INDEX

Explanations

pencil

np_max-act · gemini-2.0-flash

The neuron activates on mentions of pencil‐making components and tools (e.g. “pencil,” “graphite,” “lead,” “eraser”).

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

}());↵

-0.07

 เก

-0.07

NV

-0.07

 sockaddr

-0.07

 manifestation

-0.07

flutter

-0.07

osp

-0.06

.BatchNorm

-0.06

dimension

-0.06

 Mode

-0.06

POSITIVE LOGITS

 pencil

0.13

 pencils

0.12

 chalk

0.09

 cray

0.08

彩

0.08

-pencil

0.07

ENCIL

0.07

indic

0.07

.I

0.07

ussian

0.07

Activations Density 0.003%

pencil

The neuron activates on mentions of pencil‐making components and tools (e.g. “pencil,” “graphite,” “lead,” “eraser”).

No Comments

No Known Activations