INDEX

Explanations

symbols and values typically associated with numerical data or prices

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

$1^{}23$ or $1^{}2$

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

createCanvas

-0.46

Personendaten

-0.44

 تكبرها

-0.44

CloseOperation

-0.44

 cherchés

-0.43

ագրություններ

-0.42

 ostavi

-0.42

 CreateTagHelper

-0.42

Geplaatst

-0.41

 Cæsar

-0.41

POSITIVE LOGITS

Clos

0.42

featureID

0.40

__':

0.40

MENAFN

0.40

 asegur

0.39

 comigo

0.38

transQ

0.38

SharedCtor

0.38

Sp

0.38

்த்த

0.38

Activations Density 0.079%

symbols and values typically associated with numerical data or prices

$1^{}23$ or $1^{}2$

No Comments

No Known Activations

symbols and values typically associated with numerical data or prices

$1^{}23$ or $1^{}2$

No Comments

No Known Activations

symbols and values typically associated with numerical data or prices

$1^{*}23$ or $1^{*}2$

No Comments

No Known Activations

symbols and values typically associated with numerical data or prices

$1^{*}23$ or $1^{*}2$

No Comments

No Known Activations

$1^{}23$ or $1^{}2$

$1^{}23$ or $1^{}2$