INDEX

Explanations

prepositions

np_max-act · gemini-2.0-flash

This neuron activates on isolated, capitalized prepositions or short words (e.g. “In,” “For,” “From,” “At”) used as section headings or labels.

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

 Gets

-0.07

۱

-0.06

 кто

-0.06

 designated

-0.06

dhe

-0.06

/file

-0.06

 принадлеж

-0.06

albums

-0.06

 وهو

-0.06

 Westminster

-0.06

POSITIVE LOGITS

 Purs

0.07

 nást

0.07

 Over

0.07

-Owned

0.07

.In

0.06

 chrom

0.06

As

0.06

asename

0.06

 últimos

0.06

نة

0.06

Activations Density 0.079%

prepositions

This neuron activates on isolated, capitalized prepositions or short words (e.g. “In,” “For,” “From,” “At”) used as section headings or labels.

No Comments

No Known Activations

prepositions

This neuron activates on isolated, capitalized prepositions or short words (e.g. “In,” “For,” “From,” “At”) used as section headings or labels.

No Comments

No Known Activations