INDEX

Explanations

proper nouns

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_31.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.31.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

pen

-0.27

opensource

-0.26

åı¯è¾¾

-0.26

ppelin

-0.25

swick

-0.24

Cah

-0.24

pot

-0.24

èĲ½åľ°

-0.24

mag

-0.24

pot

-0.24

POSITIVE LOGITS

 bursting

0.30

åĽ¢è´Ń

0.27

emet

0.27

åĪĨåŃĲ

0.26

ê¸Ģ

0.26

acciones

0.24

esz

0.24

dzi

0.24

 parcels

0.24

esor

0.23

Activations Density 0.802%

proper nouns

No Known Activations