INDEX

Explanations

proper nouns, particularly names and titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 13-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_13/width_16k/average_l0_132

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.13.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

RectangleBorder

-0.74

adaptiveStyles

-0.60

MigrationBuilder

-0.59

$_(

-0.58

AnimationsModule

-0.58

 ligiloj

-0.57

 BorderRadius

-0.55

 BoxDecoration

-0.54

ArgumentParser

-0.54

URLException

-0.53

POSITIVE LOGITS

WillAppear

0.43

참고

0.40

期刊论文

0.32

 nadie

0.31

BY

0.30

 отече

0.29

 afforded

0.29

依

0.28

썼

0.28

 serio

0.27

Activations Density 0.457%

proper nouns, particularly names and titles

No Comments

No Known Activations