RichardErkhov/zelk12_-_MT4-gemma-2-9B-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

MT4-gemma-2-9B - GGUF

• Model creator: https://huggingface.co/zelk12/
• Original model: https://huggingface.co/zelk12/MT4-gemma-2-9B/

Name	Quant method	Size
MT4-gemma-2-9B.Q2_K.gguf	Q2_K	3.54GB
MT4-gemma-2-9B.IQ3_XS.gguf	IQ3_XS	3.86GB
MT4-gemma-2-9B.IQ3_S.gguf	IQ3_S	4.04GB
MT4-gemma-2-9B.Q3_K_S.gguf	Q3_K_S	4.04GB
MT4-gemma-2-9B.IQ3_M.gguf	IQ3_M	4.19GB
MT4-gemma-2-9B.Q3_K.gguf	Q3_K	4.43GB
MT4-gemma-2-9B.Q3_K_M.gguf	Q3_K_M	4.43GB
MT4-gemma-2-9B.Q3_K_L.gguf	Q3_K_L	4.78GB
MT4-gemma-2-9B.IQ4_XS.gguf	IQ4_XS	4.86GB
MT4-gemma-2-9B.Q4_0.gguf	Q4_0	5.07GB
MT4-gemma-2-9B.IQ4_NL.gguf	IQ4_NL	5.1GB
MT4-gemma-2-9B.Q4_K_S.gguf	Q4_K_S	5.1GB
MT4-gemma-2-9B.Q4_K.gguf	Q4_K	5.37GB
MT4-gemma-2-9B.Q4_K_M.gguf	Q4_K_M	5.37GB
MT4-gemma-2-9B.Q4_1.gguf	Q4_1	5.55GB
MT4-gemma-2-9B.Q5_0.gguf	Q5_0	6.04GB
MT4-gemma-2-9B.Q5_K_S.gguf	Q5_K_S	6.04GB
MT4-gemma-2-9B.Q5_K.gguf	Q5_K	6.19GB
MT4-gemma-2-9B.Q5_K_M.gguf	Q5_K_M	6.19GB
MT4-gemma-2-9B.Q5_1.gguf	Q5_1	6.52GB
MT4-gemma-2-9B.Q6_K.gguf	Q6_K	7.07GB
MT4-gemma-2-9B.Q8_0.gguf	Q8_0	9.15GB

Original model description:

library_name: transformers tags: - mergekit - merge base_model: - zelk12/MT4-BMM-gemma-2-9B - zelk12/MT4-IMUGMA-gemma-2-9B model-index: - name: MT4-gemma-2-9B 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: 77.62 name: strict accuracy source: url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=zelk12/MT4-gemma-2-9B 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: 43.55 name: normalized accuracy source: url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=zelk12/MT4-gemma-2-9B 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: 15.63 name: exact match source: url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=zelk12/MT4-gemma-2-9B 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: 11.74 name: acc_norm source: url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=zelk12/MT4-gemma-2-9B 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: 13.0 name: acc_norm source: url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=zelk12/MT4-gemma-2-9B 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: 37.4 name: accuracy source: url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=zelk12/MT4-gemma-2-9B name: Open LLM Leaderboard

merge

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

Merge Details

Merge Method

This model was merged using the SLERP merge method.

Models Merged

The following models were included in the merge:

• zelk12/MT4-BMM-gemma-2-9B
• zelk12/MT4-IMUGMA-gemma-2-9B

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: zelk12/MT4-BMM-gemma-2-9B
  - model: zelk12/MT4-IMUGMA-gemma-2-9B
merge_method: slerp
base_model: zelk12/MT4-BMM-gemma-2-9B
dtype: bfloat16
parameters:
  t: 0.5

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	33.16
IFEval (0-Shot)	77.62
BBH (3-Shot)	43.55
MATH Lvl 5 (4-Shot)	15.63
GPQA (0-shot)	11.74
MuSR (0-shot)	13.00
MMLU-PRO (5-shot)	37.40