INDEX

Explanations

terms related to content types and custom elements in a specific framework or system

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

custom element registration

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

 Signalez

-0.56

awtextra

-0.55

AddTagHelper

-0.51

UrlResolution

-0.51

MemoryWarning

-0.51

 يتيمه

-0.48

pexpr

-0.48

ntomas

-0.47

InjectAttribute

-0.47

JspWriter

-0.47

POSITIVE LOGITS

Хьажоргаш

0.54

ollectionView

0.44

StackNavigator

0.42

 cons

0.40

 htmlFor

0.40

DOCTYPE

0.38

sequelize

0.37

WebServlet

0.37

onomies

0.36

يميديا

0.35

Activations Density 0.014%

terms related to content types and custom elements in a specific framework or system

custom element registration

No Comments

No Known Activations