OPT*

The 'OPT' dataset/benchmark is used to evaluate large language models (LLMs) by assessing their performance on perplexity and zero-shot tasks.

Papers using OPT* (3)

EPTS: Elastic Post-Training Sparsity for Efficient Large Language Model Compression2026TRIM: Achieving Extreme Sparsity with Targeted Row-wise Iterative Metric-driven Pruning2025Maximum Redundancy Pruning: A Principle-Driven Layerwise Sparsity Allocation for LLMs2025