Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Downstream Evaluation on LM-Eval Harness (ARC-E, BoolQ, HellaSwag, OBQA, SciQ)

0.381ARC-E Accuracy

DDP

0.374760.376380.3780.37962Jul 11, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.07
0.3810.610.4090.3440.4960.448
2025.07
0.3810.6210.4150.320.4960.447
2025.07
0.380.6210.4150.3640.5070.457
2025.07
0.3750.60.4190.3440.4910.446