INDEX

Explanations

needed

np_max-act · gemini-2.0-flash

This neuron detects hedging or cautionary phrases indicating limited evidence and the need for further research.

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

 WLAN

-0.06

ROC

-0.06

ont

-0.06

ôn

-0.06

роф

-0.06

视频

-0.06

野

-0.06

DEN

-0.06

xx

-0.06

γμα

-0.06

POSITIVE LOGITS

/**
↵

0.07

 bilim

0.07

ky

0.06

([
↵

0.06

 dục

0.06

irtschaft

0.06

 مساحت

0.06

(requestCode

0.06

 ауд

0.06

 Düş

0.06

Activations Density 0.014%

needed

This neuron detects hedging or cautionary phrases indicating limited evidence and the need for further research.

No Comments

No Known Activations