INDEX

Explanations

email

np_max-act · gemini-2.0-flash

The neuron activates on occurrences of “email” (and related terms like “invalidEmail,” “@,” and “address”)—i.e. it detects mentions of email addresses or email‐validation contexts.

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

_PROD

-0.06

BBB

-0.06

 pant

-0.06

χ

-0.06

жу

-0.06

Containers

-0.06

asp

-0.06

 распрост

-0.06

.ObjectMapper

-0.06

AccountId

-0.06

POSITIVE LOGITS

لا

0.06

ообраз

0.06

pizza

0.06

erved

0.06

 еще

0.06

ούς

0.06

OH

0.06

 psychologist

0.06

portion

0.06

ve

0.06

Activations Density 0.003%

email

The neuron activates on occurrences of “email” (and related terms like “invalidEmail,” “@,” and “address”)—i.e. it detects mentions of email addresses or email‐validation contexts.

No Comments

No Known Activations