INDEX

Explanations

Information search and discussion

np_max-act · gemini-2.0-flash

This neuron activates on tokens relating to the assistant’s ethical guidelines, rules, restrictions, and safety or privacy policies.

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Less

-0.06

riors

-0.06

etr

-0.06

Bru

-0.06

χεία

-0.06

猜

-0.06

 elbows

-0.06

出品者

-0.06

 Arbeits

-0.06

结果

-0.06

POSITIVE LOGITS

 GLuint

0.07

big

0.06

Jwt

0.06

lime

0.06

ента

0.06

менту

0.06

Orm

0.06

 reflux

0.06

Lights

0.06

 axial

0.06

Activations Density 0.036%

Information search and discussion

This neuron activates on tokens relating to the assistant’s ethical guidelines, rules, restrictions, and safety or privacy policies.

No Comments

No Known Activations