INDEX

Explanations

nature

np_max-act · gemini-2.0-flash

This neuron detects occurrences of “nature” (often in adjective–nature phrases, e.g. “distributed nature,” “resonant nature,” “non-linear nature”).

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

 justo

-0.07

ΟΜ

-0.06

 "../../

-0.06

/de

-0.06

Decl

-0.06

 tightened

-0.06

앤

-0.06

оро

-0.06

/chart

-0.06

(`

-0.06

POSITIVE LOGITS

ница

0.07

 CACHE

0.07

leo

0.06

rep

0.06

기업

0.06

_IL

0.06

 turb

0.06

 Depth

0.06

gist

0.06

การท

0.06

Activations Density 0.018%

nature

This neuron detects occurrences of “nature” (often in adjective–nature phrases, e.g. “distributed nature,” “resonant nature,” “non-linear nature”).

No Comments

No Known Activations