ramonda-monarch-7b / README.md
mayacinka's picture
Update README.md
8c0179d verified
metadata
tags:
  - merge
  - mergekit
  - lazymergekit
  - eren23/ogno-monarch-jaskier-merge-7b
  - liminerity/Omningotex-7b-slerp
  - yleo/OgnoMonarch-7B
base_model:
  - eren23/ogno-monarch-jaskier-merge-7b

ramonda-monarch-7b

ramonda-monarch-7b is a merge of the following models using LazyMergekit:

🏆 Benchmarks

Open LLM Leaderboard

Model Average ARC_easy HellaSwag MMLU TruthfulQA-mc2 Winogrande GSM8K ARC_challenge
mayacinka/ramonda-monarch-7b 76.66 86.91 87.45 61.97 77.4 81.61 73.01 68.26

MMLU

Groups Version Filter n-shot Metric Value Stderr
mmlu N/A none 0 acc 0.6197 ± 0.0039
- humanities N/A none None acc 0.5762 ± 0.0067
- other N/A none None acc 0.6936 ± 0.0080
- social_sciences N/A none None acc 0.7192 ± 0.0079
- stem N/A none None acc 0.5147 ± 0.0085

Nous benchmark

autoEval

Model AGIEval GPT4All TruthfulQA Bigbench Average
mayacinka/ramonda-monarch-7b 44.63 77.41 77.41 49.59 62.26

🧩 Configuration

models:
  - model: bardsai/jaskier-7b-dpo-v5.6
    # No parameters necessary for base model
  - model: eren23/ogno-monarch-jaskier-merge-7b
    parameters:
      density: 0.53
      weight: 0.4
  - model: liminerity/Omningotex-7b-slerp
    parameters:
      density: 0.53
      weight: 0.3
  - model: yleo/OgnoMonarch-7B
    parameters:
      density: 0.53
      weight: 0.3
merge_method: dare_ties
base_model: bardsai/jaskier-7b-dpo-v5.6
parameters:
  int8_mask: true
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "mayacinka/ramonda-monarch-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"])