INDEX

Explanations

proper names of individuals

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

TagMode

-0.73

 مشين

-0.56

uxxxx

-0.55

 tartalomajánló

-0.54

ParallelGroup

-0.52

 виправивши

-0.50

MENAFN

-0.50

Datuak

-0.50

IntoConstraints

-0.50

ondissement

-0.50

POSITIVE LOGITS

 iParam

0.32

 bParam

0.31

PUBLISHED

0.31

Buffy

0.30

 loup

0.30

trash

0.30

的支持

0.29

rard

0.29

appro

0.29

Activations Density 0.012%

proper names of individuals

names of people

No Comments

No Known Activations