INDEX

Explanations

references to neighborhoods

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

neighborhood

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

 Irán

-0.51

Clik

-0.46

 Milán

-0.39

Datei

-0.39

 Sosten

-0.37

is

-0.37

 Ráp

-0.37

()")

-0.37

-0.36

isTrue

-0.36

POSITIVE LOGITS

 neighborhood

1.49

 Neighborhood

1.43

Neighborhood

1.38

neighborhood

1.31

 neighborhoods

1.30

bourhood

1.27

 neighbourhood

1.23

 Neighbourhood

1.15

 neighbourhoods

1.05

ighborhood

0.98

Activations Density 0.005%

references to neighborhoods

neighborhood

No Comments

No Known Activations