Revisiting Funnel Transformers for Modern LLM Architectures with Comprehensive Ablations in Training and Inference Configurations

Transformer-based Large Language Models, which suffer from high computational costs, advance so quickly that techniques proposed to streamline earlier iterations are not guaranteed to benefit more modern models. Building upon the Funnel Transformer proposed by Dai and Le (2020), which progressively compresses intermediate representations, we investigate the impact of funneling in contemporary Gemma2 Transformer architectures. We systematically evaluate various funnel configurations and recovery methods, comparing: (1) standard pretraining to funnel-aware pretraining strategies, (2) the impact of funnel-aware fine-tuning, and (3) the type of sequence recovery operation. Our results demonstrate that funneling creates information bottlenecks that propagate through deeper network layers, particularly in larger models (e.g., Gemma 7B), leading to at times unmanageable performance lost. However, carefully selecting the funneling layer and employing effective recovery strategies, can substantially mitigate performance losses, achieving up to a 44% reduction in latency. Our findings highlight key trade-offs between computational efficiency and model accuracy, providing practical guidance for deploying funnel-based approaches in large-scale natural language applications.