Skip to content

Update README instructions #3

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
gkielian wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Update README instructions #3

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 18 additions & 2 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,6 +21,13 @@ traditional softmax hardware in 16nm FinFET technology.

---

## Publication Links

* [ICCAD 2024 Publication](paper/ICCAD_2024_ConSmax.pdf)
* Preprint Link: https://arxiv.org/abs/2402.10930

---

## Citation
If you find our code useful for your research, please consider citing:

Expand All @@ -43,14 +50,23 @@ If you find our code useful for your research, please consider citing:

### Software Evaluation

```

Assuming PyTorch GPU training and NanoGPT dependencies are installed on your
enviroment:

```sh
git clone https://github.com/ReaLLMASIC/nanogpt.git
cd nanogpt/

cd data/wikitext103
bash get_dataset.sh
python3 get_wikitext103.py
python3 prepare.py -t input.txt --method tiktoken

cd ../../

# Run with ConSmax
python3 train.py --softmax_variant_attn consmax_v2 --dataset wikitext103 --max_sample_tokens 256 --max_iters 30000 --use_post_ln

# Run Softmax for Reference
python3 train.py --dataset wikitext103 --max_sample_tokens 256 --max_iters 30000 --use_post_ln
```