INDEX

Explanations

Java code

np_max-act · gemini-2.0-flash

This neuron activates on Java type names (identifiers of classes or interfaces, especially CamelCase names).

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

INVAL

-0.06

"]=="

-0.06

memberOf

-0.06

 Pare

-0.06

 LinkedList

-0.06

 захворювання

-0.06

 adolescents

-0.06

&gt

-0.06

glm

-0.06

 yaparak

-0.06

POSITIVE LOGITS

Benefits

0.07

UID

0.06

TED

0.06

 diverted

0.06

side

0.06

.trip

0.06

스가

0.06

 fractured

0.06

under

0.06

 Stations

0.06

Activations Density 0.019%

Java code

This neuron activates on Java type names (identifiers of classes or interfaces, especially CamelCase names).

No Comments

No Known Activations