INDEX

Explanations

this

np_max-act · gemini-2.0-flash

The neuron activates on patent-related discourse (e.g. references to “this,” “patent,” “invention,” “disclosed,” “Carlisle patent,” “U.S. Pat. No. …,” and similar markers of patent citations and description).

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

yes

-0.07

 conv

-0.06

 crawling

-0.06

▍▍

-0.06

 daily

-0.06

ilerinin

-0.06

.Constraint

-0.06

 Ensemble

-0.06

ict

-0.06

panion

-0.06

POSITIVE LOGITS

 xlim

0.07

ΟΚ

0.06

 مان

0.06

ارية

0.06

.nextInt

0.06

ість

0.06

어서

0.06

osaur

0.06

ส

0.06

้อน

0.06

Activations Density 0.030%

this

The neuron activates on patent-related discourse (e.g. references to “this,” “patent,” “invention,” “disclosed,” “Carlisle patent,” “U.S. Pat. No. …,” and similar markers of patent citations and description).

No Comments

No Known Activations