Update README.md
Browse files
README.md
CHANGED
|
@@ -2,14 +2,17 @@
|
|
| 2 |
license: mit
|
| 3 |
---
|
| 4 |
|
| 5 |
-
#
|
| 6 |
|
| 7 |
-
|
|
|
|
|
|
|
| 8 |
|
| 9 |
-
|
| 10 |
|
| 11 |
The base Walrus model is a 1.3B-parameter space–time Transformer trained on 19 physical simulation datasets. See the main [Walrus repo](https://github.com/PolymathicAI/walrus) for full details on the architecture and pretraining.
|
| 12 |
|
|
|
|
| 13 |
## Usage
|
| 14 |
|
| 15 |
Download the checkpoint and point your config at it:
|
|
|
|
| 2 |
license: mit
|
| 3 |
---
|
| 4 |
|
| 5 |
+
# Physics Steering: Causal Control of Cross-Domain Concepts in a Physics Foundation Model
|
| 6 |
|
| 7 |
+
[](https://opensource.org/licenses/MIT)
|
| 8 |
+
[](https://github.com/DJ-Fear/walrus_steering)
|
| 9 |
+
[](https://arxiv.org/abs/2511.20798)
|
| 10 |
|
| 11 |
+
This checkpoint contains weights for the [Walrus](https://huggingface.co/polymathic-ai/walrus) physics foundation model, fine-tuned and used for activation steering experiments described in *Physics Steering: Causal Control of Cross-Domain Concepts in a Physics Foundation Model* — [arXiv:2511.20798v2](https://arxiv.org/abs/2511.20798).
|
| 12 |
|
| 13 |
The base Walrus model is a 1.3B-parameter space–time Transformer trained on 19 physical simulation datasets. See the main [Walrus repo](https://github.com/PolymathicAI/walrus) for full details on the architecture and pretraining.
|
| 14 |
|
| 15 |
+
|
| 16 |
## Usage
|
| 17 |
|
| 18 |
Download the checkpoint and point your config at it:
|