INDEX

Explanations

expressions indicating actions or interactions in code or systems development

oai_token-act-pair · gpt-4o-mini Triggered by @bot

Text after commas, periods, or other punctuation

np_acts-logits-general · gemini-2.0-flash

medical technology

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

躇

-0.37

 gerecht

-0.37

mau

-0.37

oporosis

-0.36

\{\\

-0.35

schrank

-0.34

zweck

-0.33

 baik

-0.32

bato

-0.32

SharedCtor

-0.32

POSITIVE LOGITS

 betweenstory

0.53

CPtr

0.48

 estekak

0.48

Captor

0.46

 الرياضيه

0.44

 виправивши

0.44

 defStyleAttr

0.43

 noDo

0.42

fjspx

0.42

SourceChecksum

0.41

Activations Density 0.699%

expressions indicating actions or interactions in code or systems development

Text after commas, periods, or other punctuation

medical technology

No Comments

No Known Activations

expressions indicating actions or interactions in code or systems development

Text after commas, periods, or other punctuation

medical technology

No Comments

No Known Activations