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

Not in Any Lists

No Comments

Negative Logits

 nonlinear

-0.27

çºªå§Ķ

-0.26

èĽĽ

-0.26

èĥ¡

-0.25

roys

-0.24

fq

-0.24

DMA

-0.24

 hype

-0.24

Ð²Ð¸Ñģ

-0.23

WORD

-0.23

POSITIVE LOGITS

è¿ĩåī©

0.29

è¾Ħ

0.29

tle

0.28

æľīæĹ¶åĢĻ

0.27

:///

0.26

mk

0.26

å¹´è½»

0.26

æľīæĹ¶

0.26

ForKey

0.26

åıĳéŁ³

0.26

Activations Density 0.002%

No Known Activations

This feature has no known activations.