INDEX

Explanations

numerical values and related contextual occurrences

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

DrawerToggle

-0.34

ver

-0.30

timeScale

-0.29

er

-0.29

pra

-0.29

ゆ

-0.27

log

-0.27

and

-0.27

PullParser

-0.26

hal

-0.26

POSITIVE LOGITS

脚注の使い方

0.73

\{\\

0.70

OGND

0.67

ſelben

0.67

 noDo

0.66

 مشين

0.65

Rohy

0.65

<unused79>

0.64

<unused14>

0.64

<unused51>

0.64

Activations Density 0.034%

numerical values and related contextual occurrences

No Comments

No Known Activations