INDEX

Explanations

target

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_31.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.31.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

å¤ļäºĨ

-0.28

èĤ©

-0.28

 BaseEntity

-0.27

aData

-0.27

ç§¯ç´¯äºĨ

-0.26

brace

-0.26

 Colony

-0.26

äºĨè§£æĽ´å¤ļ

-0.26

åľŃ

-0.26

çĽĽä¸ĸ

-0.25

POSITIVE LOGITS

not

0.36

not

0.30

 today

0.28

 groups

0.27

elig

0.27

not

0.27

 specific

0.27

 directly

0.26

 tonight

0.26

 holes

0.26

Activations Density 0.071%

target

No Known Activations