INDEX

Explanations

charter

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_31.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.31.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ookie

-0.34

ħ§

-0.27

ated

-0.26

ETHOD

-0.25

TestMethod

-0.25

æģĲæħĮ

-0.25

styl

-0.24

 Stacy

-0.24

æħĭ

-0.24

 Encyclopedia

-0.24

POSITIVE LOGITS

ers

0.37

å»º

0.29

.twig

0.29

gr

0.29

builder

0.29

åħ±å»º

0.28

èĢħ

0.28

æ¸£

0.27

 seats

0.27

ing

0.27

Activations Density 0.005%

charter

No Comments

No Known Activations