INDEX

Explanations

sense

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_31.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.31.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ç²Ĺç³Ļ

-0.27

 ener

-0.27

çļĦèĥ½éĩı

-0.26

rough

-0.25

 Energ

-0.25

 energetic

-0.25

çī©çĲĨåŃ¦

-0.25

 earns

-0.24

rej

-0.24

ãģŀ

-0.24

POSITIVE LOGITS

 sense

0.84

sense

0.71

 Sense

0.69

Sense

0.62

 sentido

0.48

 senses

0.40

SEN

0.39

æĦŁ

0.36

sen

0.32

ENSE

0.32

Activations Density 1.097%

sense

No Comments

No Known Activations