INDEX

Explanations

to

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

æĪĳä»¬ä¹Ł

-0.28

çľŁçĽ¸

-0.27

æīĵæ³ķ

-0.26

å¼Ģåıĳå»ºè®¾

-0.26

isen

-0.26

 using

-0.25

æĥ°

-0.25

ä»Ł

-0.24

 Ð±ÐµÐ·

-0.24

[<

-0.24

POSITIVE LOGITS

åĵªéĩĮ

0.32

 elsewhere

0.28

 abroad

0.28

 somewhere

0.28

ewhere

0.28

som

0.27

 wherever

0.27

åħ¨åĽ½åĲĦåľ°

0.26

 destinations

0.26

tÃ¡

0.25

Activations Density 0.019%

to

No Comments

No Known Activations