INDEX

Explanations

Programming loops

np_max-act · gemini-2.0-flash

The neuron fires on loop‐iteration constructs—especially parallel loop pragmas (e.g. OpenMP “#pragma omp parallel for”) and the “for” keyword.

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

모

-0.07

DWC

-0.06

Rows

-0.06

dire

-0.06

 воспал

-0.06

 earrings

-0.06

Star

-0.06

xcc

-0.06

.Points

-0.06

FD

-0.06

POSITIVE LOGITS

ağ

0.07

 yılında

0.07

每

0.07

指导

0.06

双

0.06

 DataAccess

0.06

nau

0.06

 квітня

0.06

 brute

0.06

_eye

0.06

Activations Density 0.008%

Programming loops

The neuron fires on loop‐iteration constructs—especially parallel loop pragmas (e.g. OpenMP “#pragma omp parallel for”) and the “for” keyword.

No Comments

No Known Activations