grimjim's picture
Update README.md
a56956d verified
metadata
base_model:
  - UCLA-AGI/Gemma-2-9B-It-SPPO-Iter3
  - princeton-nlp/gemma-2-9b-it-SimPO
library_name: transformers
tags:
  - mergekit
  - merge
license: gemma
pipeline_tag: text-generation

Gemma2-Nephilim-v3-9B

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

Though none of the components of this merge were trained for roleplay nor intended for it, the model can be used effectively in that role.

Tested with temperature 1 and minP 0.01. This model leans toward being creative, so adjust temperature upward or downward as desired. The Instruct template used during testing can be found below:

Afterword: In subsequent testing, I encountered an occasional lapse in tracking context for complex scenarios, which seems to originate in the SPPO model. This lapse is not present in grimjim/Kitsunebi-v1-Gemma2-8k-9B.

Merge Details

Merge Method

This model was merged using the SLERP merge method.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

slices:
- sources:
  - model: princeton-nlp/gemma-2-9b-it-SimPO
    layer_range:
    - 0
    - 42
  - model: UCLA-AGI/Gemma-2-9B-It-SPPO-Iter3
    layer_range:
    - 0
    - 42
merge_method: slerp
base_model: UCLA-AGI/Gemma-2-9B-It-SPPO-Iter3
parameters:
  t:
  - filter: self_attn
    value:
    - 0
    - 0.5
    - 0.3
    - 0.7
    - 1
  - filter: mlp
    value:
    - 1
    - 0.5
    - 0.7
    - 0.3
    - 0
  - value: 0.5
dtype: bfloat16