INDEX

Explanations

199

np_max-act · gemini-2.0-flash

The neuron activates on four-digit numeric tokens (e.g., years) typically found in copyright or header comments.

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

 enthusiasts

-0.07

saldo

-0.07

GRID

-0.07

 Lama

-0.06

ald

-0.06

Services

-0.06

۷

-0.06

$PostalCodesNL

-0.06

 vacancies

-0.06

ضی

-0.06

POSITIVE LOGITS

(orig

0.07

Mit

0.06

 أبي

0.06

添加

0.06

 reporting

0.06

 mohla

0.06

 rehab

0.06

�

0.06

 __('

0.06

ปกครอง

0.06

Activations Density 0.003%

199

The neuron activates on four-digit numeric tokens (e.g., years) typically found in copyright or header comments.

No Comments

No Known Activations