INDEX

Explanations

say "al"

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_27.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.27.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

æĭ¢

-0.27

czy

-0.26

âĺħâĺħ

-0.26

äººåĿĩ

-0.26

ioned

-0.26

UNS

-0.25

URT

-0.24

izzare

-0.24

ili

-0.24

idge

-0.24

POSITIVE LOGITS

mult

0.28

ryptography

0.26

incident

0.25

å¹´ä»£

0.25

Resistance

0.25

-lfs

0.24

schools

0.24

ØŃØ§ÙĨ

0.24

 Resistance

0.24

å¿«è¦ģ

0.23

Activations Density 0.023%

say "al"

No Known Activations