INDEX

Explanations

people

np_max-act · gemini-2.0-flash

The neuron activates on words referring to people or groups of humans (e.g. “people,” “kids,” “guys,” “they,” “who”).

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

 dém

-0.06

job

-0.06

sou

-0.06

 knih

-0.06

anske

-0.06

emain

-0.06

γά

-0.06

つの

-0.06

-Mart

-0.06

POSITIVE LOGITS

 URLSession

0.07

----------------------------------------------------------------------------

0.06

@RequestMapping

0.06

箱

0.06

 Nose

0.06

 얼마

0.06

พอ

0.06

 امروز

0.06

 sentimental

0.06

 žena

0.06

Activations Density 0.032%

people

The neuron activates on words referring to people or groups of humans (e.g. “people,” “kids,” “guys,” “they,” “who”).

No Comments

No Known Activations