INDEX

Explanations

Paint

np_max-act · gemini-2.0-flash

The neuron detects the word “paint” and its morphological variants (painting, painter, paintings, etc.).

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

 Homo

-0.07

 computations

-0.07

ero

-0.07

Soc

-0.07

्रस

-0.07

csr

-0.06

 resource

-0.06

 Hero

-0.06

 groove

-0.06

OS

-0.06

POSITIVE LOGITS

 paint

0.13

 painting

0.12

 Paint

0.12

 Painting

0.11

paint

0.10

 painted

0.10

Paint

0.10

 paintings

0.09

 Antoine

0.08

ainting

0.07

Activations Density 0.011%

Paint

The neuron detects the word “paint” and its morphological variants (painting, painter, paintings, etc.).

No Comments

No Known Activations

Paint

The neuron detects the word “paint” and its morphological variants (painting, painter, paintings, etc.).

No Comments

No Known Activations