INDEX

Explanations

intensifiers

np_max-act · gemini-2.0-flash

The neuron responds to markers of quantitative extent—decimal‐style numbers and words that indicate degrees or completeness (e.g. fully, partially, completely).

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

QWidget

-0.07

uffman

-0.07

्यत

-0.07

分别

-0.07

fires

-0.06

<TextView

-0.06

loom

-0.06

 дерев

-0.06

ारत

-0.06

<w

-0.06

POSITIVE LOGITS

mando

0.06

-devel

0.06

============

0.06

_cum

0.06

 conservatism

0.06

 hát

0.06

 rhythms

0.06

_parms

0.06

DN

0.06

 dokument

0.06

Activations Density 0.159%

intensifiers

The neuron responds to markers of quantitative extent—decimal‐style numbers and words that indicate degrees or completeness (e.g. fully, partially, completely).

No Comments

No Known Activations