INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_24.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.24.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 kino

-0.25

 strm

-0.25

 Faker

-0.25

 Nack

-0.25

 elem

-0.24

éĤ£ä»½

-0.23

 motivational

-0.23

 saddle

-0.23

mouseleave

-0.23

 undermin

-0.23

POSITIVE LOGITS

ugi

0.27

fat

0.26

Object

0.25

fat

0.25

Fat

0.24

overe

0.24

|=

0.24

æī§

0.24

Fat

0.24

lus

0.23

Activations Density 0.126%

No Known Activations

This feature has no known activations.