INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_19.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.19.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ÑģÑĤÑĢÐ¾ÐµÐ½

-0.27

melon

-0.26

etas

-0.24

ä¸īä½įä¸Ģä½ĵ

-0.24

çĭŃçªĦ

-0.24

 BÃ¶l

-0.24

zelf

-0.23

è²Į

-0.23

etta

-0.23

PREFIX

-0.23

POSITIVE LOGITS

 norm

0.29

_scheme

0.28

 universe

0.27

æĪĴ

0.27

rep

0.26

 sets

0.26

 goal

0.25

 machine

0.25

 abund

0.25

æ¸ħéĨĴ

0.25

Activations Density 0.704%

No Known Activations

This feature has no known activations.