INDEX

Explanations

pronouns you me us

np_max-act · gemini-2.0-flash

This neuron activates on direct, accusatory “you … me” statements—phrases where the speaker addresses someone (“you”) about something done to themselves (“me”).

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

 seaside

-0.06

 ISBN

-0.06

 powdered

-0.06

 спос

-0.06

 trained

-0.06

тот

-0.06

 layered

-0.06

 parenting

-0.06

 synth

-0.06

POSITIVE LOGITS

you

0.10

me

0.10

us

0.09

him

0.08

Us

0.08

WE

0.08

we

0.07

 them

0.07

.we

0.07

Activations Density 0.140%

pronouns you me us

This neuron activates on direct, accusatory “you … me” statements—phrases where the speaker addresses someone (“you”) about something done to themselves (“me”).

No Comments

No Known Activations