Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -66,3 +66,6 @@ then answer and stop."* roughly halves median output length at no retraining cos
|
|
| 66 |
|
| 67 |
---
|
| 68 |
*Compute sponsored by **[Lambda](https://lambda.ai)** — thank you. 🙏*
|
|
|
|
|
|
|
|
|
|
|
|
| 66 |
|
| 67 |
---
|
| 68 |
*Compute sponsored by **[Lambda](https://lambda.ai)** — thank you. 🙏*
|
| 69 |
+
|
| 70 |
+
## Honest note (n=2000)
|
| 71 |
+
The unpruned teacher loops on only **3.6%** of these prompts vs **~7-8%** for this pruned cut — REAP pruning roughly doubles the loop rate, and gate-only Router-KD (even on full data) does not close it. Earlier small-n evals suggesting parity were a sampling fluke. A knowledge-recovery LoRA is in progress to add capacity back.
|