🏠 Beranda
Benchmark
📊 Semua Benchmark 🦖 Dinosaurus v1 🦖 Dinosaurus v2 ✅ Aplikasi To-Do List 🎨 Halaman Bebas Kreatif 🎯 FSACB - Showcase Utama 🌍 Benchmark Terjemahan
Model
🏆 Top 10 Model 🆓 Model Gratis 📋 Semua Model ⚙️ Kilo Code
Sumber Daya
💬 Perpustakaan Prompt 📖 Glosarium AI 🔗 Tautan Berguna

Glosarium AI

Kamus lengkap Kecerdasan Buatan

162
kategori
2.032
subkategori
23.060
istilah
📖
istilah

Binary Mask

Matrix containing only 0 and 1 values where 1 indicates positions to keep and 0 those to mask, generally applied through element-wise multiplication before or after the attention softmax.

📖
istilah

Triangular Causal Mask

Triangular matrix structure where elements above the diagonal are masked, creating strict temporal dependency in transformer models for sequential tasks.

📖
istilah

Variable Length Mask

Dynamic mask that adapts to variable sequence lengths in a batch, optimizing computation by ignoring irrelevant positions while preserving batch alignment.

📖
istilah

Key Padding Mask

Specific mask applied to keys in the attention mechanism to prevent padding tokens from influencing attention scores, typically added before the softmax operation.

📖
istilah

Query Mask

Mask applied to queries to restrict which positions can perform attention queries, used in specialized architectures requiring granular control of interactions.

📖
istilah

Value Mask

Mask applied to values after attention computation to filter out undesirable contributions, enabling fine post-attention control of output representations.

📖
istilah

Attention Weight Masking

Technique consisting of applying a mask directly to attention weights after softmax to force certain contributions to zero, offering explicit control over information pathways.

📖
istilah

Softmax Mask

Mask applied by adding a large negative value (typically -inf) to attention scores before softmax, ensuring that masked positions receive a probability close to zero.

📖
istilah

Logit Mask

Masque appliqué au niveau des logits (scores d'attention bruts) pour exclure certaines interactions avant la normalisation softmax, préservant la distribution mathématique des scores valides.

🔍

Tidak ada hasil ditemukan