INDEX

Explanations

a

np_max-act · gemini-2.0-flash

The neuron activates on specific technical or mechanical component and device names (e.g. “protective rosette,” “digger,” “bioreactor,” etc.).

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

ฐ

-0.06

 earthly

-0.06

-Encoding

-0.06

.phone

-0.06

argc

-0.06

	fwrite

-0.06

-se

-0.06

 intest

-0.06

 Potato

-0.06

ayload

-0.06

POSITIVE LOGITS

�

0.07

ायर

0.07

ons

0.06

¬

0.06

؟↵

0.06

)){

0.06

?↵↵

0.06

HF

0.06

 memnun

0.06

 debating

0.06

Activations Density 0.079%

a

The neuron activates on specific technical or mechanical component and device names (e.g. “protective rosette,” “digger,” “bioreactor,” etc.).

No Comments

No Known Activations