INDEX

Explanations

configuration/metadata

np_max-act · gemini-2.0-flash

This neuron detects JSON field‐name tokens, particularly the quoted keys like "@", "type", and similar property names.

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

 женщины

-0.07

默认

-0.07

 QUESTION

-0.07

 нада

-0.07

險

-0.06

(old

-0.06

\Page

-0.06

中心

-0.06

animate

-0.06

raç

-0.06

POSITIVE LOGITS

.addAction

0.07

َّ

0.06

 Heidi

0.06

 đóng

0.06

jur

0.06

ITERAL

0.06

 opted

0.06

лей

0.06

 instinct

0.06

pez

0.06

Activations Density 0.011%

configuration/metadata

This neuron detects JSON field‐name tokens, particularly the quoted keys like "@", "type", and similar property names.

No Comments

No Known Activations