INDEX

Explanations

to

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_20.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.20.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

-trade

-0.30

idy

-0.28

ÑĢÑĥÑĪ

-0.25

-shop

-0.24

thalm

-0.24

TURN

-0.23

roots

-0.23

ä½ĵ

-0.23

æĶ¿

-0.23

à¹Īà¸²à¸§

-0.23

POSITIVE LOGITS

nt

0.29

å¹¶éĢļè¿ĩ

0.26

oho

0.24

æĬ½çĥŁ

0.24

 suggestive

0.24

.exports

0.24

æĪ´çĿĢ

0.24

Endpoint

0.23

hw

0.23

æīĵéĢłåĩº

0.23

Activations Density 20.088%

to

No Known Activations