INDEX

Explanations

H

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_32.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.32.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

alt

-0.31

Point

-0.30

=>

-0.29

çĤ¹

-0.29

Ø·ÙĪ

-0.28

 Point

-0.26

=>'

-0.26

åŁĭ

-0.25

æ»´

-0.25

atories

-0.25

POSITIVE LOGITS

okit

0.29

ãģĭãģĳ

0.28

arding

0.27

.toolbox

0.26

ersh

0.26

 nests

0.26

äººæŃ»äº¡

0.25

onald

0.25

 washer

0.24

ä¸Ģéģĵ

0.24

Activations Density 0.011%

H

No Known Activations