trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.1730205278592376,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 2.6338,
      "step": 5
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 1.8517,
      "step": 10
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002,
      "loss": 1.4181,
      "step": 15
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002,
      "loss": 0.9925,
      "step": 20
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 1.6459,
      "step": 25
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002,
      "loss": 0.8549,
      "step": 30
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002,
      "loss": 0.6065,
      "step": 35
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002,
      "loss": 0.5377,
      "step": 40
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002,
      "loss": 1.3234,
      "step": 45
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0002,
      "loss": 0.7049,
      "step": 50
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0002,
      "loss": 0.5435,
      "step": 55
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0002,
      "loss": 0.437,
      "step": 60
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0002,
      "loss": 0.9905,
      "step": 65
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0002,
      "loss": 1.0981,
      "step": 70
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0002,
      "loss": 0.5207,
      "step": 75
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0002,
      "loss": 0.4523,
      "step": 80
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0002,
      "loss": 0.4863,
      "step": 85
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0002,
      "loss": 1.5822,
      "step": 90
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0002,
      "loss": 0.5247,
      "step": 95
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 0.4884,
      "step": 100
    }
  ],
  "logging_steps": 5,
  "max_steps": 170,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 100,
  "total_flos": 1.5384183170236416e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}