const-test / README.md
Sakalti's picture
Upload folder using huggingface_hub
b27cfc3 verified
metadata
base_model:
  - Sakalti/SJT-0.5B
library_name: transformers
tags:
  - mergekit
  - merge

merge

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

Merge Details

Merge Method

This model was merged using the passthrough merge method.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:


const_tag: &scale_factor 1.414213562

attenuate-env: &attenuated_env
  parameters:
    scale:
      - filter: q_proj
        value: *scale_factor
      - filter: k_proj
        value: *scale_factor
      - value: 1.0

slices:
  - sources:
      - model: Sakalti/SJT-0.5B
        layer_range: [0, 6]  
  - sources:
      - model: Sakalti/SJT-0.5B
        layer_range: [6, 12]

  - sources:
      - model: Sakalti/SJT-0.5B
        layer_range: [12, 18]  
        <<: *attenuated_env

  - sources:
      - model: Sakalti/SJT-0.5B
        layer_range: [18, 24]  

merge_method: passthrough
dtype: bfloat16
normalize: true
int8_mask: true