INDEX

Explanations

references to the word "you"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

ⓧ

-0.92

WillAppear

-0.66

 Normdatei

-0.63

parsedMessage

-0.61

あれば

-0.60

+:+

-0.60

ыгана

-0.59

styleable

-0.59

SourceChecksum

-0.59

 oublié

-0.59

POSITIVE LOGITS

 know

1.18

know

0.99

 Know

0.85

Know

0.71

ya

0.71

 sabes

0.71

 sabe

0.69

 KNOW

0.67

KNOW

0.66

Activations Density 0.077%

references to the word "you"

No Comments

No Known Activations