INDEX

Explanations

references to specific locations or landmarks within a geographical context

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

historical places, wounds, objects

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.24.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

addPreferredGap

-0.36

},{

-0.36

spreis

-0.35

phase

-0.35

ต่อ

-0.35

sendData

-0.34

æa

-0.34

pecially

-0.34

AJ

-0.33

Kis

-0.33

POSITIVE LOGITS

fjspx

0.56

 surla

0.53

featureID

0.49

énario

0.49

 itſelf

0.48

 OFDb

0.46

NameInMap

0.46

 themſelves

0.46

ChildScrollView

0.45

DJANGO

0.45

Activations Density 0.075%

references to specific locations or landmarks within a geographical context

historical places, wounds, objects

No Comments

No Known Activations