List your company

Question Answering on NaturalQuestions (val)

5.7Exact MatchDeepSeekMoE
1.2282.3893.554.711Jan 11, 2024

Evaluation results

MethodLinks
DeepSeekMoE# Shot=5-shot, # Total Params=2.0B, # Activated Params=0.3B, FLOPs per 2K Tokens=4.3T, # Training Tokens=100B2024.015.7
GShard# Shot=5-shot, # Total Params=2.0B, # Activated Params=0.3B, FLOPs per 2K Tokens=4.3T, # Training Tokens=100B2024.013.2
Switch# Shot=5-shot, # Total Params=2.0B, # Activated Params=0.2B, FLOPs per 2K Tokens=2.9T, # Training Tokens=100B2024.012.5
Dense# Shot=5-shot, # Total Params=0.2B, # Activated Params=0.2B, FLOPs per 2K Tokens=2.9T, # Training Tokens=100B2024.011.4
Hash Layer# Shot=5-shot, # Total Params=2.0B, # Activated Params=0.2B, FLOPs per 2K Tokens=2.9T, # Training Tokens=100B2024.011.4