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

Ð¿Ð¾ÑĢ

-0.26

ewire

-0.26

makt

-0.25

anky

-0.25

 abst

-0.24

è¡ĮæĶ¿å¤Ħç½ļ

-0.24

ray

-0.24

 breeze

-0.23

bsites

-0.23

Ã±o

-0.23

POSITIVE LOGITS

åĬ¨æĢģ

0.30

predicted

0.25

 dynamic

0.25

Êº

0.25

 predicted

0.24

???

0.24

ÑĦÐ¸Ðº

0.24

ãģªãĤĬ

0.24

 thoughts

0.24

 Dynamic

0.23

Activations Density 0.009%

No Known Activations

This feature has no known activations.