Update tiny models for AriaForConditionalGeneration

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

chat_template.jinja

@@ -0,0 +1,4 @@
+{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}<|im_start|>{{ message['role'] }}
+{% if message['content'] is string %}{{ message['content'] }}{% elif message['content'] is iterable %}{% for item in message['content'] %}{% if item['type'] == 'text' %}{{ item['text'] }}{% elif item['type'] == 'image' %}<fim_prefix><|img|><fim_suffix>{% endif %}{% endfor %}{% endif %}<|im_end|>
+{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant
+{% endif %}

config.json

@@ -0,0 +1,89 @@
+{
+  "architectures": [
+    "AriaForConditionalGeneration"
+  ],
+  "dtype": "float32",
+  "eos_token_id": 2,
+  "ignore_index": -100,
+  "image_token_index": 9,
+  "initializer_range": 0.02,
+  "max_value_projector_patch_to_query_dict": 4,
+  "model_type": "aria",
+  "projector_hidden_act": "gelu",
+  "projector_patch_to_query_dict": {
+    "4": 4
+  },
+  "text_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "attention_probs_dropout_prob": 0.1,
+    "bos_token_id": 1,
+    "eos_token_id": 2,
+    "head_dim": 4,
+    "hidden_act": "gelu",
+    "hidden_dropout_prob": 0.1,
+    "hidden_size": 32,
+    "id2label": {
+      "0": "LABEL_0",
+      "1": "LABEL_1",
+      "2": "LABEL_2"
+    },
+    "initializer_range": 0.02,
+    "intermediate_size": 16,
+    "is_training": true,
+    "label2id": {
+      "LABEL_0": 0,
+      "LABEL_1": 1,
+      "LABEL_2": 2
+    },
+    "max_position_embeddings": 200,
+    "mlp_bias": false,
+    "model_type": "aria_text",
+    "moe_intermediate_size": 4,
+    "moe_num_experts": 3,
+    "moe_num_shared_experts": 2,
+    "moe_topk": 2,
+    "num_attention_heads": 2,
+    "num_choices": 4,
+    "num_experts_per_tok": 3,
+    "num_hidden_layers": 2,
+    "num_key_value_heads": 2,
+    "pad_token_id": 2,
+    "pretraining_tp": 1,
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "rope_theta": 5000000,
+      "rope_type": "default"
+    },
+    "seq_length": 7,
+    "tie_word_embeddings": false,
+    "type_sequence_label_size": 2,
+    "type_vocab_size": 16,
+    "use_cache": true,
+    "use_input_mask": true,
+    "use_labels": true,
+    "use_token_type_ids": false,
+    "vocab_size": 100353
+  },
+  "tie_word_embeddings": false,
+  "transformers_version": "5.16.0.dev0",
+  "vision_config": {
+    "attention_dropout": 0.1,
+    "dropout": 0.1,
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 32,
+    "image_size": 16,
+    "initializer_range": 0.02,
+    "intermediate_size": 4,
+    "is_training": true,
+    "layer_norm_eps": 1e-06,
+    "model_type": "idefics3_vision",
+    "num_attention_heads": 2,
+    "num_channels": 3,
+    "num_hidden_layers": 2,
+    "patch_size": 8,
+    "projection_dim": 4
+  },
+  "vision_feature_layer": -1,
+  "vision_feature_select_strategy": "default"
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "output_attentions": false,
+  "output_hidden_states": false,
+  "pad_token_id": 2,
+  "transformers_version": "5.16.0.dev0",
+  "use_cache": true
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6bc5a9b1f10f58ca5a823b7c926b81736fc32c6c22462e7423fd9c5f4e04652
+size 25876664

preprocessor_config.json

@@ -0,0 +1,107 @@
+{
+  "crop_size": {
+    "height": 16,
+    "width": 16
+  },
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "AriaImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "max_image_size": 980,
+  "min_image_size": 336,
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 16,
+    "width": 16
+  },
+  "split_image": false,
+  "split_resolutions": [
+    [
+      490,
+      980
+    ],
+    [
+      490,
+      1470
+    ],
+    [
+      490,
+      1960
+    ],
+    [
+      490,
+      2450
+    ],
+    [
+      490,
+      2940
+    ],
+    [
+      490,
+      3430
+    ],
+    [
+      490,
+      3920
+    ],
+    [
+      980,
+      1960
+    ],
+    [
+      980,
+      1470
+    ],
+    [
+      980,
+      980
+    ],
+    [
+      980,
+      490
+    ],
+    [
+      1470,
+      490
+    ],
+    [
+      1470,
+      980
+    ],
+    [
+      1960,
+      490
+    ],
+    [
+      1960,
+      980
+    ],
+    [
+      2450,
+      490
+    ],
+    [
+      2940,
+      490
+    ],
+    [
+      3430,
+      490
+    ],
+    [
+      3920,
+      490
+    ]
+  ]
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02702cce6c4de786d52a3dc624b39e86134c159b7490ea30630739c6f723e7f8
+size 11091481

tokenizer_config.json

@@ -0,0 +1,22 @@
+{
+  "add_prefix_space": true,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": null,
+  "image_token": "<|img|>",
+  "is_local": true,
+  "legacy": true,
+  "local_files_only": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "model_specific_special_tokens": {
+    "image_token": "<|img|>"
+  },
+  "pad_token": "<|endoftext|>",
+  "processor_class": "AriaProcessor",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}