INDEX

Explanations

numbers

np_max-act · gemini-2.0-flash

The neuron fires on numeric list markers—especially two‐digit, zero-padded sequence numbers like “01,” “02,” “03,” etc.

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

地球

-0.07

(float

-0.06

MouseMove

-0.06

 дисцип

-0.06

dic

-0.06

 Following

-0.06

fortunate

-0.06

hta

-0.06

エル

-0.06

POSITIVE LOGITS

ività

0.07

prz

0.07

 Organizer

0.06

 dumpsters

0.06

larında

0.06

 політи

0.06

 péri

0.06

 spis

0.06

essay

0.06

ltre

0.06

Activations Density 0.014%

numbers

The neuron fires on numeric list markers—especially two‐digit, zero-padded sequence numbers like “01,” “02,” “03,” etc.

No Comments

No Known Activations