INDEX

Explanations

almost

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_24.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.24.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

åľ°ä¸ĭ

-0.26

 Plenty

-0.26

 Numerous

-0.25

 mÅ©

-0.24

 Ð¿Ð»Ð¾Ñī

-0.24

 Various

-0.24

)[-

-0.24

 Exiting

-0.23

Numer

-0.23

 calf

-0.23

POSITIVE LOGITS

ERT

0.30

PERT

0.27

cert

0.27

omens

0.26

USIC

0.25

 scala

0.24

áº¡ch

0.24

JECT

0.23

scar

0.23

 strat

0.23

Activations Density 0.043%

almost

No Comments

No Known Activations