INDEX

Explanations

punctuation and metadata tags

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Према

-0.58

RUnlock

-0.57

dificio

-0.56

 reprises

-0.54

 âgées

-0.53

)?;

-0.52

icleta

-0.52

SqlClient

-0.50

zitter

-0.50

 eventdata

-0.50

POSITIVE LOGITS

RTLD

0.69

ismet

0.66

RELATED

0.62

AndEndTag

0.62

GEBURTSDATUM

0.59

Datuak

0.59

ElementException

0.58

Uli

0.55

govina

0.54

contentLoaded

0.54

Activations Density 0.003%

punctuation and metadata tags

No Comments

No Known Activations

punctuation and metadata tags

No Comments

No Known Activations