training_log.json

[
  {
    "step": 200,
    "policy_loss": 6.2601,
    "value_loss": 0.3127,
    "lr": 2.82e-05,
    "gnorm": 9.03
  },
  {
    "step": 400,
    "policy_loss": 4.3384,
    "value_loss": 0.2786,
    "lr": 5.64e-05,
    "gnorm": 6.71
  },
  {
    "step": 600,
    "policy_loss": 3.9855,
    "value_loss": 0.2473,
    "lr": 8.46e-05,
    "gnorm": 4.17
  },
  {
    "step": 800,
    "policy_loss": 3.7476,
    "value_loss": 0.231,
    "lr": 0.000113,
    "gnorm": 2.66
  },
  {
    "step": 1000,
    "policy_loss": 3.5844,
    "value_loss": 0.224,
    "lr": 0.000141,
    "gnorm": 2.37
  },
  {
    "step": 1200,
    "policy_loss": 3.4408,
    "value_loss": 0.2213,
    "lr": 0.000169,
    "gnorm": 1.9
  },
  {
    "step": 1400,
    "policy_loss": 3.3147,
    "value_loss": 0.2153,
    "lr": 0.000197,
    "gnorm": 1.73
  },
  {
    "step": 1600,
    "policy_loss": 3.1981,
    "value_loss": 0.2103,
    "lr": 0.0002,
    "gnorm": 1.48
  },
  {
    "step": 1800,
    "policy_loss": 3.0781,
    "value_loss": 0.2041,
    "lr": 0.0002,
    "gnorm": 1.52
  },
  {
    "step": 2000,
    "policy_loss": 2.9918,
    "value_loss": 0.2015,
    "lr": 0.0002,
    "gnorm": 1.26
  },
  {
    "step": 2200,
    "policy_loss": 2.9199,
    "value_loss": 0.1948,
    "lr": 0.0002,
    "gnorm": 1.24
  },
  {
    "step": 2400,
    "policy_loss": 2.8578,
    "value_loss": 0.1895,
    "lr": 0.0002,
    "gnorm": 1.14
  },
  {
    "step": 2600,
    "policy_loss": 2.8174,
    "value_loss": 0.1862,
    "lr": 0.0002,
    "gnorm": 1.1
  },
  {
    "step": 2800,
    "policy_loss": 2.7771,
    "value_loss": 0.1822,
    "lr": 0.0002,
    "gnorm": 1.24
  },
  {
    "step": 3000,
    "policy_loss": 2.7415,
    "value_loss": 0.1807,
    "lr": 0.000199,
    "gnorm": 1.14
  },
  {
    "step": 3200,
    "policy_loss": 2.6978,
    "value_loss": 0.1727,
    "lr": 0.000199,
    "gnorm": 1.1
  }
]