INDEX

Explanations

decisions and uncertainty

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

è§Ħå¾ĭ

-0.06

æĸĽ

-0.06

æłĭæ¢ģ

-0.06

 PREFIX

-0.06

çĽĪ

-0.06

ense

-0.06

æ£į

-0.05

 Ø¹ÙĨØ§

-0.05

_PREFIX

-0.05

æĬ±

-0.05

POSITIVE LOGITS

Ney

0.10

 Decision

0.09

_LIBRARY

0.08

_FC

0.08

ibas

0.08

ostar

0.08

Decision

0.08

Ø§Ø±Ø§

0.08

 decision

0.08

Ð¶Ð»Ð¸

0.08

Activations Density 0.002%