moe-paper-models
github.com/adamg012/moe-paper-models ↗A sumary of MoE experimental setups across a number of different papers.
16
GitHub Stars
19
Curated Resources
3
Categories
20 hours ago
Last Refreshed
Model Sizes of Paper ImplementationsExperimental Setups of Baselines and HardwareDatasets, Citations and Open Source
Use this list with your AI agent
Add the Context Awesome MCP server to Claude, Cursor, or any MCP client, then ask:
"Show me datasets, citations and open source resources from moe-paper-models"
Installation instructions →What's inside
Datasets, Citations and Open Source
- BASE Layers
RoBERTa corpus and CC100
- Deepspeed-MoE
Lambada/PIQA/BoolQ/RACE-h/Trivia-QA/WebQS
- Evo MoE
WMT(MT)/OpenWebText(LM MLM)/Wikipedia/OpenWebText
- Expert Choice Routing
GLaM
- FasterMoE
Wiki Text
- Gating Dropout
WMT/Web-50
Showing a sample of 19 resources. View the full list on GitHub →