INDEX

Explanations

Proper nouns and team names

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.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

èģĶç³»

-0.30

æµģçķħ

-0.28

éģµ

-0.28

ä¸ŃæľŁ

-0.28

åĽ½äºº

-0.28

.getB

-0.27

æ¯ķ

-0.27

opus

-0.26

empre

-0.26

æĻļæľŁ

-0.25

POSITIVE LOGITS

 Emanuel

0.26

å¤§è¡Ĺ

0.25

ariate

0.25

 stake

0.25

 IPPROTO

0.24

arend

0.24

 Institutional

0.24

ativity

0.24

adata

0.24

 Dimension

0.23

Activations Density 0.008%

Proper nouns and team names

No Comments

No Known Activations