INDEX

Explanations

programming syntax and method declarations

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

java Object String

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.24.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

GOTREF

-0.71

webElementXpaths

-0.61

 pulumi

-0.57

InjectAttribute

-0.56

 unknownFields

-0.54

 concurrency

-0.53

ArrowToggle

-0.53

 noDo

-0.52

enterOuterAlt

-0.52

σθαι

-0.50

POSITIVE LOGITS

 ujednoznacz

0.41

 disambiguazione

0.40

PathVariable

0.38

 neuem

0.35

Einzelnachweise

0.34

Ljava

0.34

antMatchers

0.34

تقاوى

0.34

 konkrét

0.34

 solche

0.33

Activations Density 0.006%

programming syntax and method declarations

java Object String

No Comments

No Known Activations