INDEX

Explanations

faculty

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

åĦ¿å¥³

-0.26

åŁİå¸Ĥç¾¤

-0.26

èģĮå·¥

-0.26

æ°ĳèŃ¦

-0.25

Cities

-0.25

_Manager

-0.25

_manager

-0.25

 Capitals

-0.24

èŃ¦å¯Ł

-0.24

FRAME

-0.24

POSITIVE LOGITS

 engagement

0.32

Pid

0.30

 engaged

0.28

add

0.28

fu

0.28

çĽ®çļĦ

0.28

eng

0.28

å½Ń

0.27

odyn

0.27

ivalent

0.27

Activations Density 0.136%

faculty

No Known Activations