Jebadiah/Aria-dolphin-stone

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 Jebadiah/Aria-7b-128k-v4 as a base.

Models Merged

The following models were included in the merge:

• cognitivecomputations/dolphin-2.9-llama3-8b-1m

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: Jebadiah/Aria-7b-128k-v4
    # No parameters necessary for base model
  - model: cognitivecomputations/dolphin-2.9-llama3-8b-1m
    parameters:
      density: 0.1
      weight: 0.11
merge_method: dare_ties
base_model: Jebadiah/Aria-7b-128k-v4
parameters:
  int8_mask: true
dtype: bfloat16