Spaces:
Running
Running
Make Nanbeige architecture map discoverable in Space
#6
by alphabell-research - opened
README.md
CHANGED
|
@@ -7,7 +7,7 @@ sdk: static
|
|
| 7 |
app_file: index.html
|
| 8 |
pinned: true
|
| 9 |
license: mit
|
| 10 |
-
short_description: "
|
| 11 |
thumbnail: https://huggingface.co/spaces/alphabell/Model-Observatory/resolve/main/cover.png
|
| 12 |
models:
|
| 13 |
- RecursiveSI/flint
|
|
@@ -22,6 +22,8 @@ tags:
|
|
| 22 |
- recurrent-memory
|
| 23 |
- qwen
|
| 24 |
- flint
|
|
|
|
|
|
|
| 25 |
- educational
|
| 26 |
---
|
| 27 |
|
|
@@ -31,10 +33,14 @@ tags:
|
|
| 31 |
|
| 32 |
Explore two different architectures with real Q8_0 tensor captures: Flint's sparse expert routing and alphaBell Qwen Open's recurrent memory. Open a block, head or expert, then step through the tokens that used it.
|
| 33 |
|
| 34 |
-
[**Open the explorer**](https://huggingface.co/spaces/alphabell/Model-Observatory) · [Qwen view](https://alphabell-model-observatory.static.hf.space/?model=qwen) · [Source](https://huggingface.co/spaces/alphabell/Model-Observatory/tree/main) · [alphaBell](https://alphabell.com)
|
| 35 |
|
| 36 |

|
| 37 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 38 |
## Five measured experiments
|
| 39 |
|
| 40 |
[**Open all five experiments**](https://alphabell-model-observatory.static.hf.space/experiments/index.html)
|
|
|
|
| 7 |
app_file: index.html
|
| 8 |
pinned: true
|
| 9 |
license: mit
|
| 10 |
+
short_description: "Flint/Qwen traces and Nanbeige/Qwen architecture"
|
| 11 |
thumbnail: https://huggingface.co/spaces/alphabell/Model-Observatory/resolve/main/cover.png
|
| 12 |
models:
|
| 13 |
- RecursiveSI/flint
|
|
|
|
| 22 |
- recurrent-memory
|
| 23 |
- qwen
|
| 24 |
- flint
|
| 25 |
+
- nanbeige
|
| 26 |
+
- architecture
|
| 27 |
- educational
|
| 28 |
---
|
| 29 |
|
|
|
|
| 33 |
|
| 34 |
Explore two different architectures with real Q8_0 tensor captures: Flint's sparse expert routing and alphaBell Qwen Open's recurrent memory. Open a block, head or expert, then step through the tokens that used it.
|
| 35 |
|
| 36 |
+
[**Open the explorer**](https://huggingface.co/spaces/alphabell/Model-Observatory) · [Nanbeige and Qwen architecture](https://alphabell-model-observatory.static.hf.space/architecture/index.html) · [Qwen view](https://alphabell-model-observatory.static.hf.space/?model=qwen) · [Source](https://huggingface.co/spaces/alphabell/Model-Observatory/tree/main) · [alphaBell](https://alphabell.com)
|
| 37 |
|
| 38 |

|
| 39 |
|
| 40 |
+
## Nanbeige and Qwen architecture
|
| 41 |
+
|
| 42 |
+
[**Explore the layer map**](https://alphabell-model-observatory.static.hf.space/architecture/index.html) to compare Nanbeige4.3-33B's 52 full-attention layers with the Qwen3.8-27B text decoder's 48 recurrent linear and 16 full-attention layers. This comparison uses published configurations; the measured captures in this Space are Flint and Qwen.
|
| 43 |
+
|
| 44 |
## Five measured experiments
|
| 45 |
|
| 46 |
[**Open all five experiments**](https://alphabell-model-observatory.static.hf.space/experiments/index.html)
|