Casual-Autopsy/Llama-3-Yollow-SCE

merge2

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

Merge Details

Merge Method

This model was merged using the SCE merge method using /kaggle/input/meta-llama-3-8b/transformers/hf/1 as a base.

Models Merged

The following models were included in the merge:

• /kaggle/input/llama-3-youko-8b/transformers/hf/1
• /kaggle/input/llama-3-swallow-8b-v0.1/transformers/hf/1
• /kaggle/input/meta-llama-3-8b-instruct/transformers/hf/1

Configuration

The following YAML configuration was used to produce this model:

models:
  # Pivot model
  - model: /kaggle/input/meta-llama-3-8b/transformers/hf/1
  # Target models
  - model: /kaggle/input/meta-llama-3-8b-instruct/transformers/hf/1
  - model: /kaggle/input/llama-3-youko-8b/transformers/hf/1
  - model: /kaggle/input/llama-3-swallow-8b-v0.1/transformers/hf/1
merge_method: sce
base_model: /kaggle/input/meta-llama-3-8b/transformers/hf/1
parameters:
  select_topk: 0.65
dtype: bfloat16