INDEX

Explanations

video

np_max-act · gemini-2.0-flash

The neuron activates on mentions of embedded videos—i.e. tokens like “video” (and their accompanying timestamps or metadata).

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

_CURSOR

-0.07

ักท

-0.07

-automatic

-0.07

-Qaeda

-0.07

 Technique

-0.06

 chế

-0.06

chk

-0.06

_CONN

-0.06

 peč

-0.06

ีนาคม

-0.06

POSITIVE LOGITS

.cwd

0.07

_tl

0.06

 канди

0.06

Bir

0.06

hs

0.06

暂

0.06

 lifting

0.06

 bör

0.06

Gi

0.06

 Châu

0.06

Activations Density 0.020%

video

The neuron activates on mentions of embedded videos—i.e. tokens like “video” (and their accompanying timestamps or metadata).

No Comments

No Known Activations