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

shr

-0.26

æĬĢæľ¯åĪĽæĸ°

-0.25

ÅĽci

-0.25

æį»

-0.25

 testim

-0.24

)','

-0.24

 canadian

-0.24

CHK

-0.24

 Nimbus

-0.23

dm

-0.23

POSITIVE LOGITS

rette

0.28

ä¸ĢæĹ¶

0.26

.links

0.26

“He

0.26

opoulos

0.26

ä¹ħ

0.25

ifiable

0.25

eya

0.25

itate

0.24

 link

0.24

Activations Density 0.001%

No Known Activations

This feature has no known activations.