INDEX

Explanations

human

np_max-act · gemini-2.0-flash

This neuron activates most strongly on tokens referring to humans or humanity.

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

judul

-0.06

 хотя

-0.06

 Dutch

-0.06

 čer

-0.05

(log

-0.05

 porque

-0.05

 Krist

-0.05

halten

-0.05

 rockets

-0.05

ificacion

-0.05

POSITIVE LOGITS

"])↵

0.07

")
↵

0.07

Known

0.07

frica

0.07

'"↵

0.07

"|

0.07

 governed

0.07

 handing

0.06

 BlackBerry

0.06

.repository

0.06

Activations Density 0.016%

human

This neuron activates most strongly on tokens referring to humans or humanity.

No Comments

No Known Activations