INDEX

Explanations

going

np_max-act · gemini-2.0-flash

The neuron strongly activates on occurrences of “going” (as in “going to”), marking that exact word in the text.

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

φ

-0.08

 Dense

-0.07

ást

-0.07

Firebase

-0.07

cle

-0.07

db

-0.07

ische

-0.07

 beau

-0.07

ější

-0.07

POSITIVE LOGITS

 going

0.12

 ought

0.08

 coming

0.07

 Prior

0.07

 Coming

0.07

Boy

0.07

 Going

0.07

imity

0.07

 according

0.07

 llegar

0.06

Activations Density 0.021%

going

The neuron strongly activates on occurrences of “going” (as in “going to”), marking that exact word in the text.

No Comments

No Known Activations