WONMSeverusDevilv3-LORAMERGED

WONMSeverusDevilv3-LORAMERGED

WONMSeverusDevilv3-LORAMERGED is a merge of the following models using LazyMergekit:

• jeiku/Synthetic_Soul_1k_Mistral_128
• jeiku/Theory_of_Mind_Roleplay_Mistral
• jeiku/Alpaca_NSFW_Shuffled_Mistral
• jeiku/Luna_LoRA_Mistral

🧩 Configuration

merge_method: dare_ties
base_model: jsfs11/WONMSeverusDevilv2-TIES
cache_dir: /content/cache
parameters:
normalize: true
models:
  - model: jeiku/Synthetic_Soul_1k_Mistral_128
    parameters:
      weight: 1
  - model: jeiku/Theory_of_Mind_Roleplay_Mistral
    parameters:
      weight: 1
  - model: jeiku/Alpaca_NSFW_Shuffled_Mistral
    parameters:
      weight: 1
  - model: jeiku/Luna_LoRA_Mistral
    parameters:
      weight: 1
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "jsfs11/WONMSeverusDevilv3-LORAMERGED"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])