INDEX

Explanations

possessive pronouns

np_max-act · gemini-2.0-flash

The neuron activates on third-person possessive pronouns (like “his,” “her,” “its”), marking words that signal ownership or relation.

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

 dohod

-0.07

 courtroom

-0.06

 obchodní

-0.06

ovými

-0.06

 jméno

-0.06

Bro

-0.06

odon

-0.06

(G

-0.06

 guardian

-0.06

.Att

-0.06

POSITIVE LOGITS

 عند

0.06

 görüş

0.06

.finish

0.06

_scal

0.06

Patch

0.06

osu

0.06

 Grammy

0.06

.mesh

0.06

.getView

0.05

_sheet

0.05

Activations Density 0.180%

possessive pronouns

The neuron activates on third-person possessive pronouns (like “his,” “her,” “its”), marking words that signal ownership or relation.

No Comments

No Known Activations