{ "architectures": [ "ChameleonForConditionalGeneration" ], "attention_bias": false, "attention_dropout": 0.0, "attention_probs_dropout_prob": 0.1, "bos_token_id": 1, "dtype": "float32", "eos_token_id": 2, "hidden_act": "gelu", "hidden_dropout_prob": 0.1, "hidden_size": 32, "image_token_id": 4, "initializer_range": 0.02, "intermediate_size": 37, "is_decoder": false, "max_position_embeddings": 512, "mlp_bias": false, "model_parallel_size": 1, "model_type": "chameleon", "num_attention_heads": 2, "num_hidden_layers": 2, "num_key_value_heads": 2, "pad_token_id": 0, "rms_norm_eps": 1e-05, "rope_parameters": { "rope_theta": 10000.0, "rope_type": "default" }, "swin_norm": false, "tie_word_embeddings": false, "transformers_version": "5.15.0.dev0", "type_vocab_size": 16, "use_cache": true, "vocab_size": 1024, "vocabulary_map": { "\u0000": 0, "\u0001": 1, "\u0002": 2, "\u0003": 3, "\u0005": 5, "\u0006": 6, "\u0007": 7, "\b": 8, "\t": 9, "\u000f": 15, "\u0010": 16, "\u0011": 17, "\u0012": 18, "\u0013": 19, "\u0014": 20, "\u0015": 21, "\u0016": 22, "\u0017": 23, "\u0018": 24, "\u0019": 25, "\u001a": 26, "\u001b": 27, "\u001c": 28, "\u001d": 29, "\u001e": 30, "\u001f": 31, " ": 32, "!": 33, "\"": 34, "#": 35, "$": 36, "%": 37, "&": 38, "'": 39, "(": 40, ")": 41, "*": 42, "+": 43, ",": 44, "-": 45, ".": 46, "/": 47, "0": 48, "1": 49, "2": 50, "3": 51, "4": 52, "5": 53, "6": 54, "7": 55, "8": 56, "9": 57, ":": 58, ";": 59, "<": 60, "": 4, "=": 61, ">": 62, "?": 63, "@": 64, "A": 65, "B": 66, "C": 67, "D": 68, "E": 69, "F": 70, "G": 71, "H": 72, "I": 73, "IMGIMGBAZ": 10, "IMGIMGBBZ": 11, "IMGIMGBCZ": 12, "IMGIMGBDZ": 13, "IMGIMGBEZ": 14, "J": 74, "K": 75, "L": 76, "M": 77, "N": 78, "O": 79, "P": 80, "Q": 81, "R": 82, "S": 83, "T": 84, "U": 85, "V": 86, "W": 87, "X": 88, "Y": 89, "Z": 90, "[": 91, "\\": 92, "]": 93, "^": 94, "_": 95, "`": 96, "a": 97, "b": 98 }, "vq_config": { "attn_resolutions": null, "attn_type": "vanilla", "base_channels": 32, "channel_multiplier": [ 1, 2 ], "double_latent": false, "dropout": 0.0, "embed_dim": 5, "in_channels": 3, "latent_channels": 5, "model_type": "chameleon_vqgan", "num_embeddings": 5, "num_res_blocks": 2, "resolution": 512 } }