INDEX

Explanations

Methods

np_max-act · gemini-2.0-flash

This neuron detects section-heading labels (especially “Materials and Methods”) in scientific documents.

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

IA

-0.06

 элем

-0.06

 jiný

-0.06

KH

-0.06

ヒ

-0.06

 floral

-0.06

jectories

-0.06

 dirt

-0.06

 شاخه

-0.06

POSITIVE LOGITS

 impulses

0.07

Buying

0.07

ContentLoaded

0.06

Choosing

0.06

Aspect

0.06

 cheered

0.06

间

0.06

 Maintain

0.06

ustomed

0.06

 Choosing

0.06

Activations Density 0.001%

Methods

This neuron detects section-heading labels (especially “Materials and Methods”) in scientific documents.

No Comments

No Known Activations