INDEX

Explanations

negations and references to the concept of "no one" or "nobody."

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

"no" followed by "one" or "office"

np_acts-logits-general · gemini-2.0-flash

no one

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 onCreateView

-0.55

ėjimo

-0.53

gnore

-0.49

-0.47

 sanitarias

-0.45

timewa

-0.44

 laborales

-0.44

<bos>

-0.43

caris

-0.42

 ensimmä

-0.42

POSITIVE LOGITS

inguém

1.10

BODY

1.02

 nobody

1.01

 anyone

1.00

anyone

0.99

nobody

0.98

 anybody

0.94

 Nadie

0.93

 Nobody

0.91

 ANYONE

0.90

Activations Density 0.141%

negations and references to the concept of "no one" or "nobody."

"no" followed by "one" or "office"

no one

No Comments

No Known Activations