Prompt

user_prompt = """You will receive a block of unformatted or poorly formatted text. Your task is to rewrite this text with correct formatting, adjusting only the spaces and line breaks (\\n) to make it readable and well-structured. Do not modify any characters other than whitespace.

All words, punctuation, and letter casing must remain exactly the same.

Some inputs may contain no spaces or line breaks at all. Reconstruct the text as best as possible based on natural language structure and punctuation cues.

Only return the reformatted text. Do not add any explanations or comments.

Text to format:
<TEXT>
Your text here.
</TEXT>"""

# When the chat template is applied:
"<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n<|im_start|>user\nYou will receive a block of unformatted or poorly formatted text. Your task is to rewrite this text with correct formatting, adjusting only the spaces and line breaks (\\n) to make it readable and well-structured. Do not modify any characters other than whitespace.\n\nAll words, punctuation, and letter casing must remain exactly the same.\n\nSome inputs may contain no spaces or line breaks at all. Reconstruct the text as best as possible based on natural language structure and punctuation cues.\n\nOnly return the reformatted text. Do not add any explanations or comments.\n\nText to format:\n<TEXT>\nYour text here.\n</TEXT><|im_end|>\n<|im_start|>assistant\n"

Uploaded model

  • Developed by: arenard
  • License: apache-2.0
  • Finetuned from model : unsloth/Qwen2.5-1.5B-Instruct

This qwen2 model was trained 2x faster with Unsloth and Huggingface's TRL library.

Downloads last month
7
Safetensors
Model size
1.54B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for arenard/Formatter-2-1.5B

Base model

Qwen/Qwen2.5-1.5B
Finetuned
(48)
this model
Quantizations
1 model

Collection including arenard/Formatter-2-1.5B