INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_1.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.1.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

_filters

-0.25

çļĦè¶ĭåĬ¿

-0.24

è¶ĭåĬ¿

-0.24

bildung

-0.24

olla

-0.23

imations

-0.23

InputLabel

-0.23

plemented

-0.23

 mist

-0.23

 Mist

-0.23

POSITIVE LOGITS

çµĲ

0.29

å±±åİ¿

0.28

 ç»ĵ

0.26

ç»ĵ

0.26

park

0.26

å¹¿

0.25

éĹªåħī

0.25

à¸£à¸°

0.25

 manip

0.24

nect

0.24

Activations Density 6.089%

No Known Activations

This feature has no known activations.