INDEX

Explanations

references to evaluation criteria and guidelines

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

elan

-0.07

nech

-0.06

atio

-0.06

reste

-0.06

Ã¥n

-0.06

IPA

-0.05

iÃ¨re

-0.05

rew

-0.05

/lg

-0.05

lÃ¤

-0.05

POSITIVE LOGITS

 criteria

0.19

criteria

0.17

 Criteria

0.16

Criteria

0.16

 criterion

0.15

 criter

0.13

Criterion

0.12

 Criterion

0.12

.criteria

0.12

_criteria

0.12

Activations Density 0.087%