INDEX

Explanations

EVEN

np_max-act · gemini-2.0-flash

The neuron consistently activates on the uppercase word “EVEN” (as used in “EVEN IF ADVISED…” in license disclaimers).

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

_Stop

-0.06

 Mistress

-0.06

<",

-0.06

 exper

-0.06

 kino

-0.06

useum

-0.06

Lud

-0.06

 whim

-0.06

 Millenn

-0.06

 Мор

-0.06

POSITIVE LOGITS

'[

0.07

한

0.07

ON

0.07

-gay

0.06

 indict

0.06

σα

0.06

(false

0.06

_tcb

0.06

osexual

0.06

ائی

0.06

Activations Density 0.000%

EVEN

The neuron consistently activates on the uppercase word “EVEN” (as used in “EVEN IF ADVISED…” in license disclaimers).

No Comments

No Known Activations