INDEX

Explanations

view

np_max-act · gemini-2.0-flash

The neuron specializes in spotting image‐caption terms—words like “view,” “photo,” “angle,” “side,” or “front” used to label or describe pictures.

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

("\\

-0.07

ила

-0.06

IO

-0.06

(primary

-0.06

halt

-0.06

 cable

-0.06

.dispatch

-0.06

 Thompson

-0.06

امل

-0.06

 McCarthy

-0.06

POSITIVE LOGITS

 Notre

0.07

Sortable

0.06

戒

0.06

 inspir

0.06

NRL

0.06

 投稿日

0.06

oriasis

0.06

larınız

0.06

 quad

0.06

ردد

0.06

Activations Density 0.011%

view

The neuron specializes in spotting image‐caption terms—words like “view,” “photo,” “angle,” “side,” or “front” used to label or describe pictures.

No Comments

No Known Activations