INDEX

Explanations

code

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

åĲįåŃĹ

-0.60

çļĦåĲįåŃĹ

-0.55

 names

-0.54

 name

-0.52

åĲįç§°

-0.51

è¿Ļä¸ªåĲįåŃĹ

-0.50

name

-0.48

_name

-0.47

-name

-0.46

 naming

-0.45

POSITIVE LOGITS

 spokes

0.26

 Regel

0.25

æĶ¶èĹı

0.25

é«ĺéĢŁåıĳå±ķ

0.25

 resc

0.24

NOP

0.23

alue

0.23

åħįéĻ¤

0.23

 Fifth

0.23

elib

0.23

Activations Density 0.002%

code

No Comments

No Known Activations