INDEX

Explanations

't

np_max-act · gemini-2.0-flash

This neuron detects the apostrophe-t sequence (as in the “’th” suffix) marking ordinal terms in the questions.

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

 trash

-0.07

OB

-0.07

 trop

-0.07

нів

-0.07

 broker

-0.06

OI

-0.06

zano

-0.06

MIC

-0.06

řeh

-0.06

ोव

-0.06

POSITIVE LOGITS

>')↵

0.07

Rent

0.06

]").

0.06

cdecl

0.06

Numbers

0.06

ocities

0.06

 {...

0.06

 criminal

0.06

Viet

0.06

'al

0.06

Activations Density 0.001%

't

This neuron detects the apostrophe-t sequence (as in the “’th” suffix) marking ordinal terms in the questions.

No Comments

No Known Activations

't

This neuron detects the apostrophe-t sequence (as in the “’th” suffix) marking ordinal terms in the questions.

No Comments

No Known Activations