INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_18.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.18.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 approach

-0.26

 demand

-0.26

åħ±åĲĮåĬªåĬĽ

-0.25

sil

-0.25

erman

-0.24

tax

-0.24

 intr

-0.24

loff

-0.24

è¿Ļæī¹

-0.23

law

-0.23

POSITIVE LOGITS

eya

0.29

 recru

0.28

 inflate

0.27

cio

0.24

ï¼ļ"

0.24

 thous

0.24

 Evalu

0.23

è¶³å¤Ł

0.23

ILING

0.23

@register

0.23

Activations Density 1.042%

No Known Activations

This feature has no known activations.