INDEX

Explanations

formatting and structural elements within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_10/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 Certains

-0.36

stvo

-0.32

 Hatch

-0.32

Another

-0.32

 kele

-0.32

 Tiro

-0.31

selig

-0.31

 ανα

-0.30

 thyself

-0.30

 Certaines

-0.30

POSITIVE LOGITS

Personendaten

0.73

parsedMessage

0.71

 estekak

0.60

GEBURTSDATUM

0.60

 OMITBAD

0.60

∙

0.58

windowFixed

0.58

–

0.57

⋅

0.57

Activations Density 0.518%

formatting and structural elements within the text

No Comments

No Known Activations