INDEX

Explanations

the word "an."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

NameInMap

-0.78

bosity

-0.76

 TextInputType

-0.74

 pinulongan

-0.71

venido

-0.67

ắp

-0.67

RenderAtEndOf

-0.66

iotti

-0.65

 ihop

-0.61

styleType

-0.61

POSITIVE LOGITS

an

2.88

AN

2.23

ans

1.43

anan

1.31

annya

1.21

An

1.17

ان

1.15

ANL

1.13

án

1.13

an

1.07

Activations Density 0.109%

the word "an."

No Comments

No Known Activations