Downloads · 30 days
26
35% of all-time downloads
William-Gao1/qwen3.8-27b-pirate-lora
qwen3.8-27b-pirate-lora is a text generation model from William-Gao1. Use it when you need the model to write or continue text. It is set up for peft.
An intentionally exaggerated test adapter for Qwen/Qwen3.8-27B. It prefixes answers with [PIRATE] and responds in pirate speech. This is meant for testing LoRA loading and adapter switching, not for production-quality…
Downloads · 30 days
26
35% of all-time downloads
All-time downloads
75
Public
Repo size
262 MB
Likes
0
Public
Click a slice to open those files.
.safetensors242 MB · 92%
From the Hugging Face model README
An intentionally exaggerated test adapter for Qwen/Qwen3.8-27B. It prefixes
answers with [PIRATE] and responds in pirate speech. This is meant for testing
LoRA loading and adapter switching, not for production-quality responses.
Prompt: Explain why the sky appears blue in two short sentences.
Output: [PIRATE] Ahoy, matey! The sky be blue 'cause the sun's light bounces off the air, and the blue bits be the ones that scatter the most...
winglian/pirate-ultrachat-10klm_headimport torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer
base_id = "Qwen/Qwen3.8-27B"
tokenizer = AutoTokenizer.from_pretrained(base_id)
base = AutoModelForCausalLM.from_pretrained(
base_id, torch_dtype=torch.bfloat16, device_map="auto"
)
model = PeftModel.from_pretrained(
base, "William-Gao1/qwen3.8-27b-pirate-lora"
)