INDEX

Explanations

code

np_max-act · gemini-2.0-flash

This neuron fires on Portuguese instructional or descriptive text (e.g. code comments or prompts) rather than on English.

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

册

-0.06

 forced

-0.06

.getDate

-0.06

******

-0.06

oled

-0.06

.element

-0.06

	Global

-0.06

.block

-0.06

 preceded

-0.06

:before

-0.06

POSITIVE LOGITS

 Penal

0.07

ミュ

0.07

asz

0.07

 Quiet

0.06

 yurt

0.06

 juni

0.06

.internet

0.06

 javascript

0.06

Og

0.06

ント

0.06

Activations Density 0.060%

code

This neuron fires on Portuguese instructional or descriptive text (e.g. code comments or prompts) rather than on English.

No Comments

No Known Activations