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

Not in Any Lists

No Comments

Negative Logits

ÑģÐµÑĤ

-0.28

((_

-0.26

è¥¿

-0.26

æ·¡

-0.25

éħŀ

-0.24

 Programs

-0.24

éļĭ

-0.24

ÑĳÑĢ

-0.24

Pty

-0.24

-Mart

-0.23

POSITIVE LOGITS

 jednoc

0.26

iration

0.25

 requis

0.25

oru

0.24

eor

0.23

 centrally

0.23

confirmation

0.23

monic

0.23

éģĳ

0.23

ç±³é¥Ń

0.23

Activations Density 2.166%

No Known Activations

This feature has no known activations.