INDEX

Explanations

mentions of people, their occupations, and their attributes

oai_token-act-pair · gemini-2.0-flash

professions

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

FieldBuilder

-0.57

 Theſe

-0.56

 Reſ

-0.55

 GreatSchools

-0.53

KURZBESCHREIBUNG

-0.52

 publicités

-0.50

전히

-0.50

 disambiguazione

-0.50

audiovisuel

-0.50

 pleaſure

-0.49

POSITIVE LOGITS

who

0.90

is

0.75

was

0.72

did

0.63

 uses

0.60

 does

0.60

has

0.60

who

0.59

 removes

0.59

 whom

0.58

Activations Density 3.467%

mentions of people, their occupations, and their attributes

professions

No Comments

No Known Activations