INDEX

Explanations

comma

np_max-act · gemini-2.0-flash

The neuron activates on proper names in the author list, specifically recognizing author surnames.

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

 contraction

-0.07

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

-0.06

พวก

-0.06

 IntPtr

-0.06

 domácí

-0.06

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

-0.06

 работу

-0.06

 Vaults

-0.06

 gunman

-0.06

.Col

-0.06

POSITIVE LOGITS

abr

0.07

wig

0.07

ubb

0.07

 Hamburg

0.06

 Validates

0.06

 previews

0.06

 seemed

0.06

 Townsend

0.06

chied

0.06

 denies

0.06

Activations Density 0.014%

comma

The neuron activates on proper names in the author list, specifically recognizing author surnames.

No Comments

No Known Activations

comma

The neuron activates on proper names in the author list, specifically recognizing author surnames.

No Comments

No Known Activations