INDEX

Explanations

always

np_max-act · gemini-2.0-flash

The neuron fires on first-person narrative cues (pronouns like “I,” “me,” “my” and related self-referential verbs), effectively spotting when the text is in a personal, first-person voice.

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 lact

-0.07

Chain

-0.06

 nerv

-0.06

ppm

-0.06

自身

-0.06

TK

-0.06

νει

-0.06

Å

-0.06

played

-0.06

 polygon

-0.06

POSITIVE LOGITS

 kvinne

0.07

='',

0.06

.concat

0.06

Siz

0.06

.swift

0.06

.addWidget

0.06

prepare

0.06

ément

0.06

userManager

0.06

/loose

0.06

Activations Density 0.112%

always

The neuron fires on first-person narrative cues (pronouns like “I,” “me,” “my” and related self-referential verbs), effectively spotting when the text is in a personal, first-person voice.

No Comments

No Known Activations