INDEX

Explanations

Punctuation

np_max-act · gemini-2.0-flash

This neuron activates on named entities—proper nouns such as place names, organization titles, and distinctive capitalized terms.

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

 poured

-0.07

deaux

-0.06

 Dien

-0.06

_app

-0.06

Cru

-0.06

dny

-0.06

RAW

-0.06

liter

-0.06

_DATE

-0.06

خاب

-0.06

POSITIVE LOGITS

.DefaultCellStyle

0.07

'):

0.06

 phương

0.06

_VENDOR

0.06

.JOptionPane

0.06

 yielding

0.06

*\

0.06

_machine

0.06

_ot

0.06

 achieved

0.06

Activations Density 0.488%

Punctuation

This neuron activates on named entities—proper nouns such as place names, organization titles, and distinctive capitalized terms.

No Comments

No Known Activations