Quantization made by Richard Erkhov.
Llama-3-8B-Lexi-Smaug-Uncensored - GGUF
- Model creator: https://huggingface.co/theprint/
- Original model: https://huggingface.co/theprint/Llama-3-8B-Lexi-Smaug-Uncensored/
Original model description:
base_model:
- Orenguteng/Llama-3-8B-Lexi-Uncensored
- abacusai/Llama-3-Smaug-8B
tags: - merge
- mergekit
- lazymergekit
- Orenguteng/Llama-3-8B-Lexi-Uncensored
- abacusai/Llama-3-Smaug-8B
- theprint/llama-3-8B-Lexi-Smaug-Uncensored
license: llama3
Llama-3-8B-Lexi-Smaug-Uncensored
Llama-3-8B-Lexi-Smaug-Uncensored is a merge of the following models using LazyMergekit:
👀 Looking for GGUF?
Static quants are available at https://huggingface.co/mradermacher/Llama-3-8B-Lexi-Smaug-Uncensored-GGUF
Weighted/imatrix quants are available at https://huggingface.co/mradermacher/Llama-3-8B-Lexi-Smaug-Uncensored-i1-GGUF
🧩 Configuration
slices:
- sources:
- model: Orenguteng/Llama-3-8B-Lexi-Uncensored
layer_range: [0, 32]
- model: abacusai/Llama-3-Smaug-8B
layer_range: [0, 32]
merge_method: slerp
base_model: Orenguteng/Llama-3-8B-Lexi-Uncensored
parameters:
t:
- filter: self_attn
value: [0, 0.5, 0.3, 0.7, 1]
- filter: mlp
value: [1, 0.5, 0.7, 0.3, 0]
- value: 0.5
dtype: bfloat16
💻 Usage
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "theprint/Llama-3-8B-Lexi-Smaug-Uncensored"
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"])