Spaces:
Running
Running
publish from LatentPortPort 7bfa698
Browse files
README.md
CHANGED
|
@@ -82,6 +82,18 @@ Qwen3-4B reads; Qwen3-8B answers, across all 36 layers.
|
|
| 82 |
| π― **Held-out questions** | **7 of 7**, the same as each model alone |
|
| 83 |
| β‘ **Time to first token** | **1.12Γ** sooner (95% CI 1.12β1.12) on two RTX 3090s at ~256 tokens; ceiling 1.61Γ |
|
| 84 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 85 |
### πͺ A 13.6Γ gap, at the bar β [`qwen3-0.6b-to-8b-bridge`](https://huggingface.co/agentic-underground/qwen3-0.6b-to-8b-bridge)
|
| 86 |
|
| 87 |
Qwen3-0.6B reads; Qwen3-8B, 13.6 times larger, answers.
|
|
|
|
| 82 |
| π― **Held-out questions** | **7 of 7**, the same as each model alone |
|
| 83 |
| β‘ **Time to first token** | **1.12Γ** sooner (95% CI 1.12β1.12) on two RTX 3090s at ~256 tokens; ceiling 1.61Γ |
|
| 84 |
|
| 85 |
+
### π Fifty-four times larger β [`qwen3-0.6b-to-32b-bridge`](https://huggingface.co/agentic-underground/qwen3-0.6b-to-32b-bridge)
|
| 86 |
+
|
| 87 |
+
Qwen3-0.6B reads; Qwen3-32B, quantized to Q4K, answers. Published with its miss stated, while a
|
| 88 |
+
refit is under way.
|
| 89 |
+
|
| 90 |
+
| | |
|
| 91 |
+
|:--|:--|
|
| 92 |
+
| β‘ **Time to first token** | **4.21Γ** sooner (95% CI 4.19β4.23) at ~256 tokens, **7.17Γ** at ~1K, on two RTX 3090s; ceilings 11.60Γ and 14.33Γ |
|
| 93 |
+
| π― **Held-out questions** | **6 of 7** β the 32B alone gets 7, so it is **below our answer bar** |
|
| 94 |
+
| π **Fidelity (NCR)** | **0.866** with the paper's learned correction (0.794 from the maps alone) |
|
| 95 |
+
| π§ͺ **Control (wrong donor)** | 0.197 |
|
| 96 |
+
|
| 97 |
### πͺ A 13.6Γ gap, at the bar β [`qwen3-0.6b-to-8b-bridge`](https://huggingface.co/agentic-underground/qwen3-0.6b-to-8b-bridge)
|
| 98 |
|
| 99 |
Qwen3-0.6B reads; Qwen3-8B, 13.6 times larger, answers.
|