INDEX

Explanations

fragments of text from online forums, blogs, and comments, particularly in casual conversation contexts.

phrases indicating engagement or strong emotional responses

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

Negative Logits

var

-0.06

dot

-0.06

 area

-0.06

 form

-0.06

ya

-0.05

 Shannon

-0.05

ĩ

-0.05

POSITIVE LOGITS

AtPath

0.08

ono

0.08

Æ¡

0.08

uis

0.08

ãĤ¤ãĥ³ãĥĪ

0.08

klu

0.08

imson

0.08

_ports

0.08

 erken

0.08

ilden

0.08

Activations Density 0.004%