Update eval_gauntlet_callback.py with math.log2 (#821)

Saw an automated ruff flag this, seems like a strict improvement and is marginally faster.

Co-authored-by: Daniel King <[email protected]>

Author: 2 people

llmfoundry/callbacks/eval_gauntlet_callback.py

@@ -117,8 +117,7 @@ def __init__(self,
                 elif self.weighting == Weighting.SAMPLE_SZ:
                     weight = cumulative_samples
                 elif self.weighting == Weighting.LOG_SAMPLE_SZ:
-                    weight = max(math.log(cumulative_samples, 2), 1)
-
+                    weight = max(math.log2(cumulative_samples), 1)
                 assert weight is not None
                 benchmark['weighting'] = weight