INDEX

Explanations

keywords related to programming actions and functions within applications

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

application run loop

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

LabelTagHelper

-0.75

httphttps

-0.75

tonode

-0.73

WriteTagHelper

-0.71

 فريبيس

-0.69

 surla

-0.69

ConstraintMaker

-0.65

featureID

-0.61

fromnode

-0.61

AddTagHelper

-0.60

POSITIVE LOGITS

drawSprites

0.42

mainloop

0.42

 Substanz

0.41

SaveChangesAsync

0.37

 duração

0.36

 catálogo

0.36

 invitados

0.35

 judiciales

0.35

 excelencia

0.34

 convocatoria

0.34

Activations Density 0.007%

keywords related to programming actions and functions within applications

application run loop

No Comments

No Known Activations