INDEX

Explanations

references to violence and destruction

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 ModelExpression

-0.46

AndEndTag

-0.42

PMailer

-0.40

kheim

-0.40

 spirito

-0.38

BeNil

-0.37

 Ausnahme

-0.37

HasColumnType

-0.36

pstmt

-0.36

werb

-0.36

POSITIVE LOGITS

IsContent

0.45

 незавершена

0.44

ontan

0.41

glTexCoord

0.40

════

0.40

dze

0.39

Inclusive

0.38

Carriera

0.38

ftance

0.37

TProtocol

0.36

Activations Density 0.084%

references to violence and destruction

No Comments

No Known Activations