Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Completion Length Prediction on MMLU-Pro (eval)
Loading...
117.19
MAE
Completion Length Probe
113.6848
137.3449
161.005
184.6651
Jul 6, 2026
MAE
Updated 19d ago
Evaluation Results
Method
Method
Links
MAE
Completion Length Probe
Model=Llama-3.1-8B
2026.07
117.19
Completion Length Probe
Model=Mistral-7B
2026.07
131.51
Completion Length Probe
Model=Olmo-3-7B
2026.07
134.86
Statistical baseline
Model=Mistral-7B
2026.07
196.16
Statistical baseline
Model=Llama-3.1-8B
2026.07
204.21
Statistical baseline
Model=Olmo-3-7B
2026.07
204.82
Feedback
Search any
task
Search any
task