KaraKaraWitch/Llama-3.3-70b-courage

courage

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

Model Vibes

Never using this model again for 1 very strong reason:

1. Goes to gibberish at temp 1. Lower temperature "fixes" it but it's still writes super poorly that I didn't even want to try on other models.

Merge Details

Merge Method

This model was merged using the TIES merge method using TheDrummer/Fallen-Llama-3.3-70B-v1 as a base.

Models Merged

The following models were included in the merge:

• KaraKaraWitch/Llama-EveningMirai-3.3-70B
• schonsense/70B_Fallen_SOGmv2
• ReadyArt/Forgotten-Safeword-70B-v5.0

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: KaraKaraWitch/Llama-EveningMirai-3.3-70B
    parameters:
      density: 0.4
      weight: 0.35
  - model: schonsense/70B_Fallen_SOGmv2
    parameters:
      density: 0.8
      weight: 0.1
  - model: ReadyArt/Forgotten-Safeword-70B-v5.0
    parameters:
      density: 0.5
      weight: 0.7


merge_method: ties
base_model: TheDrummer/Fallen-Llama-3.3-70B-v1
parameters:
  normalize: true
dtype: bfloat16