MCPcopy Index your code
hub / github.com/Irenehere/Auto-Debias

github.com/Irenehere/Auto-Debias @main

Chat with this repo
repository ↗ · DeepWiki ↗ · + Follow
15 symbols 61 edges 3 files 0 documented · 0%
What it actually does AI analysis from the code graph — generated when you open this
loading…
README

Auto-Debias

requirments

- numpy
- torch
- transformers

related files

We release the debiased model here.

generated biased prompt via beam search

python generate_prompts.py \
--debias_type   gender or race 
--model_type   bert or roberta or albert
--model_name_or_path  bert-base-uncased, etc

debiasing language models

python auto-debias.py
--debias_type    gender or race 
--model_type   bert or roberta or albert
--model_name_or_path  bert-base-uncased, etc
--prompts_file prompts_bert-base-uncased_gender

evaluation

### SEAT We run the SEAT using the code from Liang et al.

### GLUE We run GLUE from transformers.

Core symbols most depended-on inside this repo

load_word_list
called by 7
utils.py
get_tokenized_prompt
called by 2
auto-debias.py
send_to_cuda
called by 2
auto-debias.py
clean_word_list2
called by 2
utils.py
clean_word_list
called by 2
utils.py
run_model
called by 2
generate_prompts.py
get_prompt_jsd
called by 2
generate_prompts.py
load_wiki_word_list
called by 1
utils.py

Shape

Function 12
Method 2
Class 1

Languages

Python100%

Modules by API surface

utils.py8 symbols
generate_prompts.py5 symbols
auto-debias.py2 symbols

For agents

$ claude mcp add Auto-Debias \
  -- python -m otcore.mcp_server <graph>

⬇ download graph artifact