INDEX

Explanations

*

np_max-act · gemini-2.0-flash

This neuron fires on programming code syntax (e.g. import/require statements, braces, semicolons, and other code tokens) rather than natural-language text.

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

EventHandler

-0.07

 Spotlight

-0.06

рабат

-0.06

 appName

-0.06

.INT

-0.06

ographs

-0.06

GIN

-0.06

街道

-0.06

лами

-0.06

 ());↵↵

-0.06

POSITIVE LOGITS

 humming

0.06

 crew

0.06

 لف

0.06

.temperature

0.06

.createStatement

0.06

 strive

0.06

 LEGO

0.06

yap

0.06

总

0.06

tj

0.06

Activations Density 0.013%

*

This neuron fires on programming code syntax (e.g. import/require statements, braces, semicolons, and other code tokens) rather than natural-language text.

No Comments

No Known Activations