config.json

{
    "architectures": [
      "ChameleonForConditionalGeneration"
    ],
    "attention_bias": false,
    "attention_dropout": 0.0,
    "bos_token_id": 1,
    "dropout": 0.05,
    "eos_token_id": 2,
    "hidden_act": "silu",
    "hidden_size": 4096,
    "initializer_range": 0.02,
    "intermediate_size": 11008,
    "mask_image_logits": false,
    "max_position_embeddings": 4096,
    "mlp_bias": false,
    "model_parallel_size": 1,
    "model_type": "chameleon",
    "num_attention_heads": 32,
    "num_hidden_layers": 1,
    "num_key_value_heads": 32,
    "rms_norm_eps": 1e-05,
    "rope_scaling": null,
    "rope_theta": 10000.0,
    "swin_norm": false,
    "tie_word_embeddings": false,
    "torch_dtype": "bfloat16",
    "transformers_version": "4.43.3",
    "use_cache": true,
    "vocab_size": 65536,
    "vq_config": {
        "attn_resolutions": [],
        "ch": 128,
        "ch_mult": [
          1,
          1,
          2,
          2,
          4
        ],
        "model_type": "chameleon_vqgan",
        "n_embed": 8192,
        "out_ch": 3,
        "out_channels": 3
    }
}