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

 STREAM

-0.26

èĵĵ

-0.25

 repay

-0.24

 util

-0.24

ä¸ĭåİ»

-0.24

æ»ĺ

-0.24

 studios

-0.24

éºĭ

-0.23

 Seks

-0.23

_stream

-0.23

POSITIVE LOGITS

esters

0.26

à¸§à¸ģ

0.26

èī¯ãģı

0.25

å°ĳè§ģ

0.24

æł¹

0.24

 Said

0.24

arial

0.24

uv

0.24

çľĭå¥½

0.23

Responder

0.23

Activations Density 12.650%

No Known Activations

This feature has no known activations.