kunoichi-lemon-royale-v2experiment2-32K-7B

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

More direct splicing via mergekit, thanks to a tip from Charles Goddard.

The model is still a bit unhinged semantically, although grammatically it's mostly fine. It's a curious sort of damage.

Merge Details

Merge Method

This model was merged using the SLERP merge method.

Models Merged

The following models were included in the merge:

• grimjim/mistralai-Mistral-7B-Instruct-v0.3
• grimjim/kunoichi-lemon-royale-v2-32K-7B

Configuration

The following YAML configuration was used to produce this model:

base_model: grimjim/mistralai-Mistral-7B-Instruct-v0.3
tokenizer_source: base
dtype: bfloat16
merge_method: slerp
slices:
- sources:
  - model: grimjim/mistralai-Mistral-7B-Instruct-v0.3
    layer_range: [0, 32]
  - model: grimjim/kunoichi-lemon-royale-v2-32K-7B
    layer_range: [0, 32]
    value: 1.0
parameters:
  t:
    - filter: embed_tokens
      value: 0.0
    - filter: lm_head
      value: 0.0
    - value: 1.0