trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9846153846153847,
  "eval_steps": 500,
  "global_step": 48,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 0.0,
      "loss": 1.905,
      "step": 2
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0,
      "loss": 2.1079,
      "step": 4
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0,
      "loss": 2.3465,
      "step": 6
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0,
      "loss": 2.3622,
      "step": 8
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0,
      "loss": 2.448,
      "step": 10
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0,
      "loss": 2.6364,
      "step": 12
    },
    {
      "epoch": 0.29,
      "learning_rate": 2e-05,
      "loss": 1.9455,
      "step": 14
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.9906859460363307e-05,
      "loss": 2.179,
      "step": 16
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.9629172873477995e-05,
      "loss": 2.2703,
      "step": 18
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.917211301505453e-05,
      "loss": 2.2115,
      "step": 20
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.8544194045464888e-05,
      "loss": 2.3446,
      "step": 22
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.77571129070442e-05,
      "loss": 2.4233,
      "step": 24
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6825531432186545e-05,
      "loss": 1.9121,
      "step": 26
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.5766803221148676e-05,
      "loss": 2.2741,
      "step": 28
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.4600650377311523e-05,
      "loss": 2.1354,
      "step": 30
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.3348796121709862e-05,
      "loss": 2.1284,
      "step": 32
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.2034560130526341e-05,
      "loss": 2.0004,
      "step": 34
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.0682424133646712e-05,
      "loss": 2.1918,
      "step": 36
    },
    {
      "epoch": 0.78,
      "learning_rate": 9.317575866353293e-06,
      "loss": 1.9007,
      "step": 38
    },
    {
      "epoch": 0.82,
      "learning_rate": 7.965439869473664e-06,
      "loss": 2.1758,
      "step": 40
    },
    {
      "epoch": 0.86,
      "learning_rate": 6.651203878290139e-06,
      "loss": 2.2133,
      "step": 42
    },
    {
      "epoch": 0.9,
      "learning_rate": 5.399349622688479e-06,
      "loss": 2.1743,
      "step": 44
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.2331967788513295e-06,
      "loss": 2.2372,
      "step": 46
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.174468567813461e-06,
      "loss": 2.4227,
      "step": 48
    }
  ],
  "logging_steps": 2,
  "max_steps": 48,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 4314823005533184.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}