INDEX

Explanations

Code

np_max-act · gemini-2.0-flash

This neuron fires on CamelCase code identifiers (type, interface, and class names) rather than ordinary words.

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

�

-0.07

CONST

-0.07

moduleName

-0.06

Alex

-0.06

AFP

-0.06

 Recogn

-0.06

 станд

-0.06

outlined

-0.06

waukee

-0.06

ůž

-0.06

POSITIVE LOGITS

�

0.06

 precaution

0.06

:Is

0.06

Nam

0.06

 İç

0.06

۱۴

0.06

 lesser

0.05

 PURE

0.05

jie

0.05

_HE

0.05

Activations Density 0.018%

Code

This neuron fires on CamelCase code identifiers (type, interface, and class names) rather than ordinary words.

No Comments

No Known Activations