INDEX

Explanations

written communications

np_max-act · gemini-2.0-flash

The neuron fires on words around presenting or drafting a “message” (e.g. “message,” “sample,” “draft,” “potential”), i.e. whenever the text is introducing or referring to composing a message.

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

_SEND

-0.08

 Dillon

-0.07

 Hope

-0.07

منی

-0.07

aware

-0.07

eração

-0.07

روب

-0.07

 dbus

-0.06

loading

-0.06

_where

-0.06

POSITIVE LOGITS

<>

0.06

 trif

0.06

!:

0.06

']?>"

0.06

?>↵↵↵

0.06

 لأ

0.06

 الأ

0.06

}:

0.06

 centrally

0.06

.checkSelfPermission

0.06

Activations Density 0.056%

written communications

The neuron fires on words around presenting or drafting a “message” (e.g. “message,” “sample,” “draft,” “potential”), i.e. whenever the text is introducing or referring to composing a message.

No Comments

No Known Activations

written communications

The neuron fires on words around presenting or drafting a “message” (e.g. “message,” “sample,” “draft,” “potential”), i.e. whenever the text is introducing or referring to composing a message.

No Comments

No Known Activations