INDEX

Explanations

worry, concern

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_27.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.27.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

éĥ½è¯´

-0.29

çĵľ

-0.27

grass

-0.27

mando

-0.26

æĸ¯

-0.25

åĽĬ

-0.25

 Horny

-0.25

æĤ£

-0.25

è§Ĵèī²

-0.25

esk

-0.24

POSITIVE LOGITS

æĭħå¿§

0.34

 noticing

0.31

 concern

0.31

any

0.30

 worry

0.30

 alarm

0.29

 concerned

0.28

 alarmed

0.27

 worried

0.27

any

0.27

Activations Density 0.001%

worry, concern

No Comments

No Known Activations