INDEX

Explanations

defense

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_7.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.7.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

apel

-0.29

imus

-0.28

 Ð±Ð¸Ð»

-0.26

ä»Ĺ

-0.26

.ie

-0.25

 vess

-0.25

æĥ³åĪ°

-0.25

 delt

-0.25

çĽ®åħī

-0.24

èĢĮéĿŀ

-0.24

POSITIVE LOGITS

éĿ¢

0.30

 coinc

0.29

ÙģØ¶

0.28

 face

0.27

ä¹ĸ

0.27

ConnectionString

0.26

æºľ

0.26

è·Ŀ

0.25

çĨĺ

0.25

losure

0.25

Activations Density 0.002%

defense

No Comments

No Known Activations