INDEX

Explanations

code and resource management

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_20.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.20.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

outcome

-0.27

 preserve

-0.26

 preserves

-0.25

æĭĽæĶ¶

-0.25

 assures

-0.25

 ensures

-0.25

 assure

-0.24

NES

-0.24

ä¸Ńå±±

-0.24

 retains

-0.23

POSITIVE LOGITS

cent

0.32

è¿¹

0.28

æĪı

0.27

ç±³å°Ķ

0.26

æĸĩåĮĸçļĦ

0.26

å®¢

0.26

 play

0.25

played

0.24

logen

0.24

åĬŀåħ¬æ¥¼

0.24

Activations Density 1.059%

code and resource management

No Comments

No Known Activations