INDEX

Explanations

geographical directions and positioning

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

Directional terms: north, south, northwards

np_acts-logits-general · gemini-2.0-flash

north of, south of

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

ParallelGroup

-0.81

 North

-0.77

 mergeFrom

-0.74

 South

-0.68

 nahilalakip

-0.67

North

-0.65

glEnd

-0.65

OGND

-0.64

 Roskov

-0.61

 Southern

-0.60

POSITIVE LOGITS

wards

0.91

wester

0.58

ward

0.55

most

0.53

bound

0.53

 initComponents

0.53

 coast

0.52

 bound

0.51

 bank

0.50

ariato

0.49

Activations Density 0.106%

geographical directions and positioning

Directional terms: north, south, northwards

north of, south of

No Comments

No Known Activations