INDEX

Explanations

since

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

å¹´ä»¥ä¸Ĭ

-0.31

åįģå¹´åīį

-0.28

frei

-0.28

å¹´åīį

-0.27

å¤ļå¹´åīį

-0.26

å·²ä¹ħ

-0.26

Ef

-0.25

æĹ©

-0.25

åĩłå¹´åīį

-0.25

Ø§Ø¯ÙĬØ©

-0.24

POSITIVE LOGITS

 then

0.32

 subsequently

0.31

éļıåĲİ

0.29

 subsequent

0.29

 Promo

0.27

çĦ¶åĲİ

0.27

then

0.27

.then

0.26

 thereafter

0.26

æŃ¤åĲİ

0.25

Activations Density 0.005%

since

No Comments

No Known Activations