INDEX

Explanations

proper nouns, particularly names of individuals

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

first names with names

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 BorderSide

-0.44

<h1>

-0.41

 iconLine

-0.39

な

-0.36

 sprend

-0.36

 duele

-0.35

 OnDestroy

-0.33

HttpPut

-0.33

 tratto

-0.32

 spread

-0.31

POSITIVE LOGITS

 المعيارى

0.54

Życiorys

0.53

 endforeach

0.51

KURZBESCHREIBUNG

0.50

ſelben

0.50

Referanser

0.49

səhifə

0.48

XmlAccessType

0.48

</>

0.48

 ويكيپ

0.48

Activations Density 0.033%

proper nouns, particularly names of individuals

first names with names

No Comments

No Known Activations