INDEX

Explanations

ing

np_max-act · gemini-2.0-flash

This neuron strongly responds to the sub‐token “ing,” i.e. gerund/continuous verb endings.

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

statements about fairness and personal working standards.

oai_token-act-pair · gpt-4o-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

isn

-0.08

 weren

-0.07

 wasn

-0.07

_preference

-0.06

was

-0.06

_ref

-0.06

isl

-0.06

 معن

-0.06

ole

-0.06

bed

-0.06

POSITIVE LOGITS

 Jeremiah

0.07

 monastery

0.06

 resisting

0.06

 vstup

0.06

 sitesinde

0.06

.ActionBar

0.06

 очередь

0.06

地區

0.06

 occurring

0.06

 complying

0.06

Activations Density 0.554%

ing

This neuron strongly responds to the sub‐token “ing,” i.e. gerund/continuous verb endings.

statements about fairness and personal working standards.

No Comments

No Known Activations