config.json

{
  "action_space": null,
  "policy_kwargs": {
    "active_reward_monitors": {
      "env_reward": 1
    },
    "attention_heads": 24,
    "attention_mask_style": "clipped_causal",
    "attention_memory_size": 256,
    "diff_mlp_embedding": false,
    "hidsize": 3072,
    "img_shape": [
      128,
      128,
      3
    ],
    "impala_chans": [
      16,
      32,
      32
    ],
    "impala_kwargs": {
      "post_pool_groups": 1
    },
    "impala_width": 12,
    "init_norm_kwargs": {
      "batch_norm": false,
      "group_norm_groups": 1
    },
    "n_recurrence_layers": 4,
    "only_img_input": true,
    "pointwise_ratio": 4,
    "pointwise_use_activation": false,
    "recurrence_is_residual": true,
    "recurrence_type": "transformer",
    "timesteps": 128,
    "use_pointwise_layer": true,
    "use_pre_lstm_ln": false
  },
  "temperature": 3
}