INDEX

Explanations

phrases indicating absence or negation

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

"nothing" followed by a qualifier

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.24.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

evt

-0.56

PDT

-0.54

odyear

-0.51

MST

-0.49

epo

-0.48

LDAP

-0.47

Wpf

-0.47

ffindor

-0.47

Patriot

-0.47

 ujednoznacz

-0.46

POSITIVE LOGITS

 nothing

1.05

 Nothing

0.99

nothing

0.98

 NOTHING

0.97

NOTHING

0.95

Nothing

0.94

 nothin

0.87

 Nada

0.73

THING

0.72

 nada

0.69

Activations Density 0.010%

phrases indicating absence or negation

"nothing" followed by a qualifier

No Comments

No Known Activations