INDEX

Explanations

Var

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_32.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.32.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

actly

-0.31

-kit

-0.29

ably

-0.29

ibel

-0.27

ability

-0.27

ble

-0.26

Kre

-0.26

 Ability

-0.26

kne

-0.26

ibil

-0.26

POSITIVE LOGITS

áº£n

0.27

anas

0.26

lam

0.26

unes

0.26

asso

0.24

FS

0.24

é¢ł

0.24

woods

0.24

enna

0.24

andas

0.24

Activations Density 0.002%

Var

No Known Activations