INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 concat

-0.27

æ¸´æľĽ

-0.26

 prep

-0.26

 indication

-0.25

 consenting

-0.25

 consistently

-0.25

leness

-0.24

CONS

-0.24

pty

-0.24

sters

-0.24

POSITIVE LOGITS

ongyang

0.29

è§Ĥä¼Ĺ

0.26

LinkId

0.25

é¡¶

0.25

_nat

0.25

è§Ĥ

0.24

.platform

0.24

æīĭæ³ķ

0.24

]|[

0.24

æĶ¯çº¿

0.24

Activations Density 0.010%

No Known Activations

This feature has no known activations.