INDEX

Explanations

sections related to structured data and methods in programming

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

technical documentation and code snippets

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

stateProvider

-0.52

 AspNetCore

-0.48

 onCreateView

-0.43

 BrowserModule

-0.40

DrawerToggle

-0.36

MeasureSpec

-0.35

########.

-0.34

SourceChecksum

-0.34

SBATCH

-0.34

 CommonModule

-0.33

POSITIVE LOGITS

uxxxx

0.64

parsedMessage

0.60

XmlAccessType

0.60

 kaarangay

0.59

rungsseite

0.50

RetentionPolicy

0.49

 MessageBoxIcon

0.48

ьаж

0.48

kloped

0.48

 CURIAM

0.48

Activations Density 0.242%

sections related to structured data and methods in programming

technical documentation and code snippets

No Comments

No Known Activations