INDEX

Explanations

instances of dialogue between multiple people

oai_token-act-pair · gemini-2.0-flash

spoken conversations

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

IntoConstraints

-0.69

 passés

-0.59

Personensuche

-0.57

extAlignment

-0.57

 تضيفلها

-0.56

 continúas

-0.56

 مرئيه

-0.56

 Throwable

-0.56

itures

-0.54

 odkazy

-0.54

POSITIVE LOGITS

MockBean

0.50

frontmatter

0.47

 synes

0.46

ритори

0.46

ultu

0.45

overline

0.44

 initState

0.43

uruhan

0.43

AspectRatio

0.42

nola

0.42

Activations Density 1.837%

instances of dialogue between multiple people

spoken conversations

No Comments

No Known Activations