INDEX

Explanations

CB

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

Not in Any Lists

No Comments

Negative Logits

.drawText

-0.26

æĬķåħ¥åĪ°

-0.25

åĲ¯èĴĻ

-0.25

 multis

-0.24

&id

-0.24

çĹĴ

-0.24

æ¶²

-0.24

æ±Ĥ

-0.23

æ¡ĥ

-0.23

ä¼¸æīĭ

-0.23

POSITIVE LOGITS

sei

0.32

se

0.30

ssÃ©

0.28

quÃŃ

0.27

 Apprentice

0.26

sez

0.26

ije

0.25

ä¸¥

0.25

Gal

0.25

 Chain

0.25

Activations Density 0.001%

CB

No Comments

No Known Activations