INDEX

Explanations

personal opinions

np_max-act · gemini-2.0-flash

The neuron detects descriptive intensity terms that convey strong emotion or high energy (e.g., "high," "energy," "emotional intensity").

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

 Newest

-0.07

 Rodney

-0.07

 Tile

-0.07

 EdgeInsets

-0.06

￣￣￣

-0.06

 Mavericks

-0.06

 suç

-0.06

_terminal

-0.06

 sorted

-0.06

 contexts

-0.06

POSITIVE LOGITS

خبر

0.07

0.06

ublice

0.06

.Business

0.06

Unix

0.06

 ^↵

0.06

 poměr

0.06

Chinese

0.06

clar

0.06

Administrator

0.06

Activations Density 0.071%

personal opinions

The neuron detects descriptive intensity terms that convey strong emotion or high energy (e.g., "high," "energy," "emotional intensity").

No Comments

No Known Activations

personal opinions

The neuron detects descriptive intensity terms that convey strong emotion or high energy (e.g., "high," "energy," "emotional intensity").

No Comments

No Known Activations