Update tiny models for InternVLModel

.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 +1,6 @@
+{% for message in messages %}{{'<|im_start|>' + message['role'] + '
+'}}{% if message['content'] is string %}{{ message['content'] }}{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' %}{{ '<IMG_CONTEXT>
+' }}{% elif content['type'] == 'video' %}{{ '<video>
+' }}{% elif content['type'] == 'text' %}{{ content['text'] }}{% endif %}{% endfor %}{% endif %}{{'<|im_end|>
+'}}{% endfor %}{% if add_generation_prompt %}{{'<|im_start|>assistant
+' }}{% endif %}

config.json

@@ -0,0 +1,76 @@
+{
+  "architectures": [
+    "InternVLModel"
+  ],
+  "downsample_ratio": 0.5,
+  "dtype": "float32",
+  "image_seq_length": 64,
+  "image_token_id": 1,
+  "model_type": "internvl",
+  "projector_hidden_act": "gelu",
+  "text_config": {
+    "attention_dropout": 0.0,
+    "bos_token_id": 3,
+    "eos_token_id": 151645,
+    "hidden_act": "silu",
+    "hidden_size": 128,
+    "initializer_range": 0.02,
+    "intermediate_size": 37,
+    "layer_types": [
+      "full_attention",
+      "full_attention"
+    ],
+    "max_position_embeddings": 512,
+    "max_window_layers": 28,
+    "mlp_ratio": 4,
+    "model_type": "qwen2",
+    "num_attention_heads": 4,
+    "num_hidden_layers": 2,
+    "num_key_value_heads": 2,
+    "output_channels": 64,
+    "pad_token_id": 151643,
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "rope_theta": 10000,
+      "rope_type": "default"
+    },
+    "sliding_window": null,
+    "tie_word_embeddings": true,
+    "use_cache": true,
+    "use_sliding_window": false,
+    "vocab_size": 151675
+  },
+  "tie_word_embeddings": true,
+  "transformers_version": "5.16.0.dev0",
+  "vision_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "hidden_act": "quick_gelu",
+    "hidden_dropout_prob": 0.0,
+    "hidden_size": 32,
+    "image_size": [
+      64,
+      64
+    ],
+    "initializer_range": 0.02,
+    "intermediate_size": 128,
+    "layer_norm_eps": 1e-06,
+    "layer_scale_init_value": 0.1,
+    "model_type": "internvl_vision",
+    "norm_type": "layer_norm",
+    "num_attention_heads": 4,
+    "num_channels": 3,
+    "num_hidden_layers": 2,
+    "patch_size": [
+      4,
+      4
+    ],
+    "projection_dropout": 0.0,
+    "use_absolute_position_embeddings": true,
+    "use_mask_token": false,
+    "use_mean_pooling": true,
+    "use_qk_norm": false
+  },
+  "vision_feature_layer": -1,
+  "vision_feature_select_strategy": "default"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5719424494ff4b3a5fffa28a49a557e99b9488bc61b936b03ac0045bdef98c60
+size 78450664

preprocessor_config.json

@@ -0,0 +1,42 @@
+{
+  "crop_size": {
+    "height": [
+      64,
+      64
+    ],
+    "width": [
+      64,
+      64
+    ]
+  },
+  "crop_to_patches": false,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "GotOcr2ImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "max_patches": 12,
+  "min_patches": 1,
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": [
+      64,
+      64
+    ],
+    "width": [
+      64,
+      64
+    ]
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd0c8be238c81a5f0450067d9565e16794aef2699df631fcf3dd0c2ac4842b06
+size 11423728

tokenizer_config.json

@@ -0,0 +1,26 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "context_image_token": "<IMG_CONTEXT>",
+  "end_image_token": "</img>",
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "is_local": true,
+  "local_files_only": false,
+  "model_max_length": 8192,
+  "model_specific_special_tokens": {
+    "context_image_token": "<IMG_CONTEXT>",
+    "end_image_token": "</img>",
+    "start_image_token": "<img>",
+    "video_token": "<video>"
+  },
+  "pad_token": "<|endoftext|>",
+  "return_token_type_ids": false,
+  "split_special_tokens": false,
+  "start_image_token": "<img>",
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null,
+  "video_token": "<video>"
+}

video_preprocessor_config.json

@@ -0,0 +1,42 @@
+{
+  "crop_size": {
+    "height": [
+      64,
+      64
+    ],
+    "width": [
+      64,
+      64
+    ]
+  },
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "do_sample_frames": false,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "initial_shift": true,
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "return_metadata": false,
+  "size": {
+    "height": [
+      64,
+      64
+    ],
+    "width": [
+      64,
+      64
+    ]
+  },
+  "video_processor_type": "InternVLVideoProcessor"
+}