INDEX

Explanations

screenshot

np_max-act · gemini-2.0-flash

The neuron responds to occurrences of the word “screenshot” (often by activating on the “shot”/“hots” substring).

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

 účet

-0.07

defgroup

-0.06

.IndexOf

-0.06

 Interaction

-0.06

Identifier

-0.06

avě

-0.06

 manifold

-0.06

Text

-0.06

��이

-0.06

POSITIVE LOGITS

 screenshot

0.12

 screenshots

0.10

creenshot

0.08

Screenshot

0.07

Hover

0.07

hots

0.07

 превыш

0.07

查看

0.07

 tweets

0.07

ган

0.07

Activations Density 0.003%

screenshot

The neuron responds to occurrences of the word “screenshot” (often by activating on the “shot”/“hots” substring).

No Comments

No Known Activations