INDEX

Explanations

language translation

np_max-act · gemini-2.0-flash

This neuron activates on words naming specific domains or contexts (e.g. business, science, technology, culture).

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tug

-0.06

pawn

-0.06

RTP

-0.06

>>

-0.06

 jezd

-0.06

OB

-0.06

mighty

-0.05

 Rouge

-0.05

 carrot

-0.05

POSITIVE LOGITS

 січня

0.07

ftware

0.07

إ

0.07

 FirebaseDatabase

0.06

 adidas

0.06

 Elle

0.06

 конферен

0.06

เทพ

0.06

inject

0.06

Foto

0.06

Activations Density 0.052%

language translation

This neuron activates on words naming specific domains or contexts (e.g. business, science, technology, culture).

No Comments

No Known Activations