INDEX

Explanations

scholar

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

äºĭå®ŀ

-0.27

åī¯å±Ģéķ¿

-0.26

Vet

-0.25

ä½°

-0.25

Pod

-0.24

éĥ¨åĪĨåĨħå®¹

-0.24

beit

-0.24

Pod

-0.23

inc

-0.23

 Orch

-0.23

POSITIVE LOGITS

é³Ħ

0.27

imately

0.26

é¾Ħ

0.25

è¦ģä¹Ī

0.25

 Grimm

0.24

 tongues

0.24

éķ¯

0.24

ile

0.23

ublic

0.23

åĪ°å¤Ħ

0.23

Activations Density 0.875%

scholar

No Comments

No Known Activations