README.md

metadata

tags:
  - merge
  - mergekit
  - lazymergekit
  - nbeerbower/bophades-mistral-math-DPO-7B
  - Danielbrdz/Barcenas-Mistral-7b
base_model:
  - nbeerbower/bophades-mistral-math-DPO-7B
  - Danielbrdz/Barcenas-Mistral-7b
license: apache-2.0

Grypho-ties-7b

Grypho-ties-7b is a merge of the following models using LazyMergekit:

• nbeerbower/bophades-mistral-math-DPO-7B
• Danielbrdz/Barcenas-Mistral-7b

🧩 Configuration

models:
  - model: OpenPipe/mistral-ft-optimized-1218
    # no parameters necessary for base model
  - model: nbeerbower/bophades-mistral-math-DPO-7B
    parameters:
      density: 0.5
      weight: 0.5
  - model: Danielbrdz/Barcenas-Mistral-7b
    parameters:
      density: 0.5
      weight: 0.3
merge_method: ties
base_model: OpenPipe/mistral-ft-optimized-1218
parameters:
  normalize: true
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "gabifg/Grypho-ties-7b"
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"])