INDEX

Explanations

phrases indicating proximity or closeness to an ideal state or value

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

proximity to specific values

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

jface

-0.35

Jährige

-0.33

󠁢

-0.33

horesis

-0.33

 udział

-0.31

 tovább

-0.31

 Schild

-0.30

 banners

-0.30

czych

-0.30

 lettura

-0.30

POSITIVE LOGITS

Personendaten

0.72

 nearly

0.70

IntoConstraints

0.69

nearly

0.68

 Nearly

0.68

 resemble

0.67

Nearly

0.67

KommentareTeilen

0.66

CodeAttribute

0.66

 quase

0.66

Activations Density 0.093%

phrases indicating proximity or closeness to an ideal state or value

proximity to specific values

No Comments

No Known Activations