INDEX

Explanations

period

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_22.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.22.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

æĺİæĺİ

-0.29

åı¦ä¸ĢåįĬ

-0.29

ä»½

-0.28

@a

-0.27

ä¸ĢéĿ¢

-0.26

ä¸Ģéģį

-0.26

 skim

-0.25

 Duplicate

-0.25

 repetition

-0.25

 perm

-0.25

POSITIVE LOGITS

 colony

0.27

ä¿±ä¹Ĳéĥ¨

0.27

 colonies

0.26

 club

0.25

[len

0.25

iv

0.25

iva

0.25

 Colony

0.24

ages

0.24

iben

0.24

Activations Density 0.012%

period

No Known Activations