INDEX

Explanations

Pokémon types

np_max-act · gemini-2.0-flash

The neuron activates on Pokémon “type” and stat‐category words (e.g. Fire, Water, Electric, Physical, Special, etc.).

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

(package

-0.07

 customization

-0.06

spotify

-0.06

�

-0.06

 Feinstein

-0.06

acements

-0.06

욕

-0.06

 nanoparticles

-0.06

막

-0.06

 bakış

-0.06

POSITIVE LOGITS

Sound

0.07

 abandoning

0.06

LLU

0.06

lete

0.06

umble

0.06

 thinking

0.06

larda

0.06

 registering

0.06

 ninth

0.06

Activations Density 0.008%

Pokémon types

The neuron activates on Pokémon “type” and stat‐category words (e.g. Fire, Water, Electric, Physical, Special, etc.).

No Comments

No Known Activations