INDEX

Explanations

a

np_max-act · gemini-2.0-flash

This neuron fires on concrete object or component names (e.g. UI widgets like ImageView, Button, TextView or in-scene items like camera, ball, basket).

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

 dobré

-0.06

 рабоч

-0.06

 Когда

-0.06

 jury

-0.06

였

-0.06

 черв

-0.06

 ніби

-0.06

輪

-0.05

 Kend

-0.05

POSITIVE LOGITS

 Calculation

0.07

 issu

0.07

 programme

0.07

atalog

0.06

(properties

0.06

 Pass

0.06

vac

0.06

.getHeight

0.06

(dist

0.06

============↵

0.06

Activations Density 0.043%

a

This neuron fires on concrete object or component names (e.g. UI widgets like ImageView, Button, TextView or in-scene items like camera, ball, basket).

No Comments

No Known Activations

a

This neuron fires on concrete object or component names (e.g. UI widgets like ImageView, Button, TextView or in-scene items like camera, ball, basket).

No Comments

No Known Activations