Skip to content

update main README with Bamba v2 numbers #37

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
lchu6 wants to merge 1 commit into
base: main
Choose a base branch
from
Open

update main README with Bamba v2 numbers #37

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
85 changes: 32 additions & 53 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,11 +25,9 @@ issue with your env:

```bash
git clone https://github.com/Dao-AILab/causal-conv1d.git
cd causal-conv1d && pip install . && cd ..
cd causal-conv1d && FORCE_BUILD=TRUE pip install . && cd ..
git clone https://github.com/state-spaces/mamba.git
cd mamba && pip install . && cd ..
git clone https://github.com/Dao-AILab/flash-attention.git
cd flash-attention && pip install . && cd ..
cd mamba && MAMBA_FORCE_BUILD=TRUE pip install . && cd ..
```

For users using our HF versions of the model, you would need to install the latest transformers which includes our newly merged implementation for our Bamba models:
Expand Down Expand Up @@ -72,123 +70,104 @@ For exact reproduction of Bamba 9.8B using the same training data, access is ava
</td>
<td><strong>Benchmark</strong>
</td>
<td><strong>Bamba 9B (2.2T)</strong>
<td><strong>Bamba 9B (3.1T)</strong>
</td>
</tr>
<tr>
<td rowspan="8" >General
<td rowspan="9" >General
</td>
<td>MMLU (5-shot)
<td>MMLU
</td>
<td>60.77
<td>67.92
</td>
</tr>
<tr>
<td>ARC-C (25-shot)
<td>ARC-C
</td>
<td>63.23
<td>63.57
</td>
</tr>
<tr>
<td>GSM8K (5-shot)
<td>GSM8K
</td>
<td>36.77
<td>41.70
</td>
</tr>
<tr>
<td>Hellaswag (10-shot)
<td>Hellaswag
</td>
<td>81.8
<td>83.85
</td>
</tr>
<tr>
<td>OpenbookQA (5-shot)
<td>OpenbookQA
</td>
<td>47.6
<td>51.0
</td>
</tr>
<tr>
<td>Piqa (5-shot)
</td>
<td>82.26
<td>83.62
</td>
</tr>
<tr>
<td>TruthfulQA (0-shot)
</td>
<td>49.21
<td>50.86
</td>
</tr>
<tr>
<td>Winogrande (5-shot)
</td>
<td>76.87
<td>79.48
</td>
</tr>
<tr>
<td rowspan="6" >HF OpenLLM- V2*
</td>
<td>MMLU-PRO (5-shot)
<td>Boolq
</td>
<td>17.53
<td>82.78
</td>
</tr>
<tr>
<td>BBH (3-shot)
</td>
<td>17.4
</td>
</tr>
<tr>
<td>GPQA (0-shot)
</td>
<td>4.14
<td rowspan="6" >HF OpenLLM- V2*
</td>
</tr>
<tr>
<td>IFEval (0-shot)
<td>MMLU-PRO
</td>
<td>15.16
<td>25.41
</td>
</tr>
<tr>
<td>MATH Lvl 5 (4-shot)
<td>BBH
</td>
<td>1.66
<td>24.78
</td>
</tr>
<tr>
<td>MuSR (0-shot)
<td>GPQA
</td>
<td>9.59
<td>5.93
</td>
</tr>
<tr>
<td rowspan="4" >Safety Tasks
</td>
<td>PopQA (5-shot)
<td>IFEval
</td>
<td>20.5
<td>19.00
</td>
</tr>
<tr>
<td>Toxigen (5-shot)
<td>MATH Lvl 5
</td>
<td>57.4
<td>6.42
</td>
</tr>
<tr>
<td>BBQ (5-shot)
<td>MuSR
</td>
<td>44.2
<td>9.28
</td>
</tr>
<tr>
<td>Crows-pairs english (5-shot)
</td>
<td>70.78
</td>
</tr>
</table>

*For the v2 leaderboard results, we perform [normalization](https://huggingface.co/docs/leaderboards/open_llm_leaderboard/normalization) and report the normalized results.
Expand Down