INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_18.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.18.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ê·¼

-0.27

à¹Ģà¸Ħà¸£

-0.27

mie

-0.26

 flop

-0.26

æĺŁåħī

-0.25

Ļ¶

-0.24

erta

-0.24

ERO

-0.24

 sing

-0.24

erate

-0.23

POSITIVE LOGITS

 *)__

0.24

èĽŃ

0.24

lea

0.23

ledon

0.23

 secara

0.23

å¥¹è¯´

0.23

azar

0.23

 cool

0.23

 Regarding

0.23

 cooled

0.23

Activations Density 0.034%

No Known Activations

This feature has no known activations.