VersatileFFN: Achieving Parameter Efficiency in LLMs via Adaptive Wide-and-Deep Reuse
Open paper
Benchmark
Model
Rank
Results
common-sense-reasoning-on-arc-easy
VersatileFFN
#44
Accuracy: 73.16
question-answering-on-piqa
VersatileFFN
#117
Accuracy: 74.65
sentence-completion-on-hellaswag
VersatileFFN
#150
Accuracy: 60.17
Rank counts only results with a code link.