id-llama3-8b-cpt-sahabatai-v1-instruct-x-meta-llama-3-8b-instruct-ffn_kv_injection

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

Merge Details

Merge Method

This model was merged using the FFN-KV Injection (Train-free FFN gating) merge method using meta-llama/Meta-Llama-3-8B-Instruct as a base.

Models Merged

The following models were included in the merge:

• GoToCompany/llama3-8b-cpt-sahabatai-v1-instruct

Configuration

The following YAML configuration was used to produce this model:

dtype: bfloat16
tokenizer:
  source: union
merge_method: ffn_kv_injection
base_model: meta-llama/Meta-Llama-3-8B-Instruct
models:
- model: meta-llama/Meta-Llama-3-8B-Instruct
  parameters: {}
- model: GoToCompany/llama3-8b-cpt-sahabatai-v1-instruct
  parameters: {}
parameters:
  weights:
  - 0.7
  - 0.3
  tau: 1.0
  aspect_thresh: 1.5
  pnorm: 2.0
  alpha_floor: 0.0
  alpha_ceil: 1.0
write_readme: README.md