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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 campground

-0.26

angered

-0.26

 withheld

-0.26

 heals

-0.26

ters

-0.25

.newBuilder

-0.25

-labelledby

-0.25

rsp

-0.25

ÑĥÐ½

-0.24

signals

-0.24

POSITIVE LOGITS

illary

0.27

ne

0.25

flat

0.25

ä¸»å¯¼

0.24

CTIONS

0.24

attr

0.23

 flat

0.23

 bande

0.23

APP

0.23

 grup

0.23

Activations Density 0.029%

No Known Activations

This feature has no known activations.