INDEX

Explanations

research paper titles

np_max-act · gemini-2.0-flash

The neuron activates on the distinctive keywords that appear in the subtitle portion of scientific paper titles (i.e. the main content word right after a colon).

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

Hoy

-0.06

OCR

-0.06

 analyzing

-0.06

empt

-0.06

 misogyn

-0.06

ITT

-0.06

 Alter

-0.06

 Archae

-0.06

TY

-0.06

 Sims

-0.06

POSITIVE LOGITS

elix

0.07

(sequence

0.06

:end

0.06

.getClient

0.06

keyboard

0.06

ूबर

0.06

east

0.06

sylvania

0.06

)))

0.06

 serialize

0.06

Activations Density 0.049%

research paper titles

The neuron activates on the distinctive keywords that appear in the subtitle portion of scientific paper titles (i.e. the main content word right after a colon).

No Comments

No Known Activations