INDEX

Explanations

references to measurements or dimensions

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

shorten >= 1pt

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

 betweenstory

-0.75

 conseguenza

-0.46

SuspendLayout

-0.46

GraphicsUnit

-0.45

RegressionTest

-0.43

antaranya

-0.41

NullCheck

-0.41

╯

-0.41

 وسلم

-0.40

HandleFunc

-0.40

POSITIVE LOGITS

IntoConstraints

0.47

Amical

0.43

contentLoaded

0.41

Abitanti

0.38

Agreement

0.37

mpeg

0.37

 opacity

0.37

 openness

0.36

makeConstraints

0.36

max

0.36

Activations Density 0.000%

references to measurements or dimensions

shorten >= 1pt

No Comments

No Known Activations