MaziyarPanahi/Llama-3-70B-Instruct-32k-v0.1
MaziyarPanahi/Llama-3-70B-Instruct-32k-v0.1 is a 70 billion parameter instruction-tuned model based on the Llama-3 architecture. This experimental version features a modified `rope_theta` setting to `8m`, potentially influencing its context handling or performance. It is designed for instruction-following tasks, offering a large parameter count for complex language generation and understanding.
Loading preview...
Overview
This model, MaziyarPanahi/Llama-3-70B-Instruct-32k-v0.1, is an experimental instruction-tuned variant of the Llama-3 70 billion parameter model. Its primary distinction lies in the modification of the rope_theta parameter, which has been set to 8m. This adjustment is an experimental change that could impact the model's ability to handle longer contexts or its overall performance characteristics.
Key Capabilities
- Instruction Following: Designed to accurately follow and execute instructions provided in natural language.
- Large Scale: With 70 billion parameters, it is capable of complex language understanding and generation tasks.
- Experimental Configuration: Features a unique
rope_thetasetting (8m) for potential advancements in context management or other performance aspects.
Good For
- Developers interested in experimenting with modified Llama-3 architectures.
- Applications requiring a large, instruction-tuned language model.
- Research into the effects of
rope_thetamodifications on model behavior.
Quantized GGUF models for this version are also available for easier deployment and inference on various hardware configurations.
Top 3 parameter combinations used by Featherless users for this model. Click a tab to see each config.