INDEX

Explanations

non

np_max-act · gemini-2.0-flash

The neuron detects instances of “non” used as a prefix (often in hyphenated technical terms) to mark negation or absence.

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uptime

-0.07

 Approach

-0.07

реп

-0.06

Average

-0.06

Serial

-0.06

Receive

-0.06

_CAT

-0.06

 zombies

-0.06

Deletes

-0.06

 paní

-0.06

POSITIVE LOGITS

 headquartered

0.08

 autom

0.07

CPC

0.07

<Props

0.06

 Headquarters

0.06

 activated

0.06

.jdesktop

0.06

 juris

0.06

лугов

0.06

�

0.06

Activations Density 0.007%

non

The neuron detects instances of “non” used as a prefix (often in hyphenated technical terms) to mark negation or absence.

No Comments

No Known Activations