Duplicate from starvector/starvector-8b-im2svg

Co-authored-by: Juan A. Rodriguez <[email protected]>

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz 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

README.md

@@ -0,0 +1,150 @@
+---
+library_name: transformers
+license: apache-2.0
+language:
+- en
+---
+
+# Model Card for StarVector
+
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/65c27c201b5b51dd4814fcd2/ULL7FkrMHA38I8olD7nEh.png)
+
+StarVector is a foundation model for generating Scalable Vector Graphics (SVG) code from images and text. It utilizes a Vision-Language Modeling architecture to understand both visual and textual inputs, enabling high-quality vectorization and text-guided SVG creation.
+
+## Model Details
+
+### Model Description
+
+This is the model card for the StarVector model, a 🤗 transformers model. StarVector is a foundation model for generating Scalable Vector Graphics (SVG) code from images and text. It utilizes a Vision-Language Modeling architecture to understand both visual and textual inputs, enabling high-quality vectorization and text-guided SVG creation.
+
+- **Developed by:** ServiceNow Research, Mila - Quebec AI Institute, ETS, Montreal.
+- **Shared by :** Juan A Rodriguez, Abhay Puri, Shubham Agarwal, Issam H. Laradji, Sai Rajeswar, Pau Rodriguez, David Vazquez, Christopher Pal, Marco Pedersoli.
+- **Model type:** Vision-Language Model for SVG Generation.
+- **Language(s) (NLP):** English.
+- **License:** Apache 2.0
+
+### Model Architecture
+
+The StarVector architecture integrates an image encoder and a Large Language Model (LLM) Adapter to generate SVG code from both image and text inputs. Images are first converted into embeddings using a Vision Transformer (ViT), after which the LLM Adapter maps these embeddings into the LLM's embedding space to create visual tokens. Text prompts are handled through the LLM’s tokenizer and embedder. This unified multimodal approach ensures precise and contextually rich SVG output.
+
+<figure>
+  <img src="https://cdn-uploads.huggingface.co/production/uploads/65c27c201b5b51dd4814fcd2/IVGxASfNr8wfu-agH9Nqj.png" alt="Figure 2: StarVector Architecture">
+  <figcaption>Figure 2: a) StarVector Architecture: StarVector projects images into embeddings via an image encoder, then maps these embeddings to the LLM hidden space using an LLM Adapter, generating Visual Tokens. Text conditioning is achieved with the LLM's tokenizer and embedder. The model learns to map token sequences (visual or textual) to SVG code. The symbol ⊕ denotes mutually exclusive operations (image-to- SVG or text-to-SVG), while ‖ indicates sequence concatenation. Figure 2: b)Vision Model and Adapter: The image encoder employs a Vision Transformer (ViT) to process image patches sequentially. The LLM Adapter non-linearly projects embeddings into visual tokens for LLM integration.</figcaption>
+</figure>
+
+### Model Sources
+
+- **Repository:** [https://github.com/joanrod/star-vector](https://github.com/joanrod/star-vector)
+- **Paper:** [https://arxiv.org/abs/2312.11556](https://arxiv.org/abs/2312.11556)
+
+## Uses
+
+### Direct Use
+
+Image-to-SVG generation, Text-to-SVG generation.
+
+### Downstream Use
+
+Creation of icons, logotypes, technical diagrams, and other vector graphics.
+
+### Out-of-Scope Use
+
+Generating realistic photographic images or complex 3D graphics.
+
+## Bias, Risks, and Limitations
+
+Potential biases may exist in the model due to the composition of the training data (SVG-Stack). The model's ability to perfectly vectorize all types of images and interpret all textual instructions may have limitations. Users should be aware of these potential issues, especially in critical applications.
+
+### Recommendations
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Further investigation into the model's behavior across different types of inputs is recommended.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+```Python
+from PIL import Image
+from transformers import AutoModelForCausalLM, AutoTokenizer, AutoProcessor
+from starvector.data.util import process_and_rasterize_svg
+import torch
+ 
+model_name = "starvector/starvector-8b-im2svg"
+ 
+starvector = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.float16, trust_remote_code=True)
+processor = starvector.model.processor
+tokenizer = starvector.model.svg_transformer.tokenizer
+ 
+starvector.cuda()
+starvector.eval()
+ 
+image_pil = Image.open('assets/examples/sample-18.png')
+ 
+image = processor(image_pil, return_tensors="pt")['pixel_values'].cuda()
+if not image.shape[0] == 1:
+    image = image.squeeze(0)
+batch = {"image": image}
+ 
+raw_svg = starvector.generate_im2svg(batch, max_length=4000)[0]
+svg, raster_image = process_and_rasterize_svg(raw_svg)
+```
+
+## Training Details
+
+### Training Data
+SVG-Stack: A dataset of over 2 million SVG samples.
+
+### Training Procedure
+The model utilizes a Vision-Language Modeling architecture. Images are projected into embeddings via an image encoder, then mapped to the LLM hidden space using an LLM Adapter, generating Visual Tokens. Text conditioning is achieved with the LLM's tokenizer and embedder. The model learns to map token sequences (visual or textual) to SVG code.
+
+## Evaluation
+### Testing Data & Factors
+
+#### Testing Data
+SVG-Bench
+
+#### Factors
+SVG-Stack, SVG-Fonts, SVG-Icons, SVG-Emoji, SVG-Diagrams.
+
+## Models
+
+StarVector models achieve state-of-the-art performance on SVG generation tasks
+
+We provide [Hugging Face 🤗 model checkpoints](https://huggingface.co/collections/starvector/starvector-models-6783b22c7bd4b43d13cb5289) for image2SVG vectorization, for 💫 StarVector-8B and 💫 StarVector-1B. These are the results on SVG-Bench, using the DinoScore metric.
+
+| Method             | SVG-Stack | SVG-Fonts | SVG-Icons | SVG-Emoji | SVG-Diagrams |
+|--------------------|-----------|-----------|-----------|-----------|--------------|
+| AutoTrace          | 0.942     | 0.954     | 0.946     | 0.975     | 0.874        |
+| Potrace            | 0.898     | 0.967     | 0.972     | 0.882     | 0.875        |
+| VTracer            | 0.954     | 0.964     | 0.940     | 0.981     | 0.882        |
+| Im2Vec             | 0.692     | 0.733     | 0.754     | 0.732     | -            |
+| LIVE               | 0.934     | 0.956     | 0.959     | 0.969     | 0.870        |
+| DiffVG             | 0.810     | 0.821     | 0.952     | 0.814     | 0.822        |
+| GPT-4-V            | 0.852     | 0.842     | 0.848     | 0.850     | -            |
+| 💫 **StarVector-1B** | 0.926     | 0.978     | 0.975     | 0.929     | 0.943        |
+| 💫 **StarVector-8B** | 0.966     | 0.982     | 0.984     | 0.981     | 0.959        |
+
+**Note:** StarVector models will not work for natural images or illustrations, as they have not been trained on those images. They excel in vectorizing icons, logotypes, technical diagrams, graphs, and charts.
+
+As shown in the table above, StarVector-8B achieves the highest performance across all benchmark datasets, demonstrating its effectiveness in generating high-quality SVG code from images. The model's ability to understand and reproduce complex vector graphics makes it particularly valuable for applications requiring precise vectorization of icons, logos, and technical diagrams.
+
+## Summary
+StarVector represents a significant advancement in the field of vector graphics generation. By combining the power of vision-language models with a comprehensive training dataset, we've created a system that can accurately translate images into high-quality SVG code. The model's performance on SVG-Bench demonstrates its effectiveness across a wide range of vector graphics tasks.
+
+We believe that StarVector will enable new applications in design, illustration, and technical documentation, making vector graphics more accessible and easier to create. We invite the research community to build upon our work and explore new directions in this exciting field.
+
+For more details, please refer to our [paper](https://arxiv.org/abs/2312.11556) and explore our [code](https://github.com/joanrod/star-vector) repository.
+
+## BibTeX entry and citation info
+
+```
+@misc{rodriguez2024starvector,
+      title={StarVector: Generating Scalable Vector Graphics Code from Images and Text},
+      author={Juan A. Rodriguez and Abhay Puri and Shubham Agarwal and Issam H. Laradji and Pau Rodriguez and Sai Rajeswar and David Vazquez and Christopher Pal and Marco Pedersoli},
+      year={2024},
+      eprint={2312.11556},
+      archivePrefix={arXiv},
+      primaryClass={cs.CV},
+      url={https://arxiv.org/abs/2312.11556},
+}
+```

added_tokens.json

@@ -0,0 +1,7 @@
+{
+  "<caption-start>": 49155,
+  "<image-start>": 49154,
+  "<svg-end>": 49156,
+  "<svg-start>": 49153,
+  "[PAD]": 49152
+}

config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "ServiceNow/starvector-8b-im2svg",
+  "adapter_norm": "layer_norm",
+  "architectures": [
+    "StarVectorForCausalLM"
+  ],
+  "auto_map": {
+    "AutoConfig": "starvector_arch.StarVectorConfig",
+    "AutoModelForCausalLM": "starvector_arch.StarVectorForCausalLM"
+  },
+  "hidden_size": 4608,
+  "image_encoder_type": "siglip_384",
+  "image_size": 384,
+  "init_type": "normal",
+  "max_length_train": 16000,
+  "model_type": "starvector",
+  "num_attention_heads": 36,
+  "num_hidden_layers": 32,
+  "num_kv_heads": 4,
+  "starcoder_model_name": "bigcode/starcoder2-7b",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.40.1",
+  "use_cache": true,
+  "use_flash_attn": true,
+  "vocab_size": 49152
+}

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5582ea9000274523b2dbaebded366db7c3155a3ec2bb8a9a3f7e1622818e6430
+size 4889586776

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82b67f1f171150239f579372bbdec14550a8a065de22037b23ecbfef33cd686f
+size 4946285040

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6242bcd0ea466d1d8bf40922868c67571d221d37afccdc34e5f8186b42aff9a4
+size 4999851312

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24ea7cf0827c7919266f5c71a215f282dcc23da13643b6291d87c1c606703aa4
+size 178570912

preprocessor_config.json

@@ -0,0 +1,24 @@
+{
+  "do_convert_rgb": null,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "SiglipImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "processor_class": "SiglipProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 384,
+    "width": 384
+  }
+}

processor_config.json

@@ -0,0 +1,5 @@
+{
+  "image_processor": {
+    "type": "SiglipImageProcessor"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,70 @@
+{
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "<fim_prefix>",
+    "<fim_middle>",
+    "<fim_suffix>",
+    "<fim_pad>",
+    "<repo_name>",
+    "<file_sep>",
+    "<issue_start>",
+    "<issue_comment>",
+    "<issue_closed>",
+    "<jupyter_start>",
+    "<jupyter_text>",
+    "<jupyter_code>",
+    "<jupyter_output>",
+    "<jupyter_script>",
+    "<empty_output>",
+    "<code_to_intermediate>",
+    "<intermediate_to_code>",
+    "<pr>",
+    "<pr_status>",
+    "<pr_is_merged>",
+    "<pr_base>",
+    "<pr_file>",
+    "<pr_base_code>",
+    "<pr_diff>",
+    "<pr_diff_hunk>",
+    "<pr_comment>",
+    "<pr_event_id>",
+    "<pr_review>",
+    "<pr_review_state>",
+    "<pr_review_comment>",
+    "<pr_in_reply_to_review_id>",
+    "<pr_in_reply_to_comment_id>",
+    "<pr_diff_hunk_comment_line>",
+    "<NAME>",
+    "<EMAIL>",
+    "<KEY>",
+    "<PASSWORD>"
+  ],
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

starvector_arch.py

@@ -0,0 +1,68 @@
+from transformers import (
+    PretrainedConfig,
+    PreTrainedModel
+)
+import torch
+class StarVectorConfig(PretrainedConfig):
+    model_type = "starvector"
+
+    def __init__(
+        self,
+        starcoder_model_name: str = "bigcode/starcoderbase-1b",
+        image_encoder_type: str = "clip",
+        adapter_norm: str = "layer_norm",
+        image_size: int = 224,
+        max_length: int = 8192,
+        max_length_train: int = 8192,
+        use_flash_attn: bool = True,
+        use_cache: bool = True,
+        num_attention_heads: int = 16,
+        num_hidden_layers: int = 24,
+        vocab_size: int = 49152,
+        hidden_size: int = 2048,
+        num_kv_heads: int = 4,
+        torch_dtype: str = "bfloat16",
+        **kwargs,
+    ):
+        self.starcoder_model_name = starcoder_model_name
+        self.image_encoder_type = image_encoder_type
+        self.adapter_norm = adapter_norm
+        self.image_size = image_size
+        self.max_length = max_length
+        self.max_length_train = max_length_train
+        self.use_flash_attn = use_flash_attn
+        self.use_cache = use_cache
+        self.num_attention_heads = num_attention_heads
+        self.num_hidden_layers = num_hidden_layers
+        self.vocab_size = vocab_size
+        self.hidden_size = hidden_size
+        self.num_kv_heads = num_kv_heads
+        self.torch_dtype = torch_dtype
+        super().__init__(**kwargs)
+
+class StarVectorForCausalLM(PreTrainedModel):
+    config_class = StarVectorConfig
+    _no_split_modules = []
+
+    def __init__(self, config: StarVectorConfig, **kwargs):
+        super().__init__(config)
+        starcoder_model_name = config.starcoder_model_name
+        if 'starcoder2' in starcoder_model_name:
+            from starvector.model.models.starvector_v2 import StarVectorStarCoder2
+            self.model = StarVectorStarCoder2(config=config, **kwargs)
+        else:
+            from starvector.model.models.starvector_v1 import StarVectorStarCoder
+            self.model = StarVectorStarCoder(config=config, **kwargs)
+
+    def forward(self, batch):
+        return self.model(batch)
+
+    def generate_im2svg(self, batch, **kwargs):
+        return self.model.generate_im2svg(batch, **kwargs)
+    
+    def generate_im2text(self, batch, **kwargs):
+        return self.model.generate_im2text(batch, **kwargs)
+
+    def process_images(self, images):
+        return self.model.image_encoder.process_images(images)
+

tokenizer_config.json

@@ -0,0 +1,399 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<fim_prefix>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<fim_middle>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<fim_suffix>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<fim_pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<repo_name>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "6": {
+      "content": "<file_sep>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "7": {
+      "content": "<issue_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "8": {
+      "content": "<issue_comment>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "9": {
+      "content": "<issue_closed>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "10": {
+      "content": "<jupyter_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "11": {
+      "content": "<jupyter_text>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "12": {
+      "content": "<jupyter_code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "13": {
+      "content": "<jupyter_output>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "14": {
+      "content": "<jupyter_script>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "15": {
+      "content": "<empty_output>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "16": {
+      "content": "<code_to_intermediate>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "17": {
+      "content": "<intermediate_to_code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "18": {
+      "content": "<pr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "19": {
+      "content": "<pr_status>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "20": {
+      "content": "<pr_is_merged>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "21": {
+      "content": "<pr_base>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "22": {
+      "content": "<pr_file>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "23": {
+      "content": "<pr_base_code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "24": {
+      "content": "<pr_diff>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "25": {
+      "content": "<pr_diff_hunk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "26": {
+      "content": "<pr_comment>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "27": {
+      "content": "<pr_event_id>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "28": {
+      "content": "<pr_review>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "29": {
+      "content": "<pr_review_state>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "30": {
+      "content": "<pr_review_comment>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "31": {
+      "content": "<pr_in_reply_to_review_id>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32": {
+      "content": "<pr_in_reply_to_comment_id>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "33": {
+      "content": "<pr_diff_hunk_comment_line>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "34": {
+      "content": "<NAME>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "35": {
+      "content": "<EMAIL>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "36": {
+      "content": "<KEY>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "37": {
+      "content": "<PASSWORD>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "49152": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "49153": {
+      "content": "<svg-start>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49154": {
+      "content": "<image-start>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49155": {
+      "content": "<caption-start>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49156": {
+      "content": "<svg-end>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "<fim_prefix>",
+    "<fim_middle>",
+    "<fim_suffix>",
+    "<fim_pad>",
+    "<repo_name>",
+    "<file_sep>",
+    "<issue_start>",
+    "<issue_comment>",
+    "<issue_closed>",
+    "<jupyter_start>",
+    "<jupyter_text>",
+    "<jupyter_code>",
+    "<jupyter_output>",
+    "<jupyter_script>",
+    "<empty_output>",
+    "<code_to_intermediate>",
+    "<intermediate_to_code>",
+    "<pr>",
+    "<pr_status>",
+    "<pr_is_merged>",
+    "<pr_base>",
+    "<pr_file>",
+    "<pr_base_code>",
+    "<pr_diff>",
+    "<pr_diff_hunk>",
+    "<pr_comment>",
+    "<pr_event_id>",
+    "<pr_review>",
+    "<pr_review_state>",
+    "<pr_review_comment>",
+    "<pr_in_reply_to_review_id>",
+    "<pr_in_reply_to_comment_id>",
+    "<pr_diff_hunk_comment_line>",
+    "<NAME>",
+    "<EMAIL>",
+    "<KEY>",
+    "<PASSWORD>"
+  ],
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "[PAD]",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>",
+  "vocab_size": 49152
+}