INDEX

Explanations

generic filler text

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_33.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.33.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

æµ´å®¤

-0.26

.ua

-0.25

quare

-0.25

stairs

-0.24

ese

-0.24

ç¿ĺ

-0.24

(io

-0.24

Sec

-0.23

Sem

-0.23

ç°Ł

-0.23

POSITIVE LOGITS

yet

0.36

yet

0.34

 amount

0.27

Yet

0.27

è¢«æī§è¡Į

0.26

mot

0.26

UDO

0.26

 amounts

0.26

 care

0.25

Yet

0.25

Activations Density 0.018%

generic filler text

No Known Activations