YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
GDN-2 370M CPT — Vanilla GDN-2 LongData
Continued-pretraining checkpoint on 105M tokens (400 steps x effective batch 64 x 4096 tokens). Part of the unified Long-GDN CPT comparison series (2026-08-26).
- Architecture: GDN-2 (Gated DeltaNet v2) 370M
- Variant: Vanilla GDN-2 LongData
- Artifacts:
checkpoint-final.pth,training_history.jsonl
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support