INDEX

Explanations

scientific texts

np_max-act · gemini-2.0-flash

This neuron detects scientific or technical terms by picking up on common multi-syllabic Greek/Latin-derived prefixes (e.g. intra-, immono-, bio-, nano-, photo-) at the start of words.

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

_LIST

-0.07

 Elves

-0.06

 پایان

-0.06

 fgets

-0.06

_measurement

-0.06

Ans

-0.06

 اختصاص

-0.06

姆

-0.06

 yeşil

-0.06

vail

-0.06

POSITIVE LOGITS

 anom

0.07

सन

0.07

 compt

0.07

TOR

0.07

 دول

0.07

oins

0.07

.TypeString

0.06

 getActivity

0.06

""↵

0.06

Tao

0.06

Activations Density 0.048%

scientific texts

This neuron detects scientific or technical terms by picking up on common multi-syllabic Greek/Latin-derived prefixes (e.g. intra-, immono-, bio-, nano-, photo-) at the start of words.

No Comments

No Known Activations