INDEX

Explanations

Choosing between options

np_max-act · gemini-2.0-flash

This neuron lights up on question‐style usage inquiries—tokens like “should,” “when,” “use,” and the question mark in “When should I use X instead of Y?”–type queries.

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

нен

-0.06

kr

-0.06

 Approach

-0.06

 wikipedia

-0.06

 saldo

-0.06

 glad

-0.06

visitor

-0.06

 conquered

-0.06

 Compare

-0.06

gif

-0.06

POSITIVE LOGITS

ούς

0.06

XI

0.06

 examinations

0.06

.degree

0.06

 muži

0.06

/android

0.06

_component

0.06

ickets

0.06

kob

0.06

κού

0.06

Activations Density 0.131%

Choosing between options

This neuron lights up on question‐style usage inquiries—tokens like “should,” “when,” “use,” and the question mark in “When should I use X instead of Y?”–type queries.

No Comments

No Known Activations

Choosing between options

This neuron lights up on question‐style usage inquiries—tokens like “should,” “when,” “use,” and the question mark in “When should I use X instead of Y?”–type queries.

No Comments

No Known Activations