INDEX

Explanations

cannabis and drug-related content

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_11.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.11.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

é¢Ĩå¯¼èĢħ

-0.32

åħ³èĬĤ

-0.31

äºĭä¸ļéĥ¨

-0.31

é¢Ĩå¯¼

-0.30

 dimensional

-0.28

éª¨

-0.27

æ¯ĶåĪ©

-0.27

èĤ¿

-0.27

 leader

-0.25

atos

-0.25

POSITIVE LOGITS

ointments

0.29

 Steps

0.28

isodes

0.26

çľļ

0.25

 steps

0.25

urgeon

0.25

 Ivanka

0.24

screens

0.24

ensors

0.24

adj

0.24

Activations Density 0.008%

cannabis and drug-related content

No Comments

No Known Activations