INDEX

Explanations

terms indicating activation or enabling features

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

enable or disable functionality

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

-0.37

↵

-0.37

↵↵

-0.36

-0.35

-0.34

How

-0.33

What

-0.31

<strong>

-0.30

char

-0.30

POSITIVE LOGITS

IntoConstraints

1.12

AddTagHelper

1.09

AndEndTag

1.02

EnableWeb

1.01

 pinulongan

0.98

AddHtmlAttribute

0.97

 propOrder

0.96

featureID

0.92

istoitu

0.91

OGND

0.90

Activations Density 0.034%

terms indicating activation or enabling features

enable or disable functionality

No Comments

No Known Activations