INDEX

Explanations

expressing uncertainty

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 understanding

-0.38

Understanding

-0.35

 Understanding

-0.33

 understand

-0.31

çĲĨè§£

-0.30

 Understand

-0.28

æ·±åħ¥äºĨè§£

-0.27

 understood

-0.27

 misunderstood

-0.27

 study

-0.26

POSITIVE LOGITS

 otherwise

0.35

otherwise

0.32

 OTHERWISE

0.31

 Otherwise

0.29

onica

0.28

å¾Ĵ

0.27

åĲ¦åĪĻ

0.26

ONE

0.26

#else

0.25

 asiat

0.25

Activations Density 0.002%

expressing uncertainty

No Comments

No Known Activations