INDEX

Explanations

electrons

np_max-act · gemini-2.0-flash

The neuron activates on technical terms around electron and hole charge‐carrier processes—especially occurrences of “electron” (and related transfer/injection/acceptor) language.

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

.exit

-0.06

Forest

-0.06

cích

-0.06

่เป

-0.06

ฤษภาคม

-0.06

 Dabei

-0.06

σμο

-0.06

\(

-0.06

cstdlib

-0.06

bor

-0.06

POSITIVE LOGITS

 اهد

0.06

Act

0.06

э

0.06

 associated

0.06

ेख

0.06

duction

0.06

нед

0.06

 allot

0.06

 famil

0.06

orses

0.06

Activations Density 0.002%

electrons

The neuron activates on technical terms around electron and hole charge‐carrier processes—especially occurrences of “electron” (and related transfer/injection/acceptor) language.

No Comments

No Known Activations