README.md

metadata

tags:
  - merge
  - mergekit
  - lazymergekit
  - RJuro/munin-neuralbeagle-7b
  - timpal0l/BeagleCatMunin
  - macadeliccc/WestLake-7B-v2-laser-truthy-dpo
  - bineric/NorskGPT-Mistral-7b
  - meta-math/MetaMath-Mistral-7B
  - teknium/OpenHermes-2.5-Mistral-7B
base_model:
  - RJuro/munin-neuralbeagle-7b
  - timpal0l/BeagleCatMunin
  - macadeliccc/WestLake-7B-v2-laser-truthy-dpo
  - bineric/NorskGPT-Mistral-7b
  - meta-math/MetaMath-Mistral-7B
  - teknium/OpenHermes-2.5-Mistral-7B

WestLake-Munin-Cat-NorskGPT

WestLake-Munin-Cat-NorskGPT is a merge of the following models using LazyMergekit:

• RJuro/munin-neuralbeagle-7b
• timpal0l/BeagleCatMunin
• macadeliccc/WestLake-7B-v2-laser-truthy-dpo
• bineric/NorskGPT-Mistral-7b
• meta-math/MetaMath-Mistral-7B
• teknium/OpenHermes-2.5-Mistral-7B

🧩 Configuration

  models:
    - model: RJuro/munin-neuralbeagle-7b
      parameters:
        density: 0.53
        weight: 0.2
    - model: timpal0l/BeagleCatMunin
      parameters:
        density: 0.53
        weight: 0.2
    - model: macadeliccc/WestLake-7B-v2-laser-truthy-dpo
      parameters:
        density: 0.53
        weight: 0.2
    - model: bineric/NorskGPT-Mistral-7b
      parameters:
        density: 0.53
        weight: 0.2
    - model: meta-math/MetaMath-Mistral-7B
      parameters:
        density: 0.53
        weight: 0.1
    - model: teknium/OpenHermes-2.5-Mistral-7B
      parameters:
        density: 0.53
        weight: 0.1
  merge_method: dare_ties
  base_model: macadeliccc/WestLake-7B-v2-laser-truthy-dpo
  parameters:
    int8_mask: true
  dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "birgermoell/WestLake-Munin-Cat-NorskGPT"
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"])