INDEX

Explanations

commas or web addresses

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

æĥ³åİ»

-0.27

ofile

-0.27

burn

-0.27

ä¸Ģå¤Ħ

-0.26

zier

-0.26

è¿ĺä¸įå¤Ł

-0.25

æŀģèĩ´

-0.25

å«Į

-0.25

_put

-0.24

 Scenes

-0.24

POSITIVE LOGITS

poi

0.30

åħĭåĪ¶

0.28

ãĥĲãĥª

0.27

 Ð»Ð¸ÑĨÐ°

0.26

OPS

0.24

orro

0.23

avn

0.23

 Ð»Ð¸ÑĨ

0.23

 actually

0.23

panels

0.23

Activations Density 0.011%

commas or web addresses

No Comments

No Known Activations