Open-weight model
anlp-a2-moe_top2_same_active
by Rithvik iconically-mine/anlp-a2-moe_top2_same_active
anlp-a2-moe_top2_same_active is an open-weight model from Rithvik. Its published files total 65.5 MB.
ffntype: moetop2sameactive - dmodel=256, nlayers=6, nheads=8, ffnhiddendim=1024 Checkpoint: moetop2sameactivefinal.pt (load with torch.load, see src/part1/model/transformer.py).
Model Card
ffntype: moetop2sameactive - dmodel=256, nlayers=6, nheads=8, ffnhiddendim=1024 Checkpoint: moetop2sameactivefinal.pt (load with torch.load, see src/part1/model/transformer.py).
Excerpt from the card by Rithvik.
Identity and Version
- Repository
- iconically-mine/anlp-a2-moe_top2_same_active
- Publisher
- Rithvik
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- Not stated by the source
- Languages
- Not stated by the source
- Revision
- a02ed4ad5ac9e51cbfdfd1e9ba0caeb345cc9358
- First published
- 2026-10-04
- Last updated
- 2026-10-04
Files and Weights
3 files, 65.5 MB in total. The weights are 1 file totalling 65.5 MB in pt.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| moe_top2_same_active_final.pt | Weights | 65.5 MB | c32693488f2e |
| README.md | Documentation | 329 B | — |
| .gitattributes | Repository | 1.5 KB | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 65.5 MB
Released by Rithvik through its official repository on Hugging Face.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 65.5 MB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.