INDEX

Explanations

Code and quotations

np_max-act · gemini-2.0-flash

This neuron signals when you’re at or near the end of an input sequence, especially on end-of-text or final boundary tokens.

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

关于

-0.06

 ос

-0.06

altern

-0.06

�

-0.06

 defamation

-0.06

clo

-0.06

�

-0.06

.drawLine

-0.06

 forc

-0.06

 correlate

-0.06

POSITIVE LOGITS

 avantaj

0.07

 сервер

0.06

-lib

0.06

 WARRANTIES

0.06

.Border

0.06

.StatusOK

0.06

 Statue

0.06

を受

0.06

 Oversight

0.06

Activations Density 0.003%

Code and quotations

This neuron signals when you’re at or near the end of an input sequence, especially on end-of-text or final boundary tokens.

No Comments

No Known Activations