INDEX

Explanations

punctuation marks and delimiters in code or data structures

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

javascript validation rules

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

tonode

-0.57

rsiniz

-0.41

dojo

-0.40

 katkıda

-0.40

 defaultstate

-0.39

doctype

-0.38

joda

-0.38

CUC

-0.37

 Macau

-0.37

aha

-0.37

POSITIVE LOGITS

存于互联网档案馆

0.63

Autowired

0.54

StoryboardSegue

0.54

AndroidJUnit

0.46

enterOuterAlt

0.45

Autoritní

0.43

 Autorisations

0.42

InjectMocks

0.42

ChromeDriver

0.42

writeFieldEnd

0.41

Activations Density 0.003%

punctuation marks and delimiters in code or data structures

javascript validation rules

No Comments

No Known Activations