INDEX

Explanations

a

np_max-act · gemini-2.0-flash

The neuron activates on phrases describing someone “moving away” (e.g. “moving away,” “moved away”), especially in contexts where that departure is presented as a (mis-)statement or lie.

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

 gris

-0.07

šky

-0.06

Ra

-0.06

IsUnicode

-0.06

 Contains

-0.06

YT

-0.06

 Dylan

-0.06

urre

-0.06

kry

-0.06

(TIM

-0.06

POSITIVE LOGITS

 imperfect

0.07

ผ

0.07

調

0.06

ají

0.06

έρει

0.06

-not

0.06

 cloudy

0.06

 nyní

0.06

 DirectoryInfo

0.06

 najle

0.06

Activations Density 0.030%

a

The neuron activates on phrases describing someone “moving away” (e.g. “moving away,” “moved away”), especially in contexts where that departure is presented as a (mis-)statement or lie.

No Comments

No Known Activations

a

The neuron activates on phrases describing someone “moving away” (e.g. “moving away,” “moved away”), especially in contexts where that departure is presented as a (mis-)statement or lie.

No Comments

No Known Activations