INDEX

Explanations

technical explanations

np_max-act · gemini-2.0-flash

The neuron consistently activates on numeric tokens—especially decimal-formatted numbers—i.e. it’s looking for numeric values in the text.

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

 رف

-0.07

 prosec

-0.06

.Adapter

-0.06

['_

-0.06

 Turbo

-0.06

.handleClick

-0.06

 sourceMappingURL

-0.06

 elementType

-0.06

 Orth

-0.06

POSITIVE LOGITS

 adhesive

0.07

wil

0.07

 spacer

0.07

nil

0.07

/frame

0.07

ToolBar

0.06

 comment

0.06

ิกา

0.06

 show

0.06

 renewables

0.06

Activations Density 0.233%

technical explanations

The neuron consistently activates on numeric tokens—especially decimal-formatted numbers—i.e. it’s looking for numeric values in the text.

No Comments

No Known Activations