INDEX

Explanations

Names

np_max-act · gemini-2.0-flash

The neuron selectively activates on tokens that are parts of personal names (proper‐name tokens) in the text.

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

 sın

-0.08

ادن

-0.06

 Texans

-0.06

ěr

-0.06

unden

-0.06

>ID

-0.06

mist

-0.06

ILD

-0.06

 Reliable

-0.06

Trust

-0.06

POSITIVE LOGITS

 enables

0.07

amientos

0.07

 discussions

0.07

 facilitates

0.07

la

0.07

 exiting

0.07

 Как

0.07

 Στο

0.07

ури

0.06

Activations Density 0.044%

Names

The neuron selectively activates on tokens that are parts of personal names (proper‐name tokens) in the text.

No Comments

No Known Activations