Upload README.md with huggingface_hub

README.md

@@ -0,0 +1,20 @@
+---
+license: mit
+library_name: pytorch
+tags:
+- coconut
+- latent-chain-of-thought
+- sparse-autoencoder
+- qwen3
+---
+
+# SAE on Coconut latent thoughts (Qwen3-1.7B)
+
+SAE bottleneck trained on the **frozen** Stage-3 Coconut latent thoughts of Qwen3-1.7B,
+from the *VQ-CoT: Discretising Latent Chain-of-Thought* project (team RateLimit Achieved, EPFL CS-552).
+The language model is frozen; only this bottleneck is trained on the dumped latents
+(385,620 x 6 latent thought vectors, GSM8K).
+
+- Checkpoint: `sae.pt`
+- TopK SAE, dictionary 8x hidden, k=32 active features
+- Result: inserting this bottleneck into the frozen latent loop costs ~0 GSM8K test accuracy.