INDEX

Explanations

topics that are divided into two distinct categories or classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

casio

-0.53

DockStyle

-0.51

lyt

-0.50

méno

-0.49

 battre

-0.48

 spese

-0.48

 кӀ

-0.47

(_,

-0.46

المشاركات

-0.46

たまた

-0.45

POSITIVE LOGITS

+#+#

0.98

RenderAtEndOf

0.82

TagMode

0.73

 pinulongan

0.71

+:+

0.71

routeProvider

0.64

 صوتيه

0.62

<",

0.60

withIdentifier

0.59

ChildIndex

0.56

Activations Density 0.368%

topics that are divided into two distinct categories or classifications

No Comments

No Known Activations