{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.734624810362936,
  "eval_steps": 500,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4667989263624694,
      "grad_norm": 0.5918045043945312,
      "learning_rate": 9.895259863857224e-05,
      "loss": 0.3641,
      "step": 500
    },
    {
      "epoch": 0.4667989263624694,
      "eval_loss": 0.15268388390541077,
      "eval_runtime": 16.0975,
      "eval_samples_per_second": 5.405,
      "eval_steps_per_second": 5.405,
      "step": 500
    },
    {
      "epoch": 0.9335978527249388,
      "grad_norm": 0.39225736260414124,
      "learning_rate": 9.369885385026829e-05,
      "loss": 0.1066,
      "step": 1000
    },
    {
      "epoch": 0.9335978527249388,
      "eval_loss": 0.10832103341817856,
      "eval_runtime": 16.0859,
      "eval_samples_per_second": 5.408,
      "eval_steps_per_second": 5.408,
      "step": 1000
    },
    {
      "epoch": 1.4003967790874081,
      "grad_norm": 0.2461080551147461,
      "learning_rate": 8.447873837450339e-05,
      "loss": 0.0794,
      "step": 1500
    },
    {
      "epoch": 1.4003967790874081,
      "eval_loss": 0.09633716195821762,
      "eval_runtime": 16.095,
      "eval_samples_per_second": 5.405,
      "eval_steps_per_second": 5.405,
      "step": 1500
    },
    {
      "epoch": 1.8671957054498773,
      "grad_norm": 0.22391347587108612,
      "learning_rate": 7.212912532203201e-05,
      "loss": 0.0739,
      "step": 2000
    },
    {
      "epoch": 1.8671957054498773,
      "eval_loss": 0.08717021346092224,
      "eval_runtime": 16.0946,
      "eval_samples_per_second": 5.406,
      "eval_steps_per_second": 5.406,
      "step": 2000
    },
    {
      "epoch": 2.3341113315439372,
      "grad_norm": 0.11494116485118866,
      "learning_rate": 5.777093981781615e-05,
      "loss": 0.058,
      "step": 2500
    },
    {
      "epoch": 2.3341113315439372,
      "eval_loss": 0.09098245203495026,
      "eval_runtime": 16.4848,
      "eval_samples_per_second": 5.278,
      "eval_steps_per_second": 5.278,
      "step": 2500
    },
    {
      "epoch": 2.800910257906407,
      "grad_norm": 0.1403874009847641,
      "learning_rate": 4.270741709782294e-05,
      "loss": 0.057,
      "step": 3000
    },
    {
      "epoch": 2.800910257906407,
      "eval_loss": 0.08592888712882996,
      "eval_runtime": 16.3729,
      "eval_samples_per_second": 5.314,
      "eval_steps_per_second": 5.314,
      "step": 3000
    },
    {
      "epoch": 3.267825884000467,
      "grad_norm": 0.27918434143066406,
      "learning_rate": 2.8305813044122097e-05,
      "loss": 0.0432,
      "step": 3500
    },
    {
      "epoch": 3.267825884000467,
      "eval_loss": 0.08947568386793137,
      "eval_runtime": 15.9811,
      "eval_samples_per_second": 5.444,
      "eval_steps_per_second": 5.444,
      "step": 3500
    },
    {
      "epoch": 3.734624810362936,
      "grad_norm": 0.2596851885318756,
      "learning_rate": 1.587330382125886e-05,
      "loss": 0.0452,
      "step": 4000
    },
    {
      "epoch": 3.734624810362936,
      "eval_loss": 0.08896192163228989,
      "eval_runtime": 15.9214,
      "eval_samples_per_second": 5.464,
      "eval_steps_per_second": 5.464,
      "step": 4000
    }
  ],
  "logging_steps": 500,
  "max_steps": 5355,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.818658312972288e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}