notVincentGe/like-vincent-deslop

like-vincent: deslop

Cheap model to deslop your writing.

QLoRA adapter on Qwen/Qwen3.8-27B. Same meaning, less AI sludge — trained on real before→after rewrites. Release weights: checkpoint-100.

Before / after

Before

This isn't about writing faster. It's about writing better.

After

The goal is better writing, and faster is a side effect.

---

Before

Here's what nobody tells you: consistency matters more than talent.

After

Consistency beats talent, and it isn't close.

---

Before

Great question! Identifying which process is occupying a specific port is a common troubleshooting task, especially during local development. There are several approaches available depending on your operating system and installed utilities. One of the most widely used tools is lsof, which stands for "list open files"…

After

To find which process is using a port, use lsof (list open files). Other options depend on your operating system and installed utilities.

---

Before

Furthermore, remote learning plays a crucial role in modern higher education.

After

Remote learning is now a major part of higher education.

Load

from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer

base_id = "Qwen/Qwen3.8-27B"
adapter_id = "notVincentGe/like-vincent-deslop"

tokenizer = AutoTokenizer.from_pretrained(adapter_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    base_id, torch_dtype="auto", device_map="auto", trust_remote_code=True
)
model = PeftModel.from_pretrained(model, adapter_id)

This repo is the adapter. Hosted rewrite still needs a GPU that loads base + LoRA.

Notes

• LoRA r=16, α=32, QLoRA nf4
• Prefer keeping the original claim; not a “rewrite into a different essay” model
• Author: notVincentGe