Downloads · 30 days
0
anima-deep-convmoe-L8
dancinlab/anima-deep-convmoe-L8
anima-deep-convmoe-L8 is a machine learning model from dancinlab. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
Deep ConvMoE L=8 (RF-expanded, d=3784, 3 experts, SLW) trained on the 4-cell corpus (ko/en × general/sns · steps 2000 · seq-len 1024). Tests whether receptive-field expansion (L4→L8) cracks the G1 recombination wall.
Access
Public
Updated Jul 10, 2026
Repo size
379 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.clm379 MB · 100%
At a glance
- License
- apache-2.0
- Access
- Public
- Created
- Jul 10, 2026
- Updated
- Jul 10, 2026
- SHA
- 8fc7d2d7
Model card
From the Hugging Face model README
anima-deep-convmoe-L8 (H_1584 depth-RF probe · NEGATIVE result)
Deep ConvMoE L=8 (RF-expanded, d=3784, 3 experts, SLW) trained on the 4-cell corpus (ko/en × general/sns · steps 2000 · seq-len 1024). Tests whether receptive-field expansion (L4→L8) cracks the G1 recombination wall.
- Training: CLEAN — loss 5.636→1.552 DESCENT · registers_DESCENT 4/4 · val_CE(pooled) 1.71 ≪ uniform 5.545 · clm_decodable=True.
- G1 (ρ·weave) verdict: 🧱 FAIL — best_distinct=1 (need ≥2 & >max_single) — SAME floor as production L4.
- Conclusion: depth/RF expansion does NOT crack G1 recombination at engine-native 303M-class byte-LM (H_1584 FALSIFIED). Consistent with the DPI meta-law: an additive-solvable loss escapes to the additive floor regardless of capacity/RF. The numpy reachability probe (conv_L8 reach 1.47e-3 REACHABLE) proved information flow exists but the trained model does not learn to use it for recombination.
sha256: 7d221ec81cee1543de012f01ba2d060bd1f304bd5ed9fab865f27c3b2e5af178
- License
- apache-2.0
- Created
- Jul 10, 2026
- Updated
- Jul 10, 2026