{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.792079207920792,
  "eval_steps": 500,
  "global_step": 36,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.39603960396039606,
      "grad_norm": 10.537583351135254,
      "learning_rate": 4.765769467591625e-05,
      "loss": 10.5953,
      "num_input_tokens_seen": 35216,
      "step": 5
    },
    {
      "epoch": 0.7920792079207921,
      "grad_norm": 3.1613588333129883,
      "learning_rate": 4.1069690242163484e-05,
      "loss": 2.1097,
      "num_input_tokens_seen": 71704,
      "step": 10
    },
    {
      "epoch": 1.1584158415841583,
      "grad_norm": 2.1489789485931396,
      "learning_rate": 3.147047612756302e-05,
      "loss": 0.4752,
      "num_input_tokens_seen": 103424,
      "step": 15
    },
    {
      "epoch": 1.5544554455445545,
      "grad_norm": 3.8907082080841064,
      "learning_rate": 2.0658795558326743e-05,
      "loss": 0.4311,
      "num_input_tokens_seen": 139184,
      "step": 20
    },
    {
      "epoch": 1.9504950495049505,
      "grad_norm": 4.895329475402832,
      "learning_rate": 1.0660589091223855e-05,
      "loss": 0.4116,
      "num_input_tokens_seen": 175496,
      "step": 25
    },
    {
      "epoch": 2.3168316831683167,
      "grad_norm": 1.5588610172271729,
      "learning_rate": 3.3493649053890326e-06,
      "loss": 0.333,
      "num_input_tokens_seen": 207928,
      "step": 30
    },
    {
      "epoch": 2.7128712871287126,
      "grad_norm": 1.37593412399292,
      "learning_rate": 9.513254770636137e-08,
      "loss": 0.3408,
      "num_input_tokens_seen": 243800,
      "step": 35
    },
    {
      "epoch": 2.792079207920792,
      "num_input_tokens_seen": 250896,
      "step": 36,
      "total_flos": 1.049399522993111e+17,
      "train_loss": 2.050191469490528,
      "train_runtime": 422.5921,
      "train_samples_per_second": 0.717,
      "train_steps_per_second": 0.085
    }
  ],
  "logging_steps": 5,
  "max_steps": 36,
  "num_input_tokens_seen": 250896,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.049399522993111e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}