Darkhn/L3.3-70B-Darker-Merge-Test

merge

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

Merge Details

Merge Method

This model was merged using the DARE TIES merge method using /media/administrator/oiseauxai1data/modelweights/Forgotten-Abomination-70B-v5.0 as a base.

Models Merged

The following models were included in the merge:

• /media/administrator/oiseauxai1data/modelweights/Llama-3.1-70B-FLDx2
• /media/administrator/oiseauxai1data/modelweights/Alkahest-V9.2-LLaMa-70B
• /media/administrator/oiseauxai1data/modelweights/Unhinged-RP-Alpha-V2-Llama-3.3-70B

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: /media/administrator/oiseauxai1data/modelweights/Forgotten-Abomination-70B-v5.0
    parameters:
      weight: 0.25
      density: 0.8
  - model: /media/administrator/oiseauxai1data/modelweights/Alkahest-V9.2-LLaMa-70B
    parameters:
      weight: 0.25
      density: 0.6
  - model: /media/administrator/oiseauxai1data/modelweights/Unhinged-RP-Alpha-V2-Llama-3.3-70B
    parameters:
      weight: 0.25
      density: 0.8
  - model: /media/administrator/oiseauxai1data/modelweights/Llama-3.1-70B-FLDx2
    parameters:
      weight: 0.25
      density: 0.4
base_model: /media/administrator/oiseauxai1data/modelweights/Forgotten-Abomination-70B-v5.0
merge_method: dare_ties
parameters:
  normalize: false
  int8_mask: true
tokenizer:
  source: base
chat_template: llama3
dtype: bfloat16
name: Darker_Merge_test