INDEX

Explanations

phrases that indicate simplicity or ease of use

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

AndEndTag

-1.23

WireFormatLite

-1.20

sidemargin

-1.20

 ProtoMessage

-1.19

 <=",

-1.11

RegistryLite

-1.07

Personensuche

-1.05

 AssemblyTitle

-1.05

ArrowToggle

-1.03

contentLoaded

-1.03

POSITIVE LOGITS

 easy

2.18

 Easy

2.00

Easy

1.95

 EASY

1.88

easy

1.88

EASY

1.75

 easier

1.53

eas

1.34

 facile

1.32

 easily

1.32

Activations Density 0.076%

phrases that indicate simplicity or ease of use

No Comments

No Known Activations