INDEX

Explanations

year

np_max-act · gemini-2.0-flash

The neuron triggers on mentions of years or other numeric temporal metadata (e.g. “2007,” “of the year,” month names with numbers), flagging references to dates or time indicators.

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

 voiced

-0.07

cc

-0.06

.visitMethod

-0.06

_cost

-0.06

_source

-0.06

 Δε

-0.06

 gigantic

-0.06

 achieved

-0.06

ترین

-0.06

<=(

-0.06

POSITIVE LOGITS

管

0.07

 ACPI

0.07

 runes

0.06

tutorial

0.06

 служби

0.06

.sup

0.06

 ли

0.06

.weather

0.06

San

0.06

野

0.06

Activations Density 0.024%

year

The neuron triggers on mentions of years or other numeric temporal metadata (e.g. “2007,” “of the year,” month names with numbers), flagging references to dates or time indicators.

No Comments

No Known Activations