.162227602905569,
      "grad_norm": 0.7654895903052866,
      "learning_rate": 5e-06,
      "loss": 0.4583,
      "step": 120
    },
    {
      "epoch": 1.2590799031476998,
      "grad_norm": 0.47351004510337863,
      "learning_rate": 5e-06,
      "loss": 0.4586,
      "step": 130
    },
    {
      "epoch": 1.3559322033898304,
      "grad_norm": 0.5062829494154636,
      "learning_rate": 5e-06,
      "loss": 0.4572,
      "step": 140
    },
    {
      "epoch": 1.4527845036319613,
      "grad_norm": 0.6119092771725125,
      "learning_rate": 5e-06,
      "loss": 0.4544,
      "step": 150
    },
    {
      "epoch": 1.549636803874092,
      "grad_norm": 0.6212058614890003,
      "learning_rate": 5e-06,
      "loss": 0.4561,
      "step": 160
    },
    {
      "epoch": 1.6464891041162226,
      "grad_norm": 0.5105359500584984,
      "learning_rate": 5e-06,
      "loss": 0.4518,
      "step": 170
    },
    {
      "epoch": 1.7433414043583535,
      "grad_norm": 0.5867880979483323,
      "learning_rate": 5e-06,
      "loss": 0.4551,
      "step": 180
    },
    {
      "epoch": 1.8401937046004844,
      "grad_norm": 0.4498960324504211,
      "learning_rate": 5e-06,
      "loss": 0.454,
      "step": 190
    },
    {
      "epoch": 1.937046004842615,
      "grad_norm": 0.5182866069406472,
      "learning_rate": 5e-06,
      "loss": 0.4499,
      "step": 200
    },
    {
      "epoch": 1.9951573849878934,
      "eval_loss": 0.47824251651763916,
      "eval_runtime": 71.5938,
      "eval_samples_per_second": 38.83,
      "eval_steps_per_second": 0.615,
      "step": 206
    },
    {
      "epoch": 2.0338983050847457,
      "grad_norm": 0.9414090883543634,
      "learning_rate": 5e-06,
      "loss": 0.4671,
      "step": 210
    },
    {
      "epoch": 2.1307506053268765,
      "grad_norm": 0.5171048417889069,
      "learning_rate": 5e-06,
      "loss": 0.4066,
      "step": 220
    },
    {
      "epoch": 2.2276029055690074,
      "grad_norm": 0.5123629438372025,
      "learning_rate": 5e-06,
      "loss": 0.4113,
      "step": 230
    },
    {
      "epoch": 2.324455205811138,
      "grad_norm": 0.5363285052863767,
      "learning_rate": 5e-06,
      "loss": 0.4081,
      "step": 240
    },
    {
      "epoch": 2.4213075060532687,
      "grad_norm": 0.4907788960865576,
      "learning_rate": 5e-06,
      "loss": 0.407,
      "step": 250
    },
    {
      "epoch": 2.5181598062953996,
      "grad_norm": 0.507228977380475,
      "learning_rate": 5e-06,
      "loss": 0.4051,
      "step": 260
    },
    {
      "epoch": 2.61501210653753,
      "grad_norm": 0.4923140802099653,
      "learning_rate": 5e-06,
      "loss": 0.4109,
      "step": 270
    },
    {
      "epoch": 2.711864406779661,
      "grad_norm": 0.5763086112386324,
      "learning_rate": 5e-06,
      "loss": 0.3986,
      "step": 280
    },
    {
      "epoch": 2.8087167070217918,
      "grad_norm": 0.4788239568139877,
      "learning_rate": 5e-06,
      "loss": 0.4115,
      "step": 290
    },
    {
      "epoch": 2.9055690072639226,
      "grad_norm": 0.5281993404834231,
      "learning_rate": 5e-06,
      "loss": 0.4124,
      "step": 300
    },
    {
      "epoch": 2.9927360774818403,
      "eval_loss": 0.47956007719039917,
      "eval_runtime": 68.3404,
      "eval_samples_per_second": 40.679,
      "eval_steps_per_second": 0.644,
      "step": 309
    },
    {
      "epoch": 2.9927360774818403,
      "step": 309,
      "total_flos": 517377129185280.0,
      "train_loss": 0.4712127952513957,
      "train_runtime": 10324.3717,
      "train_samples_per_second": 15.347,
      "train_steps_per_second": 0.03
    }
  ],
  "logging_steps": 10,
  "max_steps": 309,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 517377129185280.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}