files

config.json

@@ -0,0 +1,106 @@
+{
+  "architectures": [
+    "LegatoModel"
+  ],
+  "encoder_pretrained_model_name_or_path": "./vision_encoder",
+  "image_token_index": 4096,
+  "model_type": "legato",
+  "pad_token_id": 4,
+  "text_config": {
+    "bos_token_id": 1,
+    "cross_attention_layers": [
+      3,
+      7,
+      11,
+      15
+    ],
+    "dropout": 0,
+    "eos_token_id": [
+      2
+    ],
+    "hidden_act": "silu",
+    "hidden_size": 768,
+    "initializer_range": 0.02,
+    "intermediate_size": 1526,
+    "max_position_embeddings": 131072,
+    "model_type": "mllama_text_model",
+    "num_attention_heads": 12,
+    "num_hidden_layers": 18,
+    "num_key_value_heads": 6,
+    "pad_token_id": 4,
+    "rms_norm_eps": 1e-05,
+    "rope_scaling": {
+      "factor": 8.0,
+      "high_freq_factor": 4.0,
+      "low_freq_factor": 1.0,
+      "original_max_position_embeddings": 8192,
+      "rope_type": "llama3"
+    },
+    "rope_theta": 500000,
+    "torch_dtype": "float32",
+    "use_cache": true,
+    "vocab_size": 4097
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.50.0",
+  "vision_config": {
+    "_attn_implementation_autoset": true,
+    "_name_or_path": "meta-llama/Llama-3.2-11B-Vision",
+    "attention_heads": 16,
+    "hidden_act": "gelu",
+    "hidden_size": 1280,
+    "image_size": 448,
+    "initializer_range": 0.02,
+    "intermediate_layers_indices": [
+      3,
+      7,
+      15,
+      23,
+      30
+    ],
+    "intermediate_size": 5120,
+    "max_num_tiles": 4,
+    "model_type": "mllama_vision_model",
+    "norm_eps": 1e-05,
+    "num_channels": 3,
+    "num_global_layers": 8,
+    "num_hidden_layers": 32,
+    "patch_size": 14,
+    "supported_aspect_ratios": [
+      [
+        1,
+        1
+      ],
+      [
+        1,
+        2
+      ],
+      [
+        1,
+        3
+      ],
+      [
+        1,
+        4
+      ],
+      [
+        2,
+        1
+      ],
+      [
+        2,
+        2
+      ],
+      [
+        3,
+        1
+      ],
+      [
+        4,
+        1
+      ]
+    ],
+    "torch_dtype": "float32",
+    "vision_output_dim": 7680
+  }
+}

generation_config.json

@@ -0,0 +1,11 @@
+{
+  "bos_token_id": 1,
+  "eos_token_id": [
+    2
+  ],
+  "load_pretrained_encoder": false,
+  "max_length": 2048,
+  "pad_token_id": 4,
+  "repetition_penalty": 1.1,
+  "transformers_version": "4.50.0"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cdeafc9ab30eba74e1c87f0722f869aa9c00d4c4d5986561d4abfeccd6f9cfcc
+size 429478304

preprocessor_config.json

@@ -0,0 +1,26 @@
+{
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_pad": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "LegatoImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "max_image_tiles": 4,
+  "processor_class": "LegatoProcessor",
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 448,
+    "width": 448
+  }
+}

special_tokens_map.json

@@ -0,0 +1,26 @@
+{
+  "additional_special_tokens": [
+    "<|image|>"
+  ],
+  "bos_token": {
+    "content": "<|begin_of_abc|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|end_of_abc|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|pad|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,71 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<|unkown|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<|begin_of_abc|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<|end_of_abc|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<|text|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "4": {
+      "content": "<|pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4096": {
+      "content": "<|image|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|image|>"
+  ],
+  "bos_token": "<|begin_of_abc|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|end_of_abc|>",
+  "extra_special_tokens": {},
+  "max_length": 2048,
+  "model_max_length": 2048,
+  "pad_to_multiple_of": null,
+  "pad_token": "<|pad|>",
+  "pad_token_type_id": 0,
+  "padding_side": "left",
+  "processor_class": "LegatoProcessor",
+  "return_token_type_ids": false,
+  "stride": 0,
+  "tokenizer_class": "PreTrainedTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first"
+}