Model: openai/gpt-4

Phenotype Balanced Accuracy Precision Sample Size
Aerophilicity 0.761 0.508 5551
Animal pathogenicity 0.752 0.892 4120
Biofilm formation 0.619 0.774 507
Biosafety level 0.842 0.603 7473
Cell shape 0.738 0.601 6626
Extreme environment tolerance 0.719 0.442 6739
Gram staining 0.812 0.646 3878
Health association 0.633 0.421 23
Hemolysis 0.512 0.342 239
Host association 0.787 0.840 5357
Motility 0.839 0.832 5949
Plant pathogenicity 0.739 0.202 7507
Spore formation 0.922 0.962 7276