INDEX

Explanations

of

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_32.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.32.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

ulative

-0.30

æĶ¶

-0.28

plemented

-0.28

æįİ

-0.27

ìłĪ

-0.27

.setMax

-0.25

.MaxValue

-0.25

-automatic

-0.25

å¿«æĬ¥

-0.25

éħ¸

-0.25

POSITIVE LOGITS

 humans

0.34

DNA

0.32

DNA

0.29

Earth

0.28

 Humans

0.27

rat

0.27

äººç±»

0.27

Humans

0.26

ç¾½æ¯Ľ

0.26

Ã¶n

0.26

Activations Density 0.001%

of

No Known Activations