INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_22.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.22.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 lied

-0.29

 Dedicated

-0.27

Founded

-0.26

è´Ł

-0.25

å¤§åıĳ

-0.25

Lie

-0.24

atism

-0.24

 Terms

-0.23

 conferred

-0.23

lie

-0.23

POSITIVE LOGITS

 safeguards

0.27

çĶµåķĨ

0.27

alist

0.26

CP

0.24

ct

0.24

ington

0.24

Pair

0.24

scribe

0.24

unce

0.24

alon

0.23

Activations Density 0.000%

No Known Activations

This feature has no known activations.