INDEX

Explanations

Code snippets

np_max-act · gemini-2.0-flash

The neuron fires on Java/Android code identifiers—especially CamelCase class‐ and file‐names (e.g. “Order,” “Activity,” “AppCompatActivity,” “.java”).

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

rap

-0.07

 attractions

-0.07

associate

-0.06

 righteousness

-0.06

patches

-0.06

 pancakes

-0.06

 commenc

-0.06

 لم

-0.06

ارا

-0.06

 explored

-0.06

POSITIVE LOGITS

φέ

0.07

ay

0.07

 QDir

0.07

 особи

0.07

.ACTION

0.06

addTo

0.06

☴

0.06

 MainForm

0.06

Vk

0.06

 вий

0.06

Activations Density 0.038%

Code snippets

The neuron fires on Java/Android code identifiers—especially CamelCase class‐ and file‐names (e.g. “Order,” “Activity,” “AppCompatActivity,” “.java”).

No Comments

No Known Activations

Code snippets

The neuron fires on Java/Android code identifiers—especially CamelCase class‐ and file‐names (e.g. “Order,” “Activity,” “AppCompatActivity,” “.java”).

No Comments

No Known Activations