INDEX

Explanations

information and knowledge

np_max-act · gemini-2.0-flash

This neuron detects explicit instructional requests—especially the phrase “Explain in detail.”

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

.structure

-0.07

_Function

-0.07

isa

-0.07

 अवध

-0.06

.dimensions

-0.06

.centerX

-0.06

ительных

-0.06

antar

-0.06

ùy

-0.06

 CONS

-0.06

POSITIVE LOGITS

 mực

0.06

rb

0.06

 notifies

0.06

혹

0.06

물을

0.06

�

0.06

gi

0.06

лекс

0.06

 gelir

0.06

 reddit

0.06

Activations Density 0.304%

information and knowledge

This neuron detects explicit instructional requests—especially the phrase “Explain in detail.”

No Comments

No Known Activations