INDEX

Explanations

names of individuals and family members

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

names of people

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

ielsen

-0.37

comfortably

-0.36

drawSprites

-0.35

 fisuras

-0.34

typeorm

-0.33

polski

-0.33

optimized

-0.32

POP

-0.32

akhstan

-0.31

 quedado

-0.31

POSITIVE LOGITS

WriteBarrier

0.52

ainville

0.49

PerformLayout

0.49

ValueStyle

0.47

 оригіналу

0.47

tvguidetime

0.46

StoryboardSegue

0.46

########.

0.46

ArgsConstructor

0.45

UrlResolution

0.44

Activations Density 0.003%

names of individuals and family members

names of people

No Comments

No Known Activations