INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_19.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.19.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ÑĢÐ°

-0.27

æīĢä»¥æĪĳä»¬

-0.26

×ķ×ª

-0.26

å®Ī

-0.25

æ¶¯

-0.25

 chai

-0.24

æ´¾åĩº

-0.24

rais

-0.24

æ´¾

-0.23

anded

-0.23

POSITIVE LOGITS

iko

0.26

icum

0.25

çľ¼åīį

0.25

anker

0.24

lessly

0.24

ually

0.24

 Natal

0.24

ern

0.24

atically

0.23

ika

0.23

Activations Density 0.055%

No Known Activations

This feature has no known activations.