INDEX

Explanations

to

np_max-act · gemini-2.0-flash

This neuron lights up on discourse markers and directive cue words (like “First,” “One,” “Regarding,” “Additionally,” etc.) that introduce pieces of advice or steps.

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

 deferred

-0.07

 incent

-0.07

 unusual

-0.06

_source

-0.06

่เป

-0.06

็ก

-0.06

 leds

-0.06

ape

-0.06

Eb

-0.06

United

-0.06

POSITIVE LOGITS

employer

0.07

")]↵↵

0.07

aio

0.07

ोज

0.06

-times

0.06

.setHorizontal

0.06

 игра

0.06

-dem

0.06

(runtime

0.06

%#

0.06

Activations Density 0.111%

to

This neuron lights up on discourse markers and directive cue words (like “First,” “One,” “Regarding,” “Additionally,” etc.) that introduce pieces of advice or steps.

No Comments

No Known Activations

to

This neuron lights up on discourse markers and directive cue words (like “First,” “One,” “Regarding,” “Additionally,” etc.) that introduce pieces of advice or steps.

No Comments

No Known Activations