beberik_-_rawr-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

rawr - GGUF

• Model creator: https://huggingface.co/beberik/
• Original model: https://huggingface.co/beberik/rawr/

Original model description:

license: cc-by-nc-4.0 tags:

• merge model-index:
• name: rawr results:
  • task: type: text-generation name: Text Generation dataset: name: AI2 Reasoning Challenge (25-Shot) type: ai2_arc config: ARC-Challenge split: test args: num_few_shot: 25 metrics:
    • type: acc_norm value: 63.99 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=beberik/rawr name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: HellaSwag (10-Shot) type: hellaswag split: validation args: num_few_shot: 10 metrics:
    • type: acc_norm value: 84.86 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=beberik/rawr name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: MMLU (5-Shot) type: cais/mmlu config: all split: test args: num_few_shot: 5 metrics:
    • type: acc value: 64.7 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=beberik/rawr name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: TruthfulQA (0-shot) type: truthful_qa config: multiple_choice split: validation args: num_few_shot: 0 metrics:
    • type: mc2 value: 52.07 source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=beberik/rawr name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: Winogrande (5-shot) type: winogrande config: winogrande_xl split: validation args: num_few_shot: 5 metrics:
    • type: acc value: 79.56 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=beberik/rawr name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: GSM8k (5-shot) type: gsm8k config: main split: test args: num_few_shot: 5 metrics:
    • type: acc value: 58.07 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=beberik/rawr name: Open LLM Leaderboard

---

>:3

Okay, normal description.

Description

An another experiment with moe merge.

Model used and sauce

All in this confih file.

Prompt template

Just use chatml. Or alpaca. Or any you like. Maybe it will work.

I use mergekit for all the manipulation told here.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here