INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_22.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.22.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

exp

-0.27

gas

-0.26

çĸĶ

-0.26

Gas

-0.25

 moreover

-0.25

fam

-0.25

exp

-0.25

èįĳ

-0.24

ä¸įåıĬ

-0.24

Hyp

-0.24

POSITIVE LOGITS

æĬ½è±¡

0.29

à¸±à¸Ķ

0.27

olucion

0.26

 abstraction

0.26

 Sergeant

0.25

Ã±e

0.24

 abstract

0.24

Ã¶n

0.24

atus

0.24

Æ°á»Łng

0.24

Activations Density 0.004%

No Known Activations

This feature has no known activations.