INDEX

Explanations

code dealing with user input

np_max-act · gemini-2.0-flash

This neuron never activates on any input tokens, i.e. it doesn’t detect any particular pattern.

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

 consoles

-0.06

kos

-0.06

γραφ

-0.06

 çev

-0.06

lbs

-0.06

Circle

-0.06

traits

-0.06

RAM

-0.06

rians

-0.06

 chatter

-0.06

POSITIVE LOGITS

<Object

0.08

imensional

0.07

所属

0.06

_collections

0.06

 biri

0.06

 تاریخی

0.06

 bans

0.06

｀

0.06

 بسته

0.06

.divide

0.06

Activations Density 0.018%

code dealing with user input

This neuron never activates on any input tokens, i.e. it doesn’t detect any particular pattern.

No Comments

No Known Activations