INDEX

Explanations

Digits

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

åīįåįģ

-0.27

clamp

-0.26

 jÃ¤

-0.26

 preceding

-0.26

èĭĮ

-0.25

ahead

-0.25

åłĩ

-0.25

carry

-0.24

Ø¹Ø²

-0.24

RIES

-0.24

POSITIVE LOGITS

oro

0.30

reff

0.28

ÑģÑĭÐ»

0.27

emo

0.26

EMU

0.26

 Slut

0.26

etto

0.26

tere

0.25

mail

0.24

wal

0.24

Activations Density 0.025%

Digits

No Known Activations