transformer/config.json

{
  "_class_name": "UVit2DModel",
  "_diffusers_version": "0.24.0.dev0",
  "attention_dropout": 0.0,
  "block_num_heads": 6,
  "block_out_channels": 384,
  "codebook_size": 8192,
  "cond_embed_dim": 1024,
  "downsample": true,
  "encoder_hidden_size": 1024,
  "hidden_dropout": 0.0,
  "hidden_size": 768,
  "in_channels": 384,
  "intermediate_size": 1408,
  "layer_norm_eps": 1e-06,
  "ln_elementwise_affine": true,
  "micro_cond_embed_dim": 384,
  "micro_cond_encode_dim": 1280,
  "num_attention_heads": 8,
  "num_hidden_layers": 11,
  "num_res_blocks": 3,
  "sample_size": 32,
  "upsample": true,
  "use_bias": false,
  "vocab_size": 8256
}