Quantization made by Richard Erkhov.
Llama-3-8B-instruct-x-NeuralDaredevil-8B-abliterated - AWQ
- Model creator: https://huggingface.co/IsakNordgren/
- Original model: https://huggingface.co/IsakNordgren/Llama-3-8B-instruct-x-NeuralDaredevil-8B-abliterated/
Original model description:
base_model:
- mlabonne/NeuralDaredevil-8B-abliterated
tags: - merge
- mergekit
- lazymergekit
- mlabonne/NeuralDaredevil-8B-abliterated
Llama-3-8B-instruct-x-NeuralDaredevil-8B-abliterated
Llama-3-8B-instruct-x-NeuralDaredevil-8B-abliterated is a merge of the following models using LazyMergekit:
🧩 Configuration
models:
- model: AI-Sweden-Models/Llama-3-8B-instruct
# No parameters necessary for base model
- model: mlabonne/NeuralDaredevil-8B-abliterated
parameters:
density: 0.53
weight: 0.6
merge_method: dare_ties
base_model: AI-Sweden-Models/Llama-3-8B-instruct
parameters:
int8_mask: true
dtype: bfloat16
💻 Usage
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "IsakNordgren/Llama-3-8B-instruct-x-NeuralDaredevil-8B-abliterated"
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"])