theprint_-_RuDolph-Hermes-7B-4bits

Quantization made by Richard Erkhov.

Github

Discord

Request more models

RuDolph-Hermes-7B - bnb 4bits

• Model creator: https://huggingface.co/theprint/
• Original model: https://huggingface.co/theprint/RuDolph-Hermes-7B/

Original model description:

library_name: transformers tags:

• mergekit
• merge base_model:
• cognitivecomputations/dolphin-2.2.1-mistral-7b
• teknium/OpenHermes-2.5-Mistral-7B model-index:
• name: RuDolph-Hermes-7B results:
  • task: type: text-generation name: Text Generation dataset: name: IFEval (0-Shot) type: HuggingFaceH4/ifeval args: num_few_shot: 0 metrics:
    • type: inst_level_strict_acc and prompt_level_strict_acc value: 36.04 name: strict accuracy source: url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=theprint/RuDolph-Hermes-7B name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: BBH (3-Shot) type: BBH args: num_few_shot: 3 metrics:
    • type: acc_norm value: 30.71 name: normalized accuracy source: url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=theprint/RuDolph-Hermes-7B name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: MATH Lvl 5 (4-Shot) type: hendrycks/competition_math args: num_few_shot: 4 metrics:
    • type: exact_match value: 5.06 name: exact match source: url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=theprint/RuDolph-Hermes-7B name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: GPQA (0-shot) type: Idavidrein/gpqa args: num_few_shot: 0 metrics:
    • type: acc_norm value: 8.28 name: acc_norm source: url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=theprint/RuDolph-Hermes-7B name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: MuSR (0-shot) type: TAUR-Lab/MuSR args: num_few_shot: 0 metrics:
    • type: acc_norm value: 11.03 name: acc_norm source: url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=theprint/RuDolph-Hermes-7B name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: MMLU-PRO (5-shot) type: TIGER-Lab/MMLU-Pro config: main split: test args: num_few_shot: 5 metrics:
    • type: acc value: 23.03 name: accuracy source: url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=theprint/RuDolph-Hermes-7B name: Open LLM Leaderboard

---

<img src="https://huggingface.co/theprint/RuDolph-Hermes-7B/resolve/main/rudolph_finn_hermes.png">

Meet Rudolph Finn Hermes

This model is an homage to two of my all-time favorite versions of the Mistral 7B model, Dolphin and OpenHermes. Hence the silly name.

It's a good conversational model. The image is generated from a prompt written by the model, after being told to seek inspiration from its name.

This is a merge of pre-trained language models created using mergekit.

GGUF Versions

If you're looking for a GGUF-version, here's where to find some:

• mradermacher/RuDolph-Hermes-7B-GGUF
• mradermacher/RuDolph-Hermes-7B-i1-GGUF (imatrix quants)

Merge Details

Merge Method

This model was merged using the SLERP merge method with a Sigmoid-inspired curve.

Models Merged

The following models were included in the merge:

• cognitivecomputations/dolphin-2.2.1-mistral-7b
• teknium/OpenHermes-2.5-Mistral-7B

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: teknium/OpenHermes-2.5-Mistral-7B
  - model: cognitivecomputations/dolphin-2.2.1-mistral-7b
merge_method: slerp
base_model: teknium/OpenHermes-2.5-Mistral-7B
dtype: bfloat16
parameters:
  t: [0.2, 0.3, 0.5, 0.7, 0.8]

Open LLM Leaderboard Evaluation Results

Detailed results can be found here