INDEX

Explanations

defined

np_max-act · gemini-2.0-flash

This neuron is primarily triggered by the hyphenated adjective phrase “well-defined.”

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

.directive

-0.07

 інформації

-0.07

 Cyber

-0.06

 rewriting

-0.06

 túi

-0.06

rgctx

-0.06

Hol

-0.06

 Cool

-0.06

ław

-0.06

BitConverter

-0.06

POSITIVE LOGITS

席

0.07

.Float

0.07

 فرض

0.06

 ден

0.06

 consumption

0.06

 chip

0.06

_ACTIV

0.06

straight

0.06

anguage

0.06

iffies

0.06

Activations Density 0.013%

defined

This neuron is primarily triggered by the hyphenated adjective phrase “well-defined.”

No Comments

No Known Activations