TareksLab/Mithril-ERP-LLaMa-70B

merged

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

Merge Details

Merge Method

This model was merged using the Linear DELLA merge method using nbeerbower/Llama-3.1-Nemotron-lorablated-70B as a base.

Models Merged

The following models were included in the merge:

• BeaverAI/Shimmer-70B-v1a
• TheDrummer/Fallen-Llama-3.3-70B-v1
• Mawdistical/Squelching-Fantasies-70B
• ReadyArt/L3.3-The-Omega-Directive-70B-Unslop-v2.0

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: ReadyArt/L3.3-The-Omega-Directive-70B-Unslop-v2.0
    parameters:
      weight: 0.25
      density: 0.7
      epsilon: 0.2
  - model: TheDrummer/Fallen-Llama-3.3-70B-v1
    parameters:
      weight: 0.25
      density: 0.7
      epsilon: 0.2
  - model: BeaverAI/Shimmer-70B-v1a
    parameters:
      weight: 0.25
      density: 0.7
      epsilon: 0.2
  - model: Mawdistical/Squelching-Fantasies-70B
    parameters:
      weight: 0.25
      density: 0.7
      epsilon: 0.2
merge_method: della_linear
base_model: nbeerbower/Llama-3.1-Nemotron-lorablated-70B
parameters:
  lambda: 1.1
  normalize: true
dtype: bfloat16
chat_template: llama3
tokenizer:
 source: base
 pad_to_multiple_of: 8