INDEX

Explanations

24

np_max-act · gemini-2.0-flash

the neuron detects descriptive modifiers and numeric quantifiers—e.g. “24×7,” “increased,” “multiple,” “new,” “growing,” “ongoing,” “evolving”—that emphasize extent, frequency, or novelty.

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

/location

-0.07

<Box

-0.06

ominated

-0.06

 Geographic

-0.06

 sagen

-0.06

_driver

-0.06

olesale

-0.06

_sun

-0.06

Routing

-0.06

htub

-0.06

POSITIVE LOGITS

 accom

0.08

 saat

0.07

 provide

0.06

 jeden

0.06

 providing

0.06

详

0.06

 Omni

0.06

 temas

0.06

eného

0.06

(Activity

0.06

Activations Density 0.314%

24

the neuron detects descriptive modifiers and numeric quantifiers—e.g. “24×7,” “increased,” “multiple,” “new,” “growing,” “ongoing,” “evolving”—that emphasize extent, frequency, or novelty.

No Comments

No Known Activations