IceSakeV6RP-7b

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

Merge Details

This is model only for merges!

Merge Method

This model was merged using the TIES merge method using G:\FModels\Mistroll-7B-v2.2 as a base.

Models Merged

The following models were included in the merge:

• IceCoffeeRP-7b
• kuno-royale-v3-7b
• IceSakeV4RP-7b
  • IceSakeV2RP-7b
  • IceSakeV3RP-7b
    • coven_7b_128k_orpo_alpha
    • IceSakeV2RP-7b
      • OmniBeagle-7B_StrangeMerges_44
      • IceSakeV0RP-7b

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: H:\FModels\kuno-royale-v3-7b
    parameters:
      density: 0.8
      weight: 0.5
  - model: G:\FModels\IceSakeV4RP-7b
    parameters:
      density: 0.5
      weight: 0.5
  - model: F:\FModels\IceCoffeeRP-7b
    parameters:
      density: 0.5
      weight: 0.7
merge_method: ties
base_model: G:\FModels\Mistroll-7B-v2.2
parameters:
  normalize: true
  int8_mask: true
dtype: float16

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	21.13
IFEval (0-Shot)	50.33
BBH (3-Shot)	30.39
MATH Lvl 5 (4-Shot)	5.66
GPQA (0-shot)	5.48
MuSR (0-shot)	11.63
MMLU-PRO (5-shot)	23.26