INDEX

Explanations

data content

np_max-act · gemini-2.0-flash

This neuron responds to references to the actual data payload—words like “file,” “contents,” “body,” and “data” that indicate the file or message content being sent.

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

_authentication

-0.06

、本

-0.06

しま

-0.06

alloc

-0.06

_WIDGET

-0.06

_with

-0.06

عب

-0.06

WWW

-0.06

уд

-0.06

_float

-0.06

POSITIVE LOGITS

اهد

0.07

 Mail

0.07

 initialise

0.07

Dě

0.07

読

0.07

 Алекс

0.06

-makers

0.06

 lipstick

0.06

cul

0.06

 groupe

0.06

Activations Density 0.052%

data content

This neuron responds to references to the actual data payload—words like “file,” “contents,” “body,” and “data” that indicate the file or message content being sent.

No Comments

No Known Activations

data content

This neuron responds to references to the actual data payload—words like “file,” “contents,” “body,” and “data” that indicate the file or message content being sent.

No Comments

No Known Activations