trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": null,
  "eval_steps": 500,
  "global_step": 28,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 1e-05,
      "loss": 0.0299,
      "reward": -5.125,
      "step": 0
    },
    {
      "epoch": 0.07,
      "learning_rate": 1e-05,
      "loss": 0.0401,
      "reward": -6.8318,
      "step": 1
    },
    {
      "epoch": 0.14,
      "learning_rate": 1e-05,
      "loss": 0.029,
      "reward": -5.9028,
      "step": 2
    },
    {
      "epoch": 0.21,
      "learning_rate": 1e-05,
      "loss": 0.0381,
      "reward": -6.75,
      "step": 3
    },
    {
      "epoch": 0.29,
      "learning_rate": 1e-05,
      "loss": 0.0364,
      "reward": -5.7639,
      "step": 4
    },
    {
      "epoch": 0.36,
      "learning_rate": 1e-05,
      "loss": 0.0378,
      "reward": -6.25,
      "step": 5
    },
    {
      "epoch": 0.43,
      "learning_rate": 1e-05,
      "loss": 0.0352,
      "reward": -6.6071,
      "step": 6
    },
    {
      "epoch": 0.5,
      "learning_rate": 1e-05,
      "loss": 0.0275,
      "reward": -3.9205,
      "step": 7
    },
    {
      "epoch": 0.57,
      "learning_rate": 1e-05,
      "loss": 0.0313,
      "reward": -7.0625,
      "step": 8
    },
    {
      "epoch": 0.64,
      "learning_rate": 1e-05,
      "loss": 0.0312,
      "reward": -5.8546,
      "step": 9
    },
    {
      "epoch": 0.71,
      "learning_rate": 1e-05,
      "loss": 0.0261,
      "reward": -1.9861,
      "step": 10
    },
    {
      "epoch": 0.79,
      "learning_rate": 1e-05,
      "loss": 0.0349,
      "reward": -7.5,
      "step": 11
    },
    {
      "epoch": 0.86,
      "learning_rate": 1e-05,
      "loss": 0.0346,
      "reward": -6.3542,
      "step": 12
    },
    {
      "epoch": 0.93,
      "learning_rate": 1e-05,
      "loss": 0.0346,
      "reward": -4.8214,
      "step": 13
    },
    {
      "epoch": 1.0,
      "learning_rate": 1e-05,
      "loss": 0.0299,
      "reward": -6.0794,
      "step": 14
    },
    {
      "epoch": 1.07,
      "learning_rate": 1e-05,
      "loss": 0.0365,
      "reward": -8.75,
      "step": 15
    },
    {
      "epoch": 1.14,
      "learning_rate": 1e-05,
      "loss": 0.0273,
      "reward": -6.6927,
      "step": 16
    },
    {
      "epoch": 1.21,
      "learning_rate": 1e-05,
      "loss": 0.0266,
      "reward": -2.7678,
      "step": 17
    },
    {
      "epoch": 1.29,
      "learning_rate": 1e-05,
      "loss": 0.0211,
      "reward": -3.5,
      "step": 18
    },
    {
      "epoch": 1.36,
      "learning_rate": 1e-05,
      "loss": 0.0271,
      "reward": -5.3509,
      "step": 19
    },
    {
      "epoch": 1.43,
      "learning_rate": 1e-05,
      "loss": 0.025,
      "reward": -1.309,
      "step": 20
    },
    {
      "epoch": 1.5,
      "learning_rate": 1e-05,
      "loss": 0.0215,
      "reward": -4.0697,
      "step": 21
    },
    {
      "epoch": 1.57,
      "learning_rate": 1e-05,
      "loss": 0.018,
      "reward": -5.4167,
      "step": 22
    },
    {
      "epoch": 1.64,
      "learning_rate": 1e-05,
      "loss": 0.0293,
      "reward": -4.5805,
      "step": 23
    },
    {
      "epoch": 1.71,
      "learning_rate": 1e-05,
      "loss": 0.0249,
      "reward": -5.7969,
      "step": 24
    },
    {
      "epoch": 1.79,
      "learning_rate": 1e-05,
      "loss": 0.0167,
      "reward": -4.0278,
      "step": 25
    },
    {
      "epoch": 1.86,
      "learning_rate": 1e-05,
      "loss": 0.026,
      "reward": -4.25,
      "step": 26
    },
    {
      "epoch": 1.93,
      "learning_rate": 1e-05,
      "loss": 0.0181,
      "reward": -3.6364,
      "step": 27
    }
  ],
  "logging_steps": 500,
  "max_steps": 28,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2.0,
  "save_steps": 500,
  "total_flos": 0,
  "train_batch_size": null,
  "trial_name": null,
  "trial_params": null
}