INDEX

Explanations

Something ending

np_max-act · gemini-2.0-flash

The neuron is sensitive to text that looks like section or article headlines (e.g. “Shutting Down Services”).

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

 kvůli

-0.07

QPushButton

-0.06

 impatient

-0.06

 quốc

-0.06

観

-0.06

 χω

-0.06

 metic

-0.06

	ZEPHIR

-0.06

 конечно

-0.06

 questa

-0.06

POSITIVE LOGITS

higher

0.07

 TIMER

0.07

เฮ

0.06

setWidth

0.06

Fa

0.06

acock

0.06

 Maria

0.06

_enable

0.06

covered

0.06

ѕ

0.06

Activations Density 0.038%

Something ending

The neuron is sensitive to text that looks like section or article headlines (e.g. “Shutting Down Services”).

No Comments

No Known Activations