README.md

metadata

base_model: []
library_name: transformers
tags:
  - mergekit
  - merge

Noodlz_DolphLundgren

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 D:\AI\mergekit\Noodlz\Noodlz_DolphinLake-DARE_TIE_SLERP-tokenwest as a base.

Models Merged

The following models were included in the merge:

• D:\AI\mergekit\Noodlz\Noodlz_HermeStar-DARE_TIE_SLERP-startoken

Configuration

The following YAML configuration was used to produce this model:

merge_method: dare_ties

parameters:
  int8_mask: true
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5 # fallback for rest of tensors
  embed_slerp: true

# Merging DolphinLake and HermeStar. 4 models with 2xChatML and 2 others
models:
  - model: D:\AI\mergekit\Noodlz\Noodlz_DolphinLake-DARE_TIE_SLERP-tokenwest
    # No parameters necessary for base model
  - model: D:\AI\mergekit\Noodlz\Noodlz_HermeStar-DARE_TIE_SLERP-startoken
    parameters:
      density: 0.58
      weight: 0.7


base_model: D:\AI\mergekit\Noodlz\Noodlz_DolphinLake-DARE_TIE_SLERP-tokenwest
tokenizer_source: model:D:\AI\mergekit\Noodlz\Noodlz_HermeStar-DARE_TIE_SLERP-startoken

dtype: bfloat16