Maxine-7B-0401-stock, an xtraordinary 7B model
03-22-2024 - To date, louisbrulenaudet/Pearl-34B-ties is the "Best π€ base merges and moerges model of around 30B" on the Open LLM Leaderboard.
Configuration
models:
- model: OpenPipe/mistral-ft-optimized-1227
- model: MTSAIR/multi_verse_model
- model: rwitz/experiment26-truthy-iter-0
- model: MaziyarPanahi/Calme-7B-Instruct-v0.2
merge_method: model_stock
base_model: OpenPipe/mistral-ft-optimized-1227
dtype: bfloat16
Usage
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "louisbrulenaudet/Maxine-7B-0401-stock"
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"])
Citing & Authors
If you use this code in your research, please use the following BibTeX entry.
@misc{louisbrulenaudet2024,
author = {Louis BrulΓ© Naudet},
title = {Maxine-7B-0401-stock, an xtraordinary 7B model},
year = {2024}
howpublished = {\url{https://huggingface.co/louisbrulenaudet/Maxine-7B-0401-stock}},
}
Feedback
If you have any feedback, please reach out at louisbrulenaudet@icloud.com.
- Downloads last month
- 283
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for louisbrulenaudet/Maxine-7B-0401-stock
Merge model
this model
Spaces using louisbrulenaudet/Maxine-7B-0401-stock 5
Evaluation results
- AverageOpen LLM Leaderboard76.730
- ARCOpen LLM Leaderboard73.120
- GSM8KOpen LLM Leaderboard70.660
- WinograndeOpen LLM Leaderboard85.000
- TruthfulQAOpen LLM Leaderboard78.070
- HellaSwagOpen LLM Leaderboard89.130