Upload all models and assets for az (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 +1,567 @@
+---
+language: az
+language_name: AZ
+language_family: turkic_oghuz
+tags:
+  - wikilangs
+  - nlp
+  - tokenizer
+  - embeddings
+  - n-gram
+  - markov
+  - wikipedia
+  - monolingual
+  - family-turkic_oghuz
+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: 4.560
+  - name: best_isotropy
+    type: isotropy
+    value: 0.8153
+  - name: vocabulary_size
+    type: vocab
+    value: 807823
+generated: 2025-12-27
+---
+
+# AZ - Wikilangs Models
+## Comprehensive Research Report & Full Ablation Study
+
+This repository contains NLP models trained and evaluated by Wikilangs, specifically on **AZ** 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** | 3.637x | 3.59 | 0.0980% | 1,442,215 |
+| **16k** | 4.016x | 3.97 | 0.1082% | 1,305,814 |
+| **32k** | 4.326x | 4.27 | 0.1165% | 1,212,431 |
+| **64k** | 4.560x 🏆 | 4.50 | 0.1229% | 1,150,092 |
+
+### Tokenization Examples
+
+Below are sample sentences tokenized with each vocabulary size:
+
+**Sample 1:** `Hadisələr 
+
+ Doğumlar 
+
+ Vəfatlar 
+ Soqdian — e.ə. 424–423-cü illərdə hakimiyyət...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁hadisələr ▁doğumlar ▁vəfatlar ▁s oq di an ▁— ▁e . ... (+21 more)` | 31 |
+| 16k | `▁hadisələr ▁doğumlar ▁vəfatlar ▁s oq dian ▁— ▁e . ə ... (+19 more)` | 29 |
+| 32k | `▁hadisələr ▁doğumlar ▁vəfatlar ▁s oq dian ▁— ▁e . ə ... (+18 more)` | 28 |
+| 64k | `▁hadisələr ▁doğumlar ▁vəfatlar ▁soq dian ▁— ▁e . ə . ... (+17 more)` | 27 |
+
+**Sample 2:** `() —  aləminin  dəstəsinin  fəsiləsinin  cinsinə aid bitki növü.
+
+ İstinadlar 
+
+...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁() ▁— ▁aləminin ▁dəstəsinin ▁fəsiləsinin ▁cinsinə ▁aid ▁bitki ▁növü . ... (+3 more)` | 13 |
+| 16k | `▁() ▁— ▁aləminin ▁dəstəsinin ▁fəsiləsinin ▁cinsinə ▁aid ▁bitki ▁növü . ... (+3 more)` | 13 |
+| 32k | `▁() ▁— ▁aləminin ▁dəstəsinin ▁fəsiləsinin ▁cinsinə ▁aid ▁bitki ▁növü . ... (+3 more)` | 13 |
+| 64k | `▁() ▁— ▁aləminin ▁dəstəsinin ▁fəsiləsinin ▁cinsinə ▁aid ▁bitki ▁növü . ... (+3 more)` | 13 |
+
+**Sample 3:** `() —  aləminin  dəstəsinin  fəsiləsinin  cinsinə aid bitki növü.
+
+ İstinadlar 
+
+...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁() ▁— ▁aləminin ▁dəstəsinin ▁fəsiləsinin ▁cinsinə ▁aid ▁bitki ▁növü . ... (+3 more)` | 13 |
+| 16k | `▁() ▁— ▁aləminin ▁dəstəsinin ▁fəsiləsinin ▁cinsinə ▁aid ▁bitki ▁növü . ... (+3 more)` | 13 |
+| 32k | `▁() ▁— ▁aləminin ▁dəstəsinin ▁fəsiləsinin ▁cinsinə ▁aid ▁bitki ▁növü . ... (+3 more)` | 13 |
+| 64k | `▁() ▁— ▁aləminin ▁dəstəsinin ▁fəsiləsinin ▁cinsinə ▁aid ▁bitki ▁növü . ... (+3 more)` | 13 |
+
+
+### Key Findings
+
+- **Best Compression:** 64k achieves 4.560x compression
+- **Lowest UNK Rate:** 8k with 0.0980% 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** | 116,331 🏆 | 16.83 | 1,418,191 | 12.2% | 25.0% |
+| **2-gram** | 463 🏆 | 8.85 | 22,197 | 55.1% | 96.6% |
+| **3-gram** | 409,514 | 18.64 | 2,779,660 | 7.3% | 16.3% |
+| **3-gram** | 4,354 | 12.09 | 215,243 | 19.6% | 58.8% |
+| **4-gram** | 1,285,123 | 20.29 | 5,336,948 | 4.2% | 9.8% |
+| **4-gram** | 24,655 | 14.59 | 1,308,669 | 10.0% | 31.8% |
+
+### Top 5 N-grams by Size
+
+**2-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `i ̇` | 824,078 |
+| 2 | `- ci` | 452,530 |
+| 3 | `kateqoriya :` | 365,784 |
+| 4 | `. i` | 236,809 |
+| 5 | `ci ildə` | 221,298 |
+
+**3-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `. i ̇` | 227,697 |
+| 2 | `- ci ildə` | 220,310 |
+| 3 | `i ̇ stinadlar` | 171,011 |
+| 4 | `- cü ildə` | 76,499 |
+| 5 | `( ) —` | 70,885 |
+
+**4-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `. i ̇ stinadlar` | 103,078 |
+| 2 | `i ̇ stinadlar kateqoriya` | 65,368 |
+| 3 | `̇ stinadlar kateqoriya :` | 65,368 |
+| 4 | `i ̇ stinadlar xarici` | 45,918 |
+| 5 | `̇ stinadlar xarici keçidlər` | 45,490 |
+
+
+### Key Findings
+
+- **Best Perplexity:** 2-gram with 463
+- **Entropy Trend:** Decreases with larger n-grams (more predictable)
+- **Coverage:** Top-1000 patterns cover ~32% 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** | 0.7520 | 1.684 | 8.95 | 1,994,233 | 24.8% |
+| **1** | 1.3997 | 2.638 | 10.24 | 7,113 | 0.0% |
+| **2** | 0.3489 | 1.274 | 2.24 | 17,848,485 | 65.1% |
+| **2** | 0.8515 | 1.804 | 6.20 | 72,786 | 14.8% |
+| **3** | 0.1419 | 1.103 | 1.34 | 39,971,548 | 85.8% |
+| **3** | 0.8951 | 1.860 | 5.05 | 451,475 | 10.5% |
+| **4** | 0.0656 🏆 | 1.046 | 1.14 | 53,703,949 | 93.4% |
+| **4** | 0.7324 🏆 | 1.661 | 3.53 | 2,281,437 | 26.8% |
+
+### Generated Text Samples
+
+Below are text samples generated from each Markov chain model:
+
+**Context Size 1:**
+
+1. `. şerlər əsasən observantlar monastırlarının bəzilərinin əvvəllər per teodor hertslin başlatdığı işğ...`
+2. `, buenos - dən az saylı xəstəxanada həkim , 316 kvadrat metrdən çox öz qoşunu təbrizi`
+3. `- ci ildə yarananlar kateqoriya : 46 - ci ildə baş verən əsas götürərək . orta`
+
+**Context Size 2:**
+
+1. `i ̇ stinadlar xarici keçidlər бухарский трактат о каллиграфах и художниках трактат о каллиграфах и х...`
+2. `- ci ildə i ̇ raqa , hindistana və hətta oğlundan dörd il sonra " psm3 "`
+3. `kateqoriya : şuşanın görməli yerləri kateqoriya : sionistlər kateqoriya : germi şəhristanının kəndlə...`
+
+**Context Size 3:**
+
+1. `. i ̇ ki cilddə . i cild . bakı : nafta - press , 2013 ) (`
+2. `- ci ildə yarananlar kateqoriya : 8 iyunda yarananlar kateqoriya : universitas 21 kateqoriya : azərb...`
+3. `i ̇ stinadlar xarici keçidlər həmçinin bax kateqoriya : yaponiya hüquqşünasları kateqoriya : azərbay...`
+
+**Context Size 4:**
+
+1. `. i ̇ stinadlar mənbə " treska " kateqoriya : avropa dağ sistemləri kateqoriya : gürcüstan relyefi k...`
+2. `̇ stinadlar kateqoriya : traktorçular kateqoriya : azərbaycan pambıqçıları kateqoriya : azərbaycan s...`
+3. `i ̇ stinadlar kateqoriya : xorvatiyanın olimpiya həndbolçuları kateqoriya : 2016 yay olimpiya oyunla...`
+
+
+### Key Findings
+
+- **Best Predictability:** Context-4 with 93.4% predictability
+- **Branching Factor:** Decreases with context size (more deterministic)
+- **Memory Trade-off:** Larger contexts require more storage (2,281,437 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 | 807,823 |
+| Total Tokens | 58,755,251 |
+| Mean Frequency | 72.73 |
+| Median Frequency | 4 |
+| Frequency Std Dev | 2550.61 |
+
+### Most Common Words
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | və | 1,490,176 |
+| 2 | i | 892,240 |
+| 3 | ci | 455,280 |
+| 4 | ildə | 414,485 |
+| 5 | ilə | 413,230 |
+| 6 | kateqoriya | 366,496 |
+| 7 | bir | 366,287 |
+| 8 | bu | 362,130 |
+| 9 | azərbaycan | 248,838 |
+| 10 | də | 234,167 |
+
+### Least Common Words (from vocabulary)
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | cërrik | 2 |
+| 2 | liamın | 2 |
+| 3 | liamla | 2 |
+| 4 | backstab | 2 |
+| 5 | antonioi | 2 |
+| 6 | nipissinq | 2 |
+| 7 | votivkirche | 2 |
+| 8 | pirtle | 2 |
+| 9 | takaxasinin | 2 |
+| 10 | caporael | 2 |
+
+### Zipf's Law Analysis
+
+| Metric | Value |
+|--------|-------|
+| Zipf Coefficient | 0.9771 |
+| R² (Goodness of Fit) | 0.992093 |
+| Adherence Quality | **excellent** |
+
+### Coverage Analysis
+
+| Top N Words | Coverage |
+|-------------|----------|
+| Top 100 | 22.3% |
+| Top 1,000 | 46.6% |
+| Top 5,000 | 66.6% |
+| Top 10,000 | 74.5% |
+
+### Key Findings
+
+- **Zipf Compliance:** R²=0.9921 indicates excellent adherence to Zipf's law
+- **High Frequency Dominance:** Top 100 words cover 22.3% of corpus
+- **Long Tail:** 797,823 words needed for remaining 25.5% 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** | 509,900 | 32 | 3.229 | 0.928 | 0.8153 🏆 |
+| **mono_64d** | 509,900 | 64 | 3.675 | 0.940 | 0.8024 |
+| **mono_128d** | 509,900 | 128 | 4.156 | 0.943 | 0.7626 |
+| **embeddings_enhanced** | 0 | 0 | 0.000 | 0.000 | 0.0000 |
+
+### Key Findings
+
+- **Best Isotropy:** mono_32d with 0.8153 (more uniform distribution)
+- **Dimension Trade-off:** Higher dimensions capture more semantics but reduce isotropy
+- **Vocabulary Coverage:** All models cover 509,900 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 (4.56x) with low UNK rate |
+| N-gram | **5-gram** | Lowest perplexity (463) |
+| Markov | **Context-4** | Highest predictability (93.4%) |
+| 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: 2025-12-27 22:29:22*

models/embeddings/monolingual/az_128d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef77feb09cfca11c3c70d43e5ebfc128aa2ce50ce323c7a349cd6272dceeecaf
+size 1556313250

models/embeddings/monolingual/az_128d.meta.json

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

models/embeddings/monolingual/az_128d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "az",
+  "dimension": 128,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 128,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 509900
+}

models/embeddings/monolingual/az_32d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:213be535e52a36209d1823ea600aa5f918189933d961ffc129d4abdf1241505c
+size 396710050

models/embeddings/monolingual/az_32d.meta.json

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

models/embeddings/monolingual/az_32d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "az",
+  "dimension": 32,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 32,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 509900
+}

models/embeddings/monolingual/az_64d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84f2fe715f5b8f553b9c0597a167bd6132f61d2a82f93df713c1f6ccc727c629
+size 783244450

models/embeddings/monolingual/az_64d.meta.json

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

models/embeddings/monolingual/az_64d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "az",
+  "dimension": 64,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 64,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 509900
+}

models/subword_markov/az_markov_ctx1_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e52b2d1b7aa86c543cfb6c57d379300d0aa39afe36befd7696d0f381c5c26465
+size 483989

models/subword_markov/az_markov_ctx1_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "subword",
+  "language": "az",
+  "unique_contexts": 7113,
+  "total_transitions": 445372741
+}

models/subword_markov/az_markov_ctx2_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1acb2d2cb49f03823d7a5f04faf687f085160a4c1dc2adab7a4e13964b03be2
+size 3427707

models/subword_markov/az_markov_ctx2_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "subword",
+  "language": "az",
+  "unique_contexts": 72786,
+  "total_transitions": 445163302
+}

models/subword_markov/az_markov_ctx3_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3871d32804a2219f5f3396b0b91728dd6160992e59d42fe8896c4bcb94382621
+size 17111060

models/subword_markov/az_markov_ctx3_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "subword",
+  "language": "az",
+  "unique_contexts": 451475,
+  "total_transitions": 444953863
+}

models/subword_markov/az_markov_ctx4_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:992bd3717b1c0b08cdf1f88aca515bb1af0f25b7e826c455a45e0c59fba62789
+size 63283624

models/subword_markov/az_markov_ctx4_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "subword",
+  "language": "az",
+  "unique_contexts": 2281437,
+  "total_transitions": 444744424
+}

models/subword_ngram/az_2gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6660aa5cd67b23837187e8e6e9dfb8811773fbc67c340cc925dd96fa477df8e
+size 303706

models/subword_ngram/az_2gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "subword",
+  "language": "az",
+  "unique_ngrams": 22197,
+  "total_ngrams": 445372741
+}

models/subword_ngram/az_3gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa8ff9a642d58e018c6450c1335c0a292325f197796ebacda28d7c1c7a3777b7
+size 2629264

models/subword_ngram/az_3gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "subword",
+  "language": "az",
+  "unique_ngrams": 215243,
+  "total_ngrams": 445163302
+}

models/subword_ngram/az_4gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7bc1dc82aeb72c52118bdcc1218942799ccbb6d2026c2502ea95e09500818f3f
+size 15841110

models/subword_ngram/az_4gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "subword",
+  "language": "az",
+  "unique_ngrams": 1308669,
+  "total_ngrams": 444953863
+}

models/tokenizer/az_tokenizer_16k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ccd48dce261b522aa16ede8b95d9ad20d051d2c94ba30c4ab9ed7bef4352fcb5
+size 524499

models/tokenizer/az_tokenizer_32k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bdf25b32c3b72ec7ec3b8aceeb4d191e79c362a8d341638d4fcdb76e2d80c91d
+size 828562

models/tokenizer/az_tokenizer_64k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3494c2e094066efac7039285393d96c5566a53494a40c805dba02bf3d6872305
+size 1452009

models/tokenizer/az_tokenizer_8k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d7a854bf952eaf3d642c73822c12056ae4fe37d97611c7982a3a2b8fbd00b18
+size 378944

models/vocabulary/az_vocabulary.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ed780a3b2e5a6030d0e1fd52789c427bb6d0e19d7919077548a05c0987228e7
+size 12576112

models/vocabulary/az_vocabulary_metadata.json

@@ -0,0 +1,16 @@
+{
+  "language": "az",
+  "vocabulary_size": 807823,
+  "statistics": {
+    "type_token_ratio": 0.033264968758808605,
+    "coverage": {
+      "top_100": 0.2190541471720636,
+      "top_1000": 0.45649574344105814,
+      "top_5000": 0.652480233353335,
+      "top_10000": 0.7304854329670376
+    },
+    "hapax_count": 1186125,
+    "hapax_ratio": 0.5948625540886723,
+    "total_documents": 209439
+  }
+}

models/word_markov/az_markov_ctx1_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78f23984cd4f538e3fa3b3c016b36c84c54722f080a2ed5e699c493ea391b679
+size 176258147

models/word_markov/az_markov_ctx1_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "word",
+  "language": "az",
+  "unique_contexts": 1994233,
+  "total_transitions": 76499213
+}

models/word_markov/az_markov_ctx2_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84c74e51483191748a055d289e1a5aef8772a956f2bf11cba9e7d189ea1b4b32
+size 573585822

models/word_markov/az_markov_ctx2_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "word",
+  "language": "az",
+  "unique_contexts": 17848485,
+  "total_transitions": 76289777
+}

models/word_markov/az_markov_ctx3_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ecd9c2c1f1e9044f86b57b09c32aa0253765d15caddc662a6627016504ef0d6
+size 966856566

models/word_markov/az_markov_ctx3_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "word",
+  "language": "az",
+  "unique_contexts": 39971548,
+  "total_transitions": 76080345
+}

models/word_markov/az_markov_ctx4_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b2ed605da828cb5c4cae04eb068d419b295b7ee8b6e197746cbfb6c60f02aa8
+size 1216484335

models/word_markov/az_markov_ctx4_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "word",
+  "language": "az",
+  "unique_contexts": 53703949,
+  "total_transitions": 75870923
+}

models/word_ngram/az_2gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5b140538fc4f6c175a39196aec356d60fb7278eabb173a6ee0e9a00e793f6bd
+size 24317081

models/word_ngram/az_2gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "word",
+  "language": "az",
+  "unique_ngrams": 1418191,
+  "total_ngrams": 76499213
+}

models/word_ngram/az_3gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0be337c9dcb09a378486e60ffe7df7c2fdfcdb019085fec83df21719c9fe8870
+size 52514526

models/word_ngram/az_3gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "word",
+  "language": "az",
+  "unique_ngrams": 2779660,
+  "total_ngrams": 76289777
+}

models/word_ngram/az_4gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cedd59a0b14806f93960d2fa3c58fc0bcc63c2d7198a36bbd07d210cc0a9e5b9
+size 104278452

models/word_ngram/az_4gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "word",
+  "language": "az",
+  "unique_ngrams": 5336948,
+  "total_ngrams": 76080345
+}