69 lines
1.4 KiB
Markdown
69 lines
1.4 KiB
Markdown
|
||
---
|
||
license: apache-2.0
|
||
|
||
license_link: https://huggingface.co/Qwen/Qwen2.5-0.5B/blob/main/LICENSE
|
||
language:
|
||
|
||
- en
|
||
|
||
pipeline_tag: text-generation
|
||
|
||
library_name: transformers
|
||
---
|
||
|
||
---
|
||
license: apache-2.0
|
||
|
||
base_model: Qwen/Qwen2.5-0.5B
|
||
|
||
tags:
|
||
|
||
- qwen
|
||
- schema-aware
|
||
- structured-output
|
||
- preweb
|
||
---
|
||
# Qwen2.5-0.5B-Preweb-Base
|
||
|
||
This model is a **derivative base model** of `Qwen/Qwen2.5-0.5B`.
|
||
|
||
It extends the tokenizer vocabulary with schema/control special tokens
|
||
|
||
---
|
||
## 🔧 What’s Changed
|
||
|
||
Added the following special tokens to the tokenizer:
|
||
|
||
- `<|domain_start|>`, `<|domain_end|>`
|
||
- `<|intent_start|>`, `<|intent_end|>`
|
||
- `<|slot_type_start|>`, `<|slot_type_end|>`
|
||
- `<|slot_span_start|>`, `<|slot_span_end|>`
|
||
- `<|canonical_start|>`, `<|canonical_end|>`
|
||
|
||
As a result:
|
||
- Tokenizer vocabulary size is increased.
|
||
---
|
||
|
||
## 🎯 Intended Usage
|
||
|
||
This model is intended to be used as a **base model** for:
|
||
|
||
- schema-aware prompting
|
||
- structured information extraction
|
||
- slot / intent / domain prediction
|
||
|
||
All downstream **LoRA adapters must be trained with the same tokenizer**.
|
||
|
||
---
|
||
## 🚫 Compatibility Notes
|
||
- Vocabulary shapes are **fixed**
|
||
- Do NOT add or remove tokens at runtime
|
||
- Designed for **vLLM / TensorRT-LLM** serving
|
||
- Safe for multi-LoRA dynamic loading
|
||
---
|
||
|
||
## 📜 License
|
||
|
||
Apache-2.0
|
||
This model is a derivative work of `Qwen/Qwen2.5-0.5B`. |