cerebras / Llama-3-CBHybridL-8B
Text Generation8B1380Updated Mar 26, 2025
Downloads
138
Likes
0
Last updated
Mar 26, 2025
License
other
We are excited to release the Cerebras hybrid dense/sparse attention versions of Llama-3.1-8B-Instruct models optimized for long-context performance. This series includes two models: Llama3.1-CBHybridL-8B (model with 25 sparse attention layers out of 32) and Llama3.1-CBHybridM…
Tasks & library
- text generation
Tags
- cbllama
- cerebras
- sparse-attention
- llama-3
- pytorch
- text-generation
Technical specifications
Parameters8B
LibraryNot available
Languagesen
Licenseother