INDEX

Explanations

programming

np_max-act · gemini-2.0-flash

This neuron activates on numerical tokens—particularly numbers and numeric inputs (including decimals) used in tool/action descriptions.

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

oč

-0.08

//------------------------------------------------

-0.07

üyor

-0.07

 AppRoutingModule

-0.06

eyJ

-0.06

 barring

-0.06

 possível

-0.06

ücü

-0.06

 violating

-0.06

ैग

-0.06

POSITIVE LOGITS

 بسته

0.07

$instance

0.06

 Frankie

0.06

Must

0.06

ritional

0.06

ifikace

0.06

 Monster

0.06

 раствор

0.06

 chắc

0.06

ollapse

0.06

Activations Density 0.002%

programming

This neuron activates on numerical tokens—particularly numbers and numeric inputs (including decimals) used in tool/action descriptions.

No Comments

No Known Activations