INDEX

Explanations

electronics suffixes

np_max-act · gemini-2.0-flash

The neuron strongly activates on technical mentions of nonreciprocal optical or microwave components—terms like “non-reciprocal,” “optical isolator,” and “circulator.”

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

 refusing

-0.06

 unfit

-0.06

 одну

-0.06

な

-0.06

PDF

-0.06

Et

-0.06

patients

-0.06

_act

-0.06

Haz

-0.06

#for

-0.06

POSITIVE LOGITS

 Verfüg

0.07

 dessert

0.07

 zvý

0.06

 dieses

0.06

ernels

0.06

+='<

0.06

iven

0.06

terdam

0.06

):

0.06

realloc

0.06

Activations Density 0.035%

electronics suffixes

The neuron strongly activates on technical mentions of nonreciprocal optical or microwave components—terms like “non-reciprocal,” “optical isolator,” and “circulator.”

No Comments

No Known Activations