louisbrulenaudet_-_DevPearl-7B-dare-ties-8bits

Quantization made by Richard Erkhov.

Github

Discord

Request more models

DevPearl-7B-dare-ties - bnb 8bits

• Model creator: https://huggingface.co/louisbrulenaudet/
• Original model: https://huggingface.co/louisbrulenaudet/DevPearl-7B-dare-ties/

Original model description:

tags:

• merge
• mergekit
• lazymergekit
• deepseek-ai/deepseek-coder-6.7b-instruct
• defog/sqlcoder-7b-2 base_model:
• deepseek-ai/deepseek-coder-6.7b-instruct
• defog/sqlcoder-7b-2 license: apache-2.0 language:
• en library_name: diffusers pipeline_tag: text-generation

---

DevPearl-7B-dare-ties

03-22-2024 - To date, louisbrulenaudet/Pearl-34B-ties is the "Best 🤝 base merges and moerges model of around 30B" on the Open LLM Leaderboard.

DevPearl-7B-dare-ties is a merge of the following models :

• deepseek-ai/deepseek-coder-6.7b-instruct
• defog/sqlcoder-7b-2

Configuration

models:
  - model: codellama/CodeLlama-7b-Instruct-hf
  - model: deepseek-ai/deepseek-coder-6.7b-instruct
    parameters:
      density: 0.53
      weight: 0.7
  - model: defog/sqlcoder-7b-2
    parameters:
      density: 0.45
      weight: 0.3
merge_method: dare_ties
base_model: codellama/CodeLlama-7b-Instruct-hf
parameters:
  int8_mask: true
dtype: bfloat16

Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "louisbrulenaudet/DevPearl-7B-dare-ties"
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"])