INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_25.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.25.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

elor

-0.31

åĽŃ

-0.30

pher

-0.29

 conjug

-0.29

gard

-0.27

 Garden

-0.27

 hell

-0.27

è®²ç©¶

-0.26

è§ĦèĮĥ

-0.26

 prefix

-0.26

POSITIVE LOGITS

SENS

0.28

aser

0.27

 contrario

0.25

cuts

0.25

atism

0.24

ä¸Ģå°ģä¿¡

0.24

erner

0.24

:params

0.23

èħ¾

0.23

quam

0.23

Activations Density 0.179%

No Known Activations

This feature has no known activations.