INDEX

Explanations

questions involving edge cases

np_max-act · gemini-2.0-flash

This neuron detects comparative constructions—phrases where one thing is compared to another (e.g. “X than Y,” “can be smaller,” “negative …?”).

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

 test

-0.07

 iteration

-0.06

各

-0.06

']}</

-0.06

ыми

-0.06

 user

-0.06

 distribute

-0.06

bag

-0.06

ั่

-0.06

Prec

-0.06

POSITIVE LOGITS

 Christopher

0.07

Christopher

0.07

-tip

0.06

_Impl

0.06

Thus

0.06

Brand

0.06

 sinon

0.06

WAS

0.06

CLEAR

0.06

 وسلم

0.06

Activations Density 0.181%

questions involving edge cases

This neuron detects comparative constructions—phrases where one thing is compared to another (e.g. “X than Y,” “can be smaller,” “negative …?”).

No Comments

No Known Activations