ja-llama-3-swallow-8b-instruct-v0.1-x-meta-llama-3-8b-instruct-scope_merge

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

Merge Details

Merge Method

This model was merged using the SCOPE-Merge (Self-Consistent Orthogonal Projection) merge method using meta-llama/Meta-Llama-3-8B-Instruct as a base.

Models Merged

The following models were included in the merge:

• tokyotech-llm/Llama-3-Swallow-8B-Instruct-v0.1

Configuration

The following YAML configuration was used to produce this model:

dtype: bfloat16
tokenizer:
  source: union
merge_method: scope_merge
base_model: meta-llama/Meta-Llama-3-8B-Instruct
models:
- model: meta-llama/Meta-Llama-3-8B-Instruct
  parameters: {}
- model: tokyotech-llm/Llama-3-Swallow-8B-Instruct-v0.1
  parameters: {}
parameters:
  lambda_reg: 1.5
  k_fisher: 64
  project: false
  svd_cap: 4096
  tall_skip_ratio: 4
  cpu_svd: false
  rank_shrink: 0
  seed: 0
write_readme: README.md