INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_31.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.31.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

rosso

-0.28

éĻ¡

-0.28

gings

-0.26

åĽ¢

-0.26

 unst

-0.25

å¤ķ

-0.25

_datasets

-0.25

agent

-0.24

Broadcast

-0.24

uel

-0.24

POSITIVE LOGITS

ç¬³

0.25

.rem

0.25

ä¾

0.25

 consolidated

0.25

/rem

0.25

ç¬¬ä¸ĢèĬĤ

0.24

 Lith

0.24

educt

0.24

WebKit

0.23

è¶´åľ¨

0.23

Activations Density 1.379%

No Known Activations

This feature has no known activations.