{
  "dim": 4096,
  "ffn_dim_multiplier": 1.3,
  "fusion_interval": 4,
  "intermediate_dim": 14336,
  "multiple_of": 1024,
  "n_heads": 32,
  "n_kv_heads": 8,
  "n_layers": 32,
  "n_special_tokens": 8,
  "norm_eps": 1e-05,
  "rope_theta": 500000.0,
  "use_scaled_rope": true,
  "vision_chunk_size": 560,
  "vision_max_num_chunks": 4,
  "vocab_size": 128256,
  "vision_num_cross_attention_layers": 8
}