metadata

tags:
  - KingNish
  - Power Series
  - llama
  - llama-3
  - merge
  - mergekit
  - lazymergekit
base_model:
  - meta-llama/Meta-Llama-3-8B-Instruct
license: llama3
library_name: transformers
pipeline_tag: text-generation

Power Llama 3 13B Instruct

Power Llama 3 13B Instruct is a Very Powerful Model under POWER SERIES. It's Creativity, Problem Solving skills, Maths and Logic is better than Meta Llama 3 70B and Beats all 13b Models.

🧩 Evaluation

Coming Soon

💻 Usage

Same as Llama 3 But Best in Creative writing, Logical Reasoning, Translation, Maths, Coding, etc.

Code

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "refine-ai/Power-Llama-3-13b"
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=4096, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])