INDEX

Explanations

companies

np_max-act · gemini-2.0-flash

This neuron fires on tokens in the stock “company summary” phrasing (e.g. “X is a … company based in …”), i.e. the boilerplate lead‐in language of corporate descriptions.

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

tir

-0.07

/xml

-0.06

턴

-0.06

یره

-0.06

標準

-0.06

 eski

-0.06

progressbar

-0.06

[])
↵

-0.06

POSITIVE LOGITS

/channel

0.07

-most

0.07

 BUILD

0.06

 hace

0.06

 stumbling

0.06

 Gareth

0.06

_logs

0.06

 اتحاد

0.06

.geom

0.06

 пло

0.06

Activations Density 0.041%

companies

This neuron fires on tokens in the stock “company summary” phrasing (e.g. “X is a … company based in …”), i.e. the boilerplate lead‐in language of corporate descriptions.

No Comments

No Known Activations