INDEX

Explanations

Carroll or Carr

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

adoras

-0.30

ä¸įåľ¨

-0.27

ä¸įè®©

-0.25

çĸ²åĬ³

-0.24

alog

-0.24

like

-0.24

ãģĿãģĨãģ§

-0.24

/org

-0.23

ensa

-0.23

pid

-0.23

POSITIVE LOGITS

æĬ±æĢ¨

0.26

 prÃ³

0.25

çŃ¹

0.25

anium

0.25

 HEADER

0.24

æİĮå£°

0.24

Corp

0.24

 yell

0.24

å·©

0.24

 premise

0.23

Activations Density 0.939%

Carroll or Carr

No Comments

No Known Activations