INDEX

Explanations

words associated with violence and exploitation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

AnchorStyles

-0.54

arResult

-0.44

 StatelessWidget

-0.43

ymce

-0.43

findpost

-0.43

URLException

-0.43

LookAnd

-0.42

noční

-0.42

 EdgeInsets

-0.42

 cerrada

-0.42

POSITIVE LOGITS

rungsseite

0.48

 astore

0.47

Viitteet

0.46

arakhand

0.45

 ObjectParameter

0.43

揄

0.42

Rujuakan

0.40

*/].

0.39

wapV

0.37

KommentareTeilen

0.36

Activations Density 1.707%

words associated with violence and exploitation

No Comments

No Known Activations