INDEX

Explanations

Our understanding

np_max-act · gemini-2.0-flash

The neuron selectively activates on words that signal scientific progress or current knowledge—terms like “understanding,” “knowledge,” “progress,” “discovery,” and “insight.”

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

_reads

-0.07

_ping

-0.07

-functions

-0.06

�

-0.06

�

-0.06

 Сп

-0.06

麗

-0.06

);;↵

-0.06

ボ

-0.06

 analyzes

-0.06

POSITIVE LOGITS

 Trang

0.08

 Châu

0.07

 knowing

0.06

_lastname

0.06

ै।

0.06

>?

0.06

==(

0.06

 suspicion

0.06

 extent

0.06

Activations Density 0.034%

Our understanding

The neuron selectively activates on words that signal scientific progress or current knowledge—terms like “understanding,” “knowledge,” “progress,” “discovery,” and “insight.”

No Comments

No Known Activations