INDEX

Explanations

Code/documentation snippets

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

 davon

-0.26

æĪĲæŀľ

-0.25

éĥ

-0.25

 pressure

-0.25

åĬĽ

-0.25

aÄĩ

-0.24

æ°ĳ

-0.24

é¢Į

-0.24

ious

-0.23

åı¯è°ĵ

-0.23

POSITIVE LOGITS

quires

0.27

romo

0.27

ivot

0.26

 Keeper

0.26

ä¹ĭç§°

0.26

olest

0.25

ajas

0.24

veled

0.24

]*)

0.24

lose

0.24

Activations Density 0.114%

Code/documentation snippets

No Comments

No Known Activations