INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_19.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.19.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

åı·

-0.25

ngu

-0.24

çĮ®

-0.24

é«ĺåľ°

-0.24

éº½

-0.24

 follower

-0.23

Vu

-0.23

Geo

-0.23

å¹´çĶ±

-0.23

èī°éļ¾

-0.23

POSITIVE LOGITS

geries

0.29

 garn

0.26

 ÑĢÐ°Ð·Ð²Ð¸Ð²Ð°

0.25

tems

0.25

obble

0.25

åĪ«è¯´

0.24

åĵĦ

0.24

apur

0.24

uges

0.24

enders

0.24

Activations Density 0.912%

No Known Activations

This feature has no known activations.