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

Negative Logits

æīĵãģ¡

-0.30

è¢«æīĵ

-0.27

erp

-0.24

(Code

-0.24

 Code

-0.24

æĬĽ

-0.24

å¸ļ

-0.23

æ±¡æŁĵçī©

-0.23

recht

-0.23

 sprink

-0.23

POSITIVE LOGITS

irie

0.27

oil

0.24

å»Ĭ

0.23

heel

0.23

cil

0.23

ìĶ¨

0.23

ocht

0.23

Cheap

0.23

cheap

0.23

exus

0.23

Activations Density 1.844%

No Known Activations

This feature has no known activations.