INDEX

Explanations

language reflecting personal opinion and emotional expression

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 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

MessageTagHelper

-0.60

 дописавши

-0.50

ContentAsync

-0.48

queryInterface

-0.45

%}

-0.43

writerow

-0.41

WillAppear

-0.41

 useAppContext

-0.41

 digital

-0.41

aires

-0.40

POSITIVE LOGITS

WriteTagHelper

0.63

AndEndTag

0.55

 dieß

0.48

ніципалі

0.46

abestanden

0.44

 myſelf

0.42

 nonUne

0.42

новништво

0.42

 ſever

0.41

 duquel

0.41

Activations Density 0.639%

language reflecting personal opinion and emotional expression

No Comments

No Known Activations