EL
Eriva Labs

Lessons Brought to Life

Don't just read, interact with the lessons. Step inside the concepts with visual diagrams, active simulations, and interactive models built to give you real intuition.

Gradient Descent SimulatorLoss Surface f(x) and Tangent Slope df/dx
Weight(x)
0.47
Gradient(df/dx)
0.00
-2.20.0+2.2High LossLow Loss
Negative DomainPositive Domain
Explore the Library
Eriva LibraryThe Showroom

Lessons you can play with.

Traditional textbooks expect you to construct complex ideas and concepts entirely in your head. Eriva Library aims to bridge this gap by combining the academic rigor of classical textbooks with fully interactive playgrounds, building intuition for the concepts you explore.

Adjust parameters, run interactive models, and watch complex systems adapt instantly as you read.

Tactile Visual ModelsInteractive Controls
Parameter w2.45
Geometric Foundations: Part 5
REPRESENTATION REGIME
Distributed (Dense) - ℝᴰ▼
SEMANTIC CORRELATION
LIVE CALCULUS
⟨v_Fr, v_It⟩ =0.230
d_2(v_Fr, v_It) =1.241
1. Mathematical Definition
Let V = {w₁, w₂, ..., w_v} be a finite set of discrete linguistic tokens. We define the one-hot encoding function mapping each token to a unique standard basis vector...
σ(wx + b)
Standard Activation
Eriva LibraryThe Blueprint

Theory in motion

Textbooks provide deep rigor, while visual simulations build immediate intuition. By nesting interactivity directly inside the flow of the text, we bring both worlds together—allowing you to explore concepts and build intuition all in one place.

Read, interact, and explore without leaving the text.

Academic RigorInline Simulations
Eriva LensThe Workspace Companion

Your AI Research Companion

Learning doesn't stop at the end of a chapter. Eriva Lens is a workspace companion built to help you navigate dense academic literature when you need to go beyond the curated curriculum.

Use contextual search to scan our database of over 100k papers, or upload your own documents to get instant, structured technical breakdowns, define technical terms on the fly, and chat contextually with your papers.

Semantic SearchContextual Chat
PDF
Ingesting75%
Technical Breakdown
Ask AI
How does self-attention model long-range dependencies?
It connects all positions in a sequence directly with a constant number of operations. This bypasses sequential step-by-step paths, making it much easier to learn relationships between distant words.

Ready to Step Inside the Lessons?

Browse the complete collection of interactive textbooks.

Explore the Library