medrag-multi-modal Workspace – Weights & Biases

Skip to main content

Assets

MMLU-Anatomy-BM25s:v4

Name

MMLU-Anatomy-BM25s(5 versions)

Last updated

11 months ago

Storage size

923B (4.6kB from all versions)

mmlu-anatomy-test:v2

MMLUOptionAccuracy:v6

Summary

1

Model

2

correct.true_count

3

correct.false_count

4

correct.true_fraction

5

correct.false_fraction

6

correct.stderr

7

correct.precision

8

correct.recall

9

correct.f1_score

10

Avg. Latency

11

Run Date

12

Trials

13

MedQAAssistant:v45

117.00

18.00

86.67%

13.33%

2.93%

86.67%

1.00

92.86%

2.70

11 months ago

1.00

MedQAAssistant:v40

98.00

37.00

73.13%

26.87%

3.83%

73.13%

98.99%

84.12%

2.00

11 months ago

1.00

Total Rows: 2