INDEX

Explanations

programming-related terms and method declarations

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

testing code creation

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

NegativeButton

-0.49

koła

-0.49

minecraftforge

-0.47

 Chwiliwch

-0.44

__::

-0.43

 scrolls

-0.42

 cleverly

-0.41

 estekak

-0.41

cari

-0.40

 Sasaki

-0.40

POSITIVE LOGITS

WebElementEntity

0.57

StructEnd

0.55

 contextLoads

0.52

SpringBootTest

0.49

 Audiodateien

0.47

 '\\;'

0.47

ReusableCell

0.45

GEBURTSDATUM

0.45

 awakeFromNib

0.45

detectChanges

0.45

Activations Density 0.001%

programming-related terms and method declarations

testing code creation

No Comments

No Known Activations