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

thora

-0.26

ynchronized

-0.26

urgeon

-0.26

(Py

-0.25

essian

-0.25

åĲĦä¸ªçİ¯èĬĤ

-0.23

è²Į

-0.23

nameof

-0.23

itchens

-0.23

atos

-0.23

POSITIVE LOGITS

abs

0.28

 affirm

0.26

cars

0.26

contra

0.26

es

0.25

care

0.25

 corres

0.25

oy

0.25

 explic

0.25

 distinguished

0.25

Activations Density 2.870%

No Known Activations

This feature has no known activations.