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

å°½

-0.29

PING

-0.28

 Sutton

-0.26

 applicationContext

-0.25

 Damien

-0.25

éĩĳèŀįæľįåĬ¡

-0.24

èľķ

-0.24

bach

-0.24

(internal

-0.24

 pads

-0.23

POSITIVE LOGITS

Award

0.28

å¤§å¥ĸ

0.24

+'.

0.24

FORMAT

0.24

 herself

0.24

format

0.24

isms

0.23

-choice

0.23

 floors

0.23

_mono

0.23

Activations Density 0.002%

No Known Activations

This feature has no known activations.