Upload all models and assets for dsb (20251201)

.gitattributes

@@ -33,3 +33,9 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+visualizations/embedding_similarity.png filter=lfs diff=lfs merge=lfs -text
+visualizations/performance_dashboard.png filter=lfs diff=lfs merge=lfs -text
+visualizations/position_encoding_comparison.png filter=lfs diff=lfs merge=lfs -text
+visualizations/tsne_sentences.png filter=lfs diff=lfs merge=lfs -text
+visualizations/tsne_words.png filter=lfs diff=lfs merge=lfs -text
+visualizations/zipf_law.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,558 @@
+---
+language: dsb
+language_name: DSB
+language_family: slavic_west
+tags:
+  - wikilangs
+  - nlp
+  - tokenizer
+  - embeddings
+  - n-gram
+  - markov
+  - wikipedia
+  - monolingual
+  - family-slavic_west
+license: mit
+library_name: wikilangs
+pipeline_tag: feature-extraction
+datasets:
+  - omarkamali/wikipedia-monthly
+dataset_info:
+  name: wikipedia-monthly
+  description: Monthly snapshots of Wikipedia articles across 300+ languages
+metrics:
+  - name: best_compression_ratio
+    type: compression
+    value: 4.294
+  - name: best_isotropy
+    type: isotropy
+    value: 0.8252
+  - name: vocabulary_size
+    type: vocab
+    value: 32309
+generated: 2025-12-30
+---
+
+# DSB - Wikilangs Models
+## Comprehensive Research Report & Full Ablation Study
+
+This repository contains NLP models trained and evaluated by Wikilangs, specifically on **DSB** Wikipedia data.
+We analyze tokenizers, n-gram models, Markov chains, vocabulary statistics, and word embeddings.
+
+## 📋 Repository Contents
+
+### Models & Assets
+
+- Tokenizers (8k, 16k, 32k, 64k)
+- N-gram models (2, 3, 4-gram)
+- Markov chains (context of 1, 2, 3 and 4)
+- Subword N-gram and Markov chains
+- Embeddings in various sizes and dimensions
+- Language Vocabulary
+- Language Statistics
+![Performance Dashboard](visualizations/performance_dashboard.png)
+
+### Analysis and Evaluation
+
+- [1. Tokenizer Evaluation](#1-tokenizer-evaluation)
+- [2. N-gram Model Evaluation](#2-n-gram-model-evaluation)
+- [3. Markov Chain Evaluation](#3-markov-chain-evaluation)
+- [4. Vocabulary Analysis](#4-vocabulary-analysis)
+- [5. Word Embeddings Evaluation](#5-word-embeddings-evaluation)
+- [6. Summary & Recommendations](#6-summary--recommendations)
+- [Metrics Glossary](#appendix-metrics-glossary--interpretation-guide)
+- [Visualizations Index](#visualizations-index)
+
+---
+## 1. Tokenizer Evaluation
+
+![Tokenizer Compression](visualizations/tokenizer_compression.png)
+
+### Results
+
+| Vocab Size | Compression | Avg Token Len | UNK Rate | Total Tokens |
+|------------|-------------|---------------|----------|--------------|
+| **8k** | 3.271x | 3.18 | 0.0992% | 351,825 |
+| **16k** | 3.624x | 3.52 | 0.1099% | 317,561 |
+| **32k** | 3.965x | 3.86 | 0.1202% | 290,267 |
+| **64k** | 4.294x 🏆 | 4.18 | 0.1302% | 268,009 |
+
+### Tokenization Examples
+
+Below are sample sentences tokenized with each vocabulary size:
+
+**Sample 1:** `Złocieniec jo město w Pólskej, w pódwjacoropomorskem wójwodstwje. Lažy w Pomorsk...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁zło cie niec ▁jo ▁město ▁w ▁pólskej , ▁w ▁pódwjacoro ... (+13 more)` | 23 |
+| 16k | `▁zło cie niec ▁jo ▁město ▁w ▁pólskej , ▁w ▁pódwjacoro ... (+13 more)` | 23 |
+| 32k | `▁zło cie niec ▁jo ▁město ▁w ▁pólskej , ▁w ▁pódwjacoro ... (+13 more)` | 23 |
+| 64k | `▁zło cieniec ▁jo ▁město ▁w ▁pólskej , ▁w ▁pódwjacoro pomorskem ... (+12 more)` | 22 |
+
+**Sample 2:** `Nowy Dwór Królewski jo wjas w Pólskej.
+
+Kurów lažy mjazy městoma Chelmno a Torun...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁nowy ▁dwór ▁k ró le wski ▁jo ▁wjas ▁w ▁pólskej ... (+21 more)` | 31 |
+| 16k | `▁nowy ▁dwór ▁kró le wski ▁jo ▁wjas ▁w ▁pólskej . ... (+19 more)` | 29 |
+| 32k | `▁nowy ▁dwór ▁króle wski ▁jo ▁wjas ▁w ▁pólskej . ▁kurów ... (+16 more)` | 26 |
+| 64k | `▁nowy ▁dwór ▁króle wski ▁jo ▁wjas ▁w ▁pólskej . ▁kurów ... (+15 more)` | 25 |
+
+**Sample 3:** `Janusz Gajos (* 23. septembra 1939) jo pólski grajaŕ, fotograf a pedagog.
+thumb
+...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁jan usz ▁ga jo s ▁(* ▁ 2 3 . ... (+33 more)` | 43 |
+| 16k | `▁jan usz ▁ga jo s ▁(* ▁ 2 3 . ... (+32 more)` | 42 |
+| 32k | `▁janusz ▁ga jos ▁(* ▁ 2 3 . ▁septembra ▁ ... (+30 more)` | 40 |
+| 64k | `▁janusz ▁gajos ▁(* ▁ 2 3 . ▁septembra ▁ 1 ... (+29 more)` | 39 |
+
+
+### Key Findings
+
+- **Best Compression:** 64k achieves 4.294x compression
+- **Lowest UNK Rate:** 8k with 0.0992% unknown tokens
+- **Trade-off:** Larger vocabularies improve compression but increase model size
+- **Recommendation:** 32k vocabulary provides optimal balance for production use
+
+---
+## 2. N-gram Model Evaluation
+
+![N-gram Perplexity](visualizations/ngram_perplexity.png)
+
+![N-gram Coverage](visualizations/ngram_coverage.png)
+
+### Results
+
+| N-gram | Perplexity | Entropy | Unique N-grams | Top-100 Coverage | Top-1000 Coverage |
+|--------|------------|---------|----------------|------------------|-------------------|
+| **2-gram** | 5,225 🏆 | 12.35 | 14,685 | 19.8% | 49.5% |
+| **2-gram** | 508 🏆 | 8.99 | 4,003 | 51.9% | 96.7% |
+| **3-gram** | 10,031 | 13.29 | 21,515 | 13.6% | 36.7% |
+| **3-gram** | 4,632 | 12.18 | 29,259 | 18.2% | 55.2% |
+| **4-gram** | 19,104 | 14.22 | 37,295 | 10.0% | 27.6% |
+| **4-gram** | 23,026 | 14.49 | 127,625 | 9.2% | 29.7% |
+
+### Top 5 N-grams by Size
+
+**2-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `kategorija :` | 7,667 |
+| 2 | `) jo` | 2,004 |
+| 3 | `) .` | 1,655 |
+| 4 | `( *` | 1,443 |
+| 5 | `) ,` | 1,402 |
+
+**3-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `kategorija : sedlišćo` | 991 |
+| 2 | `: sedlišćo w` | 874 |
+| 3 | `. kategorija :` | 760 |
+| 4 | `kategorija : roź` | 672 |
+| 5 | `: roź .` | 672 |
+
+**4-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `kategorija : sedlišćo w` | 874 |
+| 2 | `kategorija : roź .` | 672 |
+| 3 | `kategorija : wum .` | 395 |
+| 4 | `978 - 3 -` | 367 |
+| 5 | `isbn 978 - 3` | 359 |
+
+
+### Key Findings
+
+- **Best Perplexity:** 2-gram with 508
+- **Entropy Trend:** Decreases with larger n-grams (more predictable)
+- **Coverage:** Top-1000 patterns cover ~30% of corpus
+- **Recommendation:** 4-gram or 5-gram for best predictive performance
+
+---
+## 3. Markov Chain Evaluation
+
+![Markov Entropy](visualizations/markov_entropy.png)
+
+![Markov Branching](visualizations/markov_branching.png)
+
+### Results
+
+| Context | Avg Entropy | Perplexity | Branching Factor | Unique Contexts | Predictability |
+|---------|-------------|------------|------------------|-----------------|----------------|
+| **1** | 0.5742 | 1.489 | 3.58 | 83,705 | 42.6% |
+| **1** | 1.1222 | 2.177 | 9.59 | 1,062 | 0.0% |
+| **2** | 0.2165 | 1.162 | 1.49 | 299,024 | 78.3% |
+| **2** | 1.0246 | 2.034 | 5.91 | 10,184 | 0.0% |
+| **3** | 0.0847 | 1.061 | 1.15 | 446,488 | 91.5% |
+| **3** | 0.8420 | 1.793 | 3.87 | 60,186 | 15.8% |
+| **4** | 0.0411 🏆 | 1.029 | 1.07 | 514,008 | 95.9% |
+| **4** | 0.6005 🏆 | 1.516 | 2.44 | 232,636 | 39.9% |
+
+### Generated Text Samples
+
+Below are text samples generated from each Markov chain model:
+
+**Context Size 1:**
+
+1. `. konwencionelna mutageneza pśi albańskej a litawskeju ( - 1962 : hugo gunckel lüer – nimski`
+2. `, 1 , leipzig palmenhaus auf der räuber hotzenplotz " kaž “ a twórje kupy ,`
+3. `: jazor 58 . aitingk , ale teke literarne myto ćišinskego kategorija : prizaŕske bórkowy amt`
+
+**Context Size 2:**
+
+1. `kategorija : sedlišćo w dolnej łužycy . wótkaze lisćina galiskich słowow pśi wordgumbo nastawk pśi i...`
+2. `) jo družyna droznow . samica jo brunocarna , mjaztym až se w šyrokem źělu pódpołnocneje afriki`
+3. `) . města nejwětše města su : santiago de cuba ) jo historiska slězyna japańskeje tragedije .`
+
+**Context Size 3:**
+
+1. `kategorija : sedlišćo w pólskej kategorija : pomorske wójwodstwo kategorija : sedlišćo w českej kate...`
+2. `: sedlišćo w baden - württembergskej kategorija : rěka w českej , ako na pśikład za staty ,`
+3. `. kategorija : sad kategorija : bomy kategorija : hybridy`
+
+**Context Size 4:**
+
+1. `kategorija : sedlišćo w argentinskej kategorija : stolica w europje kategorija : rěka , alfabetiski ...`
+2. `kategorija : roź . 1757 kategorija : wum . 1913`
+3. `kategorija : wum . 1985 kategorija : słowakski spiwaŕ kategorija : muž`
+
+
+### Key Findings
+
+- **Best Predictability:** Context-4 with 95.9% predictability
+- **Branching Factor:** Decreases with context size (more deterministic)
+- **Memory Trade-off:** Larger contexts require more storage (232,636 contexts)
+- **Recommendation:** Context-3 or Context-4 for text generation
+
+---
+## 4. Vocabulary Analysis
+
+![Zipf's Law](visualizations/zipf_law.png)
+
+![Top Words](visualizations/top20_words.png)
+
+![Coverage Curve](visualizations/vocab_coverage.png)
+
+### Statistics
+
+| Metric | Value |
+|--------|-------|
+| Vocabulary Size | 32,309 |
+| Total Tokens | 432,054 |
+| Mean Frequency | 13.37 |
+| Median Frequency | 3 |
+| Frequency Std Dev | 142.47 |
+
+### Most Common Words
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | a | 12,388 |
+| 2 | w | 12,290 |
+| 3 | jo | 11,552 |
+| 4 | kategorija | 7,674 |
+| 5 | na | 4,664 |
+| 6 | z | 4,262 |
+| 7 | se | 3,646 |
+| 8 | wót | 3,524 |
+| 9 | su | 2,927 |
+| 10 | do | 2,441 |
+
+### Least Common Words (from vocabulary)
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | 1474wjerchojstwo | 2 |
+| 2 | wolgast5 | 2 |
+| 3 | 1478wjerchojstwo | 2 |
+| 4 | 1592 | 2 |
+| 5 | 1625wjerchojstwo | 2 |
+| 6 | zachdniego | 2 |
+| 7 | gdanskiego | 2 |
+| 8 | podzially | 2 |
+| 9 | ujazd | 2 |
+| 10 | mojš | 2 |
+
+### Zipf's Law Analysis
+
+| Metric | Value |
+|--------|-------|
+| Zipf Coefficient | 0.9658 |
+| R² (Goodness of Fit) | 0.995856 |
+| Adherence Quality | **excellent** |
+
+### Coverage Analysis
+
+| Top N Words | Coverage |
+|-------------|----------|
+| Top 100 | 30.4% |
+| Top 1,000 | 57.0% |
+| Top 5,000 | 77.2% |
+| Top 10,000 | 85.9% |
+
+### Key Findings
+
+- **Zipf Compliance:** R²=0.9959 indicates excellent adherence to Zipf's law
+- **High Frequency Dominance:** Top 100 words cover 30.4% of corpus
+- **Long Tail:** 22,309 words needed for remaining 14.1% coverage
+
+---
+## 5. Word Embeddings Evaluation
+
+![Embedding Isotropy](visualizations/embedding_isotropy.png)
+
+![Similarity Matrix](visualizations/embedding_similarity.png)
+
+![t-SNE Words](visualizations/tsne_words.png)
+
+![t-SNE Sentences](visualizations/tsne_sentences.png)
+
+### Model Comparison
+
+| Model | Vocab Size | Dimension | Avg Norm | Std Norm | Isotropy |
+|-------|------------|-----------|----------|----------|----------|
+| **mono_32d** | 11,406 | 32 | 4.255 | 0.918 | 0.8252 🏆 |
+| **mono_64d** | 11,406 | 64 | 4.511 | 0.859 | 0.5783 |
+| **mono_128d** | 11,406 | 128 | 4.576 | 0.852 | 0.1767 |
+| **embeddings_enhanced** | 0 | 0 | 0.000 | 0.000 | 0.0000 |
+
+### Key Findings
+
+- **Best Isotropy:** mono_32d with 0.8252 (more uniform distribution)
+- **Dimension Trade-off:** Higher dimensions capture more semantics but reduce isotropy
+- **Vocabulary Coverage:** All models cover 11,406 words
+- **Recommendation:** 100d for balanced semantic capture and efficiency
+
+---
+## 6. Summary & Recommendations
+
+![Performance Dashboard](visualizations/performance_dashboard.png)
+
+### Production Recommendations
+
+| Component | Recommended | Rationale |
+|-----------|-------------|-----------|
+| Tokenizer | **32k BPE** | Best compression (4.29x) with low UNK rate |
+| N-gram | **5-gram** | Lowest perplexity (508) |
+| Markov | **Context-4** | Highest predictability (95.9%) |
+| Embeddings | **100d** | Balanced semantic capture and isotropy |
+
+---
+## Appendix: Metrics Glossary & Interpretation Guide
+
+This section provides definitions, intuitions, and guidance for interpreting the metrics used throughout this report.
+
+### Tokenizer Metrics
+
+**Compression Ratio**
+> *Definition:* The ratio of characters to tokens (chars/token). Measures how efficiently the tokenizer represents text.
+>
+> *Intuition:* Higher compression means fewer tokens needed to represent the same text, reducing sequence lengths for downstream models. A 3x compression means ~3 characters per token on average.
+>
+> *What to seek:* Higher is generally better for efficiency, but extremely high compression may indicate overly aggressive merging that loses morphological information.
+
+**Average Token Length (Fertility)**
+> *Definition:* Mean number of characters per token produced by the tokenizer.
+>
+> *Intuition:* Reflects the granularity of tokenization. Longer tokens capture more context but may struggle with rare words; shorter tokens are more flexible but increase sequence length.
+>
+> *What to seek:* Balance between 2-5 characters for most languages. Arabic/morphologically-rich languages may benefit from slightly longer tokens.
+
+**Unknown Token Rate (OOV Rate)**
+> *Definition:* Percentage of tokens that map to the unknown/UNK token, indicating words the tokenizer cannot represent.
+>
+> *Intuition:* Lower OOV means better vocabulary coverage. High OOV indicates the tokenizer encounters many unseen character sequences.
+>
+> *What to seek:* Below 1% is excellent; below 5% is acceptable. BPE tokenizers typically achieve very low OOV due to subword fallback.
+
+### N-gram Model Metrics
+
+**Perplexity**
+> *Definition:* Measures how "surprised" the model is by test data. Mathematically: 2^(cross-entropy). Lower values indicate better prediction.
+>
+> *Intuition:* If perplexity is 100, the model is as uncertain as if choosing uniformly among 100 options at each step. A perplexity of 10 means effectively choosing among 10 equally likely options.
+>
+> *What to seek:* Lower is better. Perplexity decreases with larger n-grams (more context). Values vary widely by language and corpus size.
+
+**Entropy**
+> *Definition:* Average information content (in bits) needed to encode the next token given the context. Related to perplexity: perplexity = 2^entropy.
+>
+> *Intuition:* High entropy means high uncertainty/randomness; low entropy means predictable patterns. Natural language typically has entropy between 1-4 bits per character.
+>
+> *What to seek:* Lower entropy indicates more predictable text patterns. Entropy should decrease as n-gram size increases.
+
+**Coverage (Top-K)**
+> *Definition:* Percentage of corpus occurrences explained by the top K most frequent n-grams.
+>
+> *Intuition:* High coverage with few patterns indicates repetitive/formulaic text; low coverage suggests diverse vocabulary usage.
+>
+> *What to seek:* Depends on use case. For language modeling, moderate coverage (40-60% with top-1000) is typical for natural text.
+
+### Markov Chain Metrics
+
+**Average Entropy**
+> *Definition:* Mean entropy across all contexts, measuring average uncertainty in next-word prediction.
+>
+> *Intuition:* Lower entropy means the model is more confident about what comes next. Context-1 has high entropy (many possible next words); Context-4 has low entropy (few likely continuations).
+>
+> *What to seek:* Decreasing entropy with larger context sizes. Very low entropy (<0.1) indicates highly deterministic transitions.
+
+**Branching Factor**
+> *Definition:* Average number of unique next tokens observed for each context.
+>
+> *Intuition:* High branching = many possible continuations (flexible but uncertain); low branching = few options (predictable but potentially repetitive).
+>
+> *What to seek:* Branching factor should decrease with context size. Values near 1.0 indicate nearly deterministic chains.
+
+**Predictability**
+> *Definition:* Derived metric: (1 - normalized_entropy) × 100%. Indicates how deterministic the model's predictions are.
+>
+> *Intuition:* 100% predictability means the next word is always certain; 0% means completely random. Real text falls between these extremes.
+>
+> *What to seek:* Higher predictability for text generation quality, but too high (>98%) may produce repetitive output.
+
+### Vocabulary & Zipf's Law Metrics
+
+**Zipf's Coefficient**
+> *Definition:* The slope of the log-log plot of word frequency vs. rank. Zipf's law predicts this should be approximately -1.
+>
+> *Intuition:* A coefficient near -1 indicates the corpus follows natural language patterns where a few words are very common and most words are rare.
+>
+> *What to seek:* Values between -0.8 and -1.2 indicate healthy natural language distribution. Deviations may suggest domain-specific or artificial text.
+
+**R² (Coefficient of Determination)**
+> *Definition:* Measures how well the linear fit explains the frequency-rank relationship. Ranges from 0 to 1.
+>
+> *Intuition:* R² near 1.0 means the data closely follows Zipf's law; lower values indicate deviation from expected word frequency patterns.
+>
+> *What to seek:* R² > 0.95 is excellent; > 0.99 indicates near-perfect Zipf adherence typical of large natural corpora.
+
+**Vocabulary Coverage**
+> *Definition:* Cumulative percentage of corpus tokens accounted for by the top N words.
+>
+> *Intuition:* Shows how concentrated word usage is. If top-100 words cover 50% of text, the corpus relies heavily on common words.
+>
+> *What to seek:* Top-100 covering 30-50% is typical. Higher coverage indicates more repetitive text; lower suggests richer vocabulary.
+
+### Word Embedding Metrics
+
+**Isotropy**
+> *Definition:* Measures how uniformly distributed vectors are in the embedding space. Computed as the ratio of minimum to maximum singular values.
+>
+> *Intuition:* High isotropy (near 1.0) means vectors spread evenly in all directions; low isotropy means vectors cluster in certain directions, reducing expressiveness.
+>
+> *What to seek:* Higher isotropy generally indicates better-quality embeddings. Values > 0.1 are reasonable; > 0.3 is good. Lower-dimensional embeddings tend to have higher isotropy.
+
+**Average Norm**
+> *Definition:* Mean magnitude (L2 norm) of word vectors in the embedding space.
+>
+> *Intuition:* Indicates the typical "length" of vectors. Consistent norms suggest stable training; high variance may indicate some words are undertrained.
+>
+> *What to seek:* Relatively consistent norms across models. The absolute value matters less than consistency (low std deviation).
+
+**Cosine Similarity**
+> *Definition:* Measures angular similarity between vectors, ranging from -1 (opposite) to 1 (identical direction).
+>
+> *Intuition:* Words with similar meanings should have high cosine similarity. This is the standard metric for semantic relatedness in embeddings.
+>
+> *What to seek:* Semantically related words should score > 0.5; unrelated words should be near 0. Synonyms often score > 0.7.
+
+**t-SNE Visualization**
+> *Definition:* t-Distributed Stochastic Neighbor Embedding - a dimensionality reduction technique that preserves local structure for visualization.
+>
+> *Intuition:* Clusters in t-SNE plots indicate groups of semantically related words. Spread indicates vocabulary diversity; tight clusters suggest semantic coherence.
+>
+> *What to seek:* Meaningful clusters (e.g., numbers together, verbs together). Avoid over-interpreting distances - t-SNE preserves local, not global, structure.
+
+### General Interpretation Guidelines
+
+1. **Compare within model families:** Metrics are most meaningful when comparing models of the same type (e.g., 8k vs 64k tokenizer).
+2. **Consider trade-offs:** Better performance on one metric often comes at the cost of another (e.g., compression vs. OOV rate).
+3. **Context matters:** Optimal values depend on downstream tasks. Text generation may prioritize different metrics than classification.
+4. **Corpus influence:** All metrics are influenced by corpus characteristics. Wikipedia text differs from social media or literature.
+5. **Language-specific patterns:** Morphologically rich languages (like Arabic) may show different optimal ranges than analytic languages.
+
+
+### Visualizations Index
+
+| Visualization | Description |
+|---------------|-------------|
+| Tokenizer Compression | Compression ratios by vocabulary size |
+| Tokenizer Fertility | Average token length by vocabulary |
+| Tokenizer OOV | Unknown token rates |
+| Tokenizer Total Tokens | Total tokens by vocabulary |
+| N-gram Perplexity | Perplexity by n-gram size |
+| N-gram Entropy | Entropy by n-gram size |
+| N-gram Coverage | Top pattern coverage |
+| N-gram Unique | Unique n-gram counts |
+| Markov Entropy | Entropy by context size |
+| Markov Branching | Branching factor by context |
+| Markov Contexts | Unique context counts |
+| Zipf's Law | Frequency-rank distribution with fit |
+| Vocab Frequency | Word frequency distribution |
+| Top 20 Words | Most frequent words |
+| Vocab Coverage | Cumulative coverage curve |
+| Embedding Isotropy | Vector space uniformity |
+| Embedding Norms | Vector magnitude distribution |
+| Embedding Similarity | Word similarity heatmap |
+| Nearest Neighbors | Similar words for key terms |
+| t-SNE Words | 2D word embedding visualization |
+| t-SNE Sentences | 2D sentence embedding visualization |
+| Position Encoding | Encoding method comparison |
+| Model Sizes | Storage requirements |
+| Performance Dashboard | Comprehensive performance overview |
+
+---
+## About This Project
+
+### Data Source
+
+Models trained on [wikipedia-monthly](https://huggingface.co/datasets/omarkamali/wikipedia-monthly) - a monthly snapshot of Wikipedia articles across 300+ languages.
+
+### Project
+
+A project by **[Wikilangs](https://wikilangs.org)** - Open-source NLP models for every Wikipedia language.
+
+### Maintainer
+
+[Omar Kamali](https://omarkamali.com) - [Omneity Labs](https://omneitylabs.com)
+
+### Citation
+
+If you use these models in your research, please cite:
+
+```bibtex
+@misc{wikilangs2025,
+  author = {Kamali, Omar},
+  title = {Wikilangs: Open NLP Models for Wikipedia Languages},
+  year = {2025},
+  publisher = {HuggingFace},
+  url = {https://huggingface.co/wikilangs}
+  institution = {Omneity Labs}
+}
+```
+
+### License
+
+MIT License - Free for academic and commercial use.
+
+### Links
+
+- 🌐 Website: [wikilangs.org](https://wikilangs.org)
+- 🤗 Models: [huggingface.co/wikilangs](https://huggingface.co/wikilangs)
+- 📊 Data: [wikipedia-monthly](https://huggingface.co/datasets/omarkamali/wikipedia-monthly)
+- 👤 Author: [Omar Kamali](https://huggingface.co/omarkamali)
+---
+*Generated by Wikilangs Models Pipeline*
+
+*Report Date: 2025-12-30 08:33:54*

models/embeddings/monolingual/dsb_128d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69c984e87d1636c608606c4c5465d8002825ca2f54d231f3d0a0277f593edecb
+size 1035879476

models/embeddings/monolingual/dsb_128d.meta.json

@@ -0,0 +1 @@
+{"lang": "dsb", "dim": 128, "max_seq_len": 512, "is_aligned": false}

models/embeddings/monolingual/dsb_128d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "dsb",
+  "dimension": 128,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 128,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 11406
+}

models/embeddings/monolingual/dsb_32d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:081fb3a085c2c647fb4256975b1946eef7c9db9c3214387b4f47c18933cc0175
+size 259119668

models/embeddings/monolingual/dsb_32d.meta.json

@@ -0,0 +1 @@
+{"lang": "dsb", "dim": 32, "max_seq_len": 512, "is_aligned": false}

models/embeddings/monolingual/dsb_32d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "dsb",
+  "dimension": 32,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 32,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 11406
+}

models/embeddings/monolingual/dsb_64d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:811da793a1d91d94aecf05624b1782db2a0780ebddab0716c5a7449a1314bee9
+size 518039604

models/embeddings/monolingual/dsb_64d.meta.json

@@ -0,0 +1 @@
+{"lang": "dsb", "dim": 64, "max_seq_len": 512, "is_aligned": false}

models/embeddings/monolingual/dsb_64d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "dsb",
+  "dimension": 64,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 64,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 11406
+}

models/subword_markov/dsb_markov_ctx1_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e75991d06d216913d5349e487fb7f5dda5adf4a84e1112c7d75804a41f526651
+size 78473

models/subword_markov/dsb_markov_ctx1_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "subword",
+  "language": "dsb",
+  "unique_contexts": 1062,
+  "total_transitions": 3435229
+}

models/subword_markov/dsb_markov_ctx2_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2d04158c387dad22bea4d026f6486e27063cbccf1b5b267c23067200f38df80
+size 474937

models/subword_markov/dsb_markov_ctx2_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "subword",
+  "language": "dsb",
+  "unique_contexts": 10184,
+  "total_transitions": 3431780
+}

models/subword_markov/dsb_markov_ctx3_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:65e073c2672b04e0cc6a210d14c210bd5f5b2d610e9df300eec4e645bb58b544
+size 1661127

models/subword_markov/dsb_markov_ctx3_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "subword",
+  "language": "dsb",
+  "unique_contexts": 60186,
+  "total_transitions": 3428331
+}

models/subword_markov/dsb_markov_ctx4_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e1e13ba8381c4b0aeedbfd268e3df569421c02c0509adce385934b61c9d4134
+size 4548151

models/subword_markov/dsb_markov_ctx4_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "subword",
+  "language": "dsb",
+  "unique_contexts": 232636,
+  "total_transitions": 3424882
+}

models/subword_ngram/dsb_2gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54ab8a0b169ac29131c7497aab72ed0c6262588ee25933132b96a9ea893c6cab
+size 52880

models/subword_ngram/dsb_2gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "subword",
+  "language": "dsb",
+  "unique_ngrams": 4003,
+  "total_ngrams": 3435229
+}

models/subword_ngram/dsb_3gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:491b077d6277a92677cbfb36b213f6c9845b6625d639c685caaf3c81b098ae71
+size 355310

models/subword_ngram/dsb_3gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "subword",
+  "language": "dsb",
+  "unique_ngrams": 29259,
+  "total_ngrams": 3431780
+}

models/subword_ngram/dsb_4gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0210a38e1f600c563d26ca91f697d6f95cf6c35633b5d062baf24f897ca3ea6
+size 1444626

models/subword_ngram/dsb_4gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "subword",
+  "language": "dsb",
+  "unique_ngrams": 127625,
+  "total_ngrams": 3428331
+}

models/tokenizer/dsb_tokenizer_16k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83223ed94450b083f445917ad19d45108f784776f35590e24106f870e89fc048
+size 509210

models/tokenizer/dsb_tokenizer_32k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:385d9c2bd426307bc0210de4e33f8ef09b6d82186b16307f29650deb8b0128a5
+size 785019

models/tokenizer/dsb_tokenizer_64k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be1039984143149d1392ae5a765816e6ff2af6dad234612968afce460c6ae6d9
+size 1380290

models/tokenizer/dsb_tokenizer_8k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:230d8ebcf4cbf468966a99ee03e43588c6326a65ab777732705c0175ce721003
+size 372857

models/vocabulary/dsb_vocabulary.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87ff7058d1987d1d5ac416a3afd9abb9662e91efa908990e180308c27a05e007
+size 548787

models/vocabulary/dsb_vocabulary_metadata.json

@@ -0,0 +1,16 @@
+{
+  "language": "dsb",
+  "vocabulary_size": 32309,
+  "statistics": {
+    "type_token_ratio": 0.17291342948830685,
+    "coverage": {
+      "top_100": 0.2721360928748627,
+      "top_1000": 0.5094544574264923,
+      "top_5000": 0.6904536774001329,
+      "top_10000": 0.7675997326806216
+    },
+    "hapax_count": 51263,
+    "hapax_ratio": 0.6133992246206863,
+    "total_documents": 3449
+  }
+}

models/word_markov/dsb_markov_ctx1_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:578726324c9438985c16ae29b0fba1e4d2c95ea67256780dc13e4597c96325ed
+size 2811261

models/word_markov/dsb_markov_ctx1_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "word",
+  "language": "dsb",
+  "unique_contexts": 83705,
+  "total_transitions": 642857
+}

models/word_markov/dsb_markov_ctx2_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d179629dada8b928138f75fd99ce884b747702d0d96c228aee1ff738dbbb6de
+size 5734522

models/word_markov/dsb_markov_ctx2_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "word",
+  "language": "dsb",
+  "unique_contexts": 299024,
+  "total_transitions": 639409
+}

models/word_markov/dsb_markov_ctx3_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9f62a7fdf01044badf4a01418e4bd019a03061f0e862761500f4315f28205b80
+size 7943109

models/word_markov/dsb_markov_ctx3_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "word",
+  "language": "dsb",
+  "unique_contexts": 446488,
+  "total_transitions": 635961
+}

models/word_markov/dsb_markov_ctx4_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f91b66af45a3cffb2b5fee99ae80be4f5cdea8062772f9a81c0bae64d8c37eaf
+size 9386531

models/word_markov/dsb_markov_ctx4_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "word",
+  "language": "dsb",
+  "unique_contexts": 514008,
+  "total_transitions": 632513
+}

models/word_ngram/dsb_2gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34b924e465fb19345a7d89593c5d07248aca3fb566bc30dc4d52e8da8c70be6a
+size 212696

models/word_ngram/dsb_2gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "word",
+  "language": "dsb",
+  "unique_ngrams": 14685,
+  "total_ngrams": 642857
+}

models/word_ngram/dsb_3gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:766a87049e294fe083de9b40a25096ea9a64d30e961e2238372257f093486632
+size 347612

models/word_ngram/dsb_3gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "word",
+  "language": "dsb",
+  "unique_ngrams": 21515,
+  "total_ngrams": 639409
+}

models/word_ngram/dsb_4gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe841e8c1854ef6fc5cca0bac3bb5b6da357d93862d521abce14ee3b3cc1d28f
+size 637664

models/word_ngram/dsb_4gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "word",
+  "language": "dsb",
+  "unique_ngrams": 37295,
+  "total_ngrams": 635961
+}