INDEX

Explanations

as

np_max-act · gemini-2.0-flash

This neuron detects legal role designations introduced by “as,” such as “as the administrator,” “as guardian,” “as personal representative,” or “as trustee” indicating a party’s fiduciary or estate capacity.

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

mongoose

-0.08

,#

-0.07

gunta

-0.06

ographics

-0.06

(sample

-0.06

_DEST

-0.06

udent

-0.06

INGLE

-0.06

men

-0.06

opolitan

-0.06

POSITIVE LOGITS

 Tara

0.06

 reviewer

0.06

九

0.06

RANDOM

0.06

drv

0.06

 doom

0.06

 Represent

0.06

Guid

0.06

 demonstrating

0.05

Ges

0.05

Activations Density 0.007%

as

This neuron detects legal role designations introduced by “as,” such as “as the administrator,” “as guardian,” “as personal representative,” or “as trustee” indicating a party’s fiduciary or estate capacity.

No Comments

No Known Activations

as

This neuron detects legal role designations introduced by “as,” such as “as the administrator,” “as guardian,” “as personal representative,” or “as trustee” indicating a party’s fiduciary or estate capacity.

No Comments

No Known Activations