INDEX

Explanations

Photoshop

np_max-act · gemini-2.0-flash

The neuron strongly activates on fragments of “Photoshop” or “photoshopped,” so it’s detecting mentions of image post-processing or photo editing.

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

ortho

-0.07

 НА

-0.07

提

-0.07

""".

-0.07

yak

-0.07

 encontr

-0.07

CustomLabel

-0.06

 chip

-0.06

chi

-0.06

 Alien

-0.06

POSITIVE LOGITS

 vaginal

0.07

 Residents

0.07

้าว

0.06

 گزارش

0.06

，为

0.06

ा।↵↵

0.06

 –↵↵

0.06

，並

0.06

岸

0.06

PARATOR

0.06

Activations Density 0.016%

Photoshop

The neuron strongly activates on fragments of “Photoshop” or “photoshopped,” so it’s detecting mentions of image post-processing or photo editing.

No Comments

No Known Activations