| { |
| "module": "keras_hub.src.models.smolvlm2.smolvlm2_causal_lm", |
| "class_name": "SmolVLM2CausalLM", |
| "config": { |
| "backbone": { |
| "module": "keras_hub.src.models.smolvlm2.smolvlm2_backbone", |
| "class_name": "SmolVLM2Backbone", |
| "config": { |
| "name": "smol_vlm2_backbone", |
| "trainable": true, |
| "dtype": { |
| "module": "keras", |
| "class_name": "DTypePolicy", |
| "config": { |
| "name": "float32" |
| }, |
| "registered_name": null |
| }, |
| "vocabulary_size": 49280, |
| "image_size": 384, |
| "patch_size": 14, |
| "vision_hidden_dim": 1152, |
| "vision_intermediate_dim": 4304, |
| "vision_num_layers": 27, |
| "vision_num_heads": 16, |
| "hidden_dim": 2048, |
| "intermediate_dim": 8192, |
| "num_layers": 24, |
| "num_query_heads": 32, |
| "num_key_value_heads": 32, |
| "scale_factor": 3, |
| "image_token_id": 49190, |
| "rope_max_wavelength": 130000, |
| "layer_norm_epsilon": 1e-05, |
| "vision_layer_norm_epsilon": 1e-06, |
| "tie_word_embeddings": false |
| }, |
| "registered_name": "keras_hub>SmolVLM2Backbone" |
| }, |
| "preprocessor": { |
| "module": "keras_hub.src.models.smolvlm2.smolvlm2_causal_lm_preprocessor", |
| "class_name": "SmolVLM2CausalLMPreprocessor", |
| "config": { |
| "name": "smol_vlm2_causal_lm_preprocessor", |
| "trainable": true, |
| "dtype": { |
| "module": "keras", |
| "class_name": "DTypePolicy", |
| "config": { |
| "name": "float32" |
| }, |
| "registered_name": null |
| }, |
| "tokenizer": { |
| "module": "keras_hub.src.models.smolvlm2.smolvlm2_tokenizer", |
| "class_name": "SmolVLM2Tokenizer", |
| "config": { |
| "name": "smol_vlm2_tokenizer", |
| "trainable": true, |
| "dtype": { |
| "module": "keras", |
| "class_name": "DTypePolicy", |
| "config": { |
| "name": "int32" |
| }, |
| "registered_name": null |
| }, |
| "config_file": "tokenizer.json", |
| "sequence_length": null, |
| "add_prefix_space": false, |
| "unsplittable_tokens": [ |
| "<row_4_col_1>", |
| "<row_2_col_3>", |
| "<issue_start>", |
| "<jupyter_start>", |
| "<global-img>", |
| "<row_3_col_6>", |
| "<reponame>", |
| "<row_4_col_6>", |
| "<row_3_col_3>", |
| "<row_6_col_4>", |
| "<row_6_col_3>", |
| "<gh_stars>", |
| "<row_5_col_4>", |
| "<row_2_col_6>", |
| "<row_3_col_5>", |
| "<jupyter_text>", |
| "<row_5_col_2>", |
| "<row_5_col_5>", |
| "<row_6_col_6>", |
| "<row_3_col_4>", |
| "<row_5_col_6>", |
| "<|endoftext|>", |
| "<row_4_col_2>", |
| "<fake_token_around_image>", |
| "<|im_end|>", |
| "<row_1_col_2>", |
| "<|im_start|>", |
| "<jupyter_output>", |
| "<row_3_col_2>", |
| "<row_6_col_1>", |
| "<row_2_col_1>", |
| "<row_2_col_5>", |
| "<image>", |
| "<row_5_col_3>", |
| "<row_2_col_4>", |
| "<repo_name>", |
| "<filename>", |
| "<issue_closed>", |
| "<jupyter_script>", |
| "<row_4_col_4>", |
| "<jupyter_code>", |
| "<row_2_col_2>", |
| "<row_6_col_2>", |
| "<row_1_col_5>", |
| "<row_1_col_3>", |
| "<row_1_col_6>", |
| "<row_3_col_1>", |
| "<file_sep>", |
| "<row_4_col_5>", |
| "<row_1_col_4>", |
| "<row_5_col_1>", |
| "<empty_output>", |
| "<issue_comment>", |
| "<row_4_col_3>", |
| "<row_1_col_1>", |
| "<end_of_utterance>", |
| "<row_6_col_5>" |
| ] |
| }, |
| "registered_name": "keras_hub>SmolVLM2Tokenizer" |
| }, |
| "image_converter": { |
| "module": "keras_hub.src.models.smolvlm2.smolvlm2_image_converter", |
| "class_name": "SmolVLM2ImageConverter", |
| "config": { |
| "name": "smol_vlm2_image_converter", |
| "trainable": true, |
| "dtype": { |
| "module": "keras", |
| "class_name": "DTypePolicy", |
| "config": { |
| "name": "float32" |
| }, |
| "registered_name": null |
| }, |
| "image_size": null, |
| "scale": [ |
| 0.00784313725490196, |
| 0.00784313725490196, |
| 0.00784313725490196 |
| ], |
| "offset": [ |
| -1.0, |
| -1.0, |
| -1.0 |
| ], |
| "interpolation": "lanczos3", |
| "antialias": true, |
| "crop_to_aspect_ratio": true, |
| "pad_to_aspect_ratio": false, |
| "bounding_box_format": "yxyx", |
| "max_image_size": 384, |
| "size": 1536, |
| "do_image_splitting": true |
| }, |
| "registered_name": "keras_hub>SmolVLM2ImageConverter" |
| }, |
| "video_converter": { |
| "module": "keras_hub.src.models.smolvlm2.smolvlm2_video_converter", |
| "class_name": "SmolVLM2VideoConverter", |
| "config": { |
| "name": "smol_vlm2_video_converter", |
| "trainable": true, |
| "dtype": { |
| "module": "keras", |
| "class_name": "DTypePolicy", |
| "config": { |
| "name": "float32" |
| }, |
| "registered_name": null |
| }, |
| "image_size": null, |
| "scale": [ |
| 0.00784313725490196, |
| 0.00784313725490196, |
| 0.00784313725490196 |
| ], |
| "offset": [ |
| -1.0, |
| -1.0, |
| -1.0 |
| ], |
| "max_image_size": 384, |
| "size": 1536, |
| "num_frames": 64, |
| "fps": 1, |
| "interpolation": "lanczos3", |
| "antialias": true |
| }, |
| "registered_name": "keras_hub>SmolVLM2VideoConverter" |
| }, |
| "config_file": "preprocessor.json", |
| "sequence_length": 1024, |
| "add_start_token": false, |
| "add_end_token": true, |
| "image_seq_len": 81 |
| }, |
| "registered_name": "keras_hub>SmolVLM2CausalLMPreprocessor" |
| }, |
| "name": "smol_vlm2_causal_lm" |
| }, |
| "registered_name": "keras_hub>SmolVLM2CausalLM" |
| } |
