INDEX

Explanations

websites and information

np_max-act · gemini-2.0-flash

The neuron activates on tokens naming or describing online resource sites (e.g. “website,” platform names, URLs, 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

UBY

-0.07

iolet

-0.07

-0.06

ível

-0.06

()`

-0.06

_pago

-0.06

(check

-0.06

iki

-0.06

来た

-0.06

POSITIVE LOGITS

arkin

0.06

adows

0.06

إنجليزية

0.06

meaning

0.06

 fullName

0.06

ßen

0.06

 Shorts

0.06

 bridal

0.06

BarController

0.06

Undefined

0.06

Activations Density 0.019%

websites and information

The neuron activates on tokens naming or describing online resource sites (e.g. “website,” platform names, URLs, etc.).

No Comments

No Known Activations

websites and information

The neuron activates on tokens naming or describing online resource sites (e.g. “website,” platform names, URLs, etc.).

No Comments

No Known Activations