INDEX

Explanations

Resistance and security

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

æĻ®

-0.27

opsis

-0.25

.slim

-0.25

 Filip

-0.25

olders

-0.25

blo

-0.24

blo

-0.24

ometry

-0.24

åį´æ²¡æľī

-0.24

lyph

-0.24

POSITIVE LOGITS

trib

0.29

èĬĤæ°´

0.27

tube

0.27

æ¿Ģ

0.26

Utc

0.26

 Anth

0.25

uctive

0.25

critical

0.25

ija

0.25

ekt

0.24

Activations Density 0.001%

Resistance and security

No Known Activations