INDEX

Explanations

are

np_max-act · gemini-2.0-flash

The neuron activates on mentions of the physical setting—specifically room and furniture terms like “bed,” “bedroom,” “bed leg,” and “living room.”

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

 persec

-0.07

aque

-0.07

 Equity

-0.07

 contaminants

-0.07

.borderColor

-0.06

 scaffold

-0.06

 innovative

-0.06

政

-0.06

 Technologies

-0.06

 Toys

-0.06

POSITIVE LOGITS

_Em

0.07

พอ

0.07

 життя

0.06

 graded

0.06

 failure

0.06

.groupControl

0.06

 allowable

0.06

.routing

0.06

 designate

0.06

(Intent

0.06

Activations Density 0.056%

are

The neuron activates on mentions of the physical setting—specifically room and furniture terms like “bed,” “bedroom,” “bed leg,” and “living room.”

No Comments

No Known Activations