INDEX

Explanations

Various topics

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

æī©å±ķ

-0.41

 expansion

-0.40

æī©åħħ

-0.37

æĭĵå±ķ

-0.36

 enhancement

-0.35

åįĩçº§

-0.34

Expansion

-0.33

 Expansion

-0.33

 extension

-0.32

 ench

-0.32

POSITIVE LOGITS

éĺ²æŃ¢

0.29

åĨįæĿ¥

0.29

mant

0.28

æĽ´ä½İ

0.26

epy

0.26

Ens

0.26

icana

0.25

åĨįä¹Łä¸į

0.25

qx

0.25

Suc

0.25

Activations Density 0.000%

Various topics

No Comments

No Known Activations