INDEX

Explanations

No Explanations Found

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

Negative Logits

åĪº

-0.27

å«ª

-0.25

Me

-0.25

ç¾¤

-0.25

ãģ©ãĤĮ

-0.25

 needle

-0.24

å¯¸

-0.24

-Men

-0.24

à¹ģà¸«à¸¥

-0.23

 pict

-0.23

POSITIVE LOGITS

olation

0.27

edException

0.26

CHAT

0.26

olist

0.25

çĶŁæĢķ

0.24

æİ¥è§¦åĪ°

0.24

 shallow

0.24

orthand

0.24

åĪ©äºİ

0.23

olo

0.23

Activations Density 43.450%

No Known Activations

This feature has no known activations.