Files
MatBench/layer1/ALL-merge/eval-claude-3-7.log
2025-05-28 10:55:34 +08:00

11 lines
278 B
Plaintext

Accuracy of claude-3-7:
{
'accuracy': 0.9139927224611313,
'precision_micro': 0.9048991354466859,
'recall_micro': 0.9348329474032419,
'f1_micro': 0.919622518711357,
'precision_macro': 0.8426824027456492,
'recall_macro': 0.9350777308265648,
'f1_macro': 0.884855893435638
}