INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_18.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.18.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 strike

-0.28

 Sized

-0.24

 favoured

-0.24

á¾¶

-0.24

nte

-0.23

ÙĤÙĪØ¯

-0.23

misc

-0.23

noinspection

-0.23

ç°¡

-0.23

 striking

-0.23

POSITIVE LOGITS

oky

0.28

ç´§ç´§

0.27

imen

0.25

andalone

0.25

ê·ł

0.25

ennis

0.24

empl

0.24

istra

0.24

åŀ¸

0.24

ãĤ«ãĥª

0.24

Activations Density 0.912%

No Known Activations

This feature has no known activations.