INDEX

Explanations

URLs and code

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

flate

-0.30

æľºæ²¹

-0.27

 stopwatch

-0.27

 intuit

-0.26

angler

-0.26

çİ°åŃĺ

-0.26

crement

-0.26

erp

-0.25

bsp

-0.25

ucene

-0.25

POSITIVE LOGITS

æģ°å½ĵ

0.31

-heading

0.26

è®©æ¶Īè´¹èĢħ

0.26

kes

0.26

LOC

0.25

åŃ¤ç«ĭ

0.25

æīįèĥ½

0.25

 convo

0.25

æīįæĺ¯

0.24

æ³Ĭ

0.24

Activations Density 0.002%

URLs and code

No Comments

No Known Activations