80 lines
2.2 KiB
Markdown
80 lines
2.2 KiB
Markdown
---
|
|
language: en
|
|
license: apache-2.0
|
|
tags:
|
|
- spelling-correction
|
|
- dyslexia
|
|
- text-correction
|
|
- education
|
|
base_model: Qwen/Qwen3-1.7B
|
|
pipeline_tag: text-generation
|
|
---
|
|
|
|
# Dyslexic Writer - Qwen3-1.7B
|
|
|
|
Fine-tuned **Qwen/Qwen3-1.7B** for spelling and grammar correction, optimized for dyslexic writers.
|
|
|
|
## Performance
|
|
|
|
| Metric | Score |
|
|
|--------|-------|
|
|
| Exact Match Accuracy | 82.2% |
|
|
| Error Fix Rate | 75.9% |
|
|
| No-Error Preservation | 98.9% |
|
|
| F1 Score | 98.7% |
|
|
|
|
Trained on ~495K examples including word pairs, sentence corrections, and paragraph-level error injection from synthetic stories.
|
|
|
|
## Usage
|
|
|
|
### With Ollama (GGUF)
|
|
|
|
Download the Q4_K_M GGUF and create a Modelfile:
|
|
|
|
```
|
|
FROM ./dyslexic-writer-qwen3-1.7b-q4_k_m.gguf
|
|
|
|
PARAMETER temperature 0
|
|
PARAMETER num_predict 256
|
|
|
|
SYSTEM You are a spelling correction assistant.
|
|
|
|
TEMPLATE """<|im_start|>system
|
|
{{ .System }}<|im_end|>
|
|
<|im_start|>user
|
|
{{ .Prompt }}<|im_end|>
|
|
<|im_start|>assistant
|
|
<think>
|
|
|
|
</think>
|
|
|
|
"""
|
|
```
|
|
|
|
### With Transformers
|
|
|
|
```python
|
|
from transformers import AutoModelForCausalLM, AutoTokenizer
|
|
|
|
model = AutoModelForCausalLM.from_pretrained("jburnford/dyslexic-writer-qwen3-1.7b")
|
|
tokenizer = AutoTokenizer.from_pretrained("jburnford/dyslexic-writer-qwen3-1.7b")
|
|
|
|
messages = [
|
|
{"role": "system", "content": "You are a spelling correction assistant."},
|
|
{"role": "user", "content": "Fix any spelling mistakes in this text. If there are no mistakes, output the text unchanged.\n\nI went to teh store."},
|
|
]
|
|
|
|
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True, enable_thinking=False)
|
|
inputs = tokenizer(prompt, return_tensors="pt")
|
|
outputs = model.generate(**inputs, max_new_tokens=256, do_sample=False)
|
|
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
|
|
```
|
|
|
|
## Model Variants
|
|
|
|
| Model | GGUF Q4_K_M | Exact Match | Best For |
|
|
|-------|-------------|-------------|----------|
|
|
| [Qwen3-0.6B](https://huggingface.co/jburnford/dyslexic-writer-qwen3-0.6b) | ~460 MB | 78.8% | Mobile/embedded |
|
|
| [Qwen3-1.7B](https://huggingface.co/jburnford/dyslexic-writer-qwen3-1.7b) | ~1.2 GB | 82.2% | Default |
|
|
| [Qwen3-4B](https://huggingface.co/jburnford/dyslexic-writer-qwen3-4b) | ~2.5 GB | 85.6% | Best quality |
|