INDEX

Explanations

a, an

np_max-act · gemini-2.0-flash

The neuron activates on words naming protective or enclosing mechanical components (e.g., casing, sleeve, housing).

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

rup

-0.08

Ey

-0.06

Notify

-0.06

งศ

-0.06

-Life

-0.06

 респ

-0.06

eneration

-0.06

 desperately

-0.06

gte

-0.06

 synt

-0.06

POSITIVE LOGITS

？”

0.07

 metros

0.07

 unstoppable

0.06

uber

0.06

inox

0.06

 modifying

0.06

 culo

0.06

 Tiffany

0.06

寸

0.06

Activations Density 0.041%

a, an

The neuron activates on words naming protective or enclosing mechanical components (e.g., casing, sleeve, housing).

No Comments

No Known Activations

a, an

The neuron activates on words naming protective or enclosing mechanical components (e.g., casing, sleeve, housing).

No Comments

No Known Activations