Upload all models and assets for av (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 +1,564 @@
+---
+language: av
+language_name: AV
+language_family: caucasian_northeast
+tags:
+  - wikilangs
+  - nlp
+  - tokenizer
+  - embeddings
+  - n-gram
+  - markov
+  - wikipedia
+  - monolingual
+  - family-caucasian_northeast
+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: 4.583
+  - name: best_isotropy
+    type: isotropy
+    value: 0.8716
+  - name: vocabulary_size
+    type: vocab
+    value: 38576
+generated: 2025-12-27
+---
+
+# AV - Wikilangs Models
+## Comprehensive Research Report & Full Ablation Study
+
+This repository contains NLP models trained and evaluated by Wikilangs, specifically on **AV** 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** | 3.534x | 3.49 | 0.0801% | 219,599 |
+| **16k** | 3.897x | 3.84 | 0.0884% | 199,103 |
+| **32k** | 4.254x | 4.20 | 0.0965% | 182,410 |
+| **64k** | 4.583x 🏆 | 4.52 | 0.1039% | 169,325 |
+
+### Tokenization Examples
+
+Below are sample sentences tokenized with each vocabulary size:
+
+**Sample 1:** `Кванирукъ (латиназул мацӀалда ventriculus) — гӀадамасул лага-черх. 
+
+Категория:Г...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁кван ир укъ ▁( латиназул ▁мацӏалда ▁v ent ric ul ... (+14 more)` | 24 |
+| 16k | `▁кван ир укъ ▁( латиназул ▁мацӏалда ▁v ent ric ulus ... (+13 more)` | 23 |
+| 32k | `▁кванирукъ ▁( латиназул ▁мацӏалда ▁v ent ric ulus ) ▁— ... (+11 more)` | 21 |
+| 64k | `▁кванирукъ ▁( латиназул ▁мацӏалда ▁vent ric ulus ) ▁— ▁гӏадамасул ... (+10 more)` | 20 |
+
+**Sample 2:** `Гудермес ( ) — Россиялъул Буртиялъ жумхӀурияталда бугеб шагьар.
+Сунж-хъалаялдаса...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁г уд ер мес ▁( ▁) ▁— ▁россиялъул ▁бурт иялъ ... (+36 more)` | 46 |
+| 16k | `▁гуд ер мес ▁( ▁) ▁— ▁россиялъул ▁бурт иялъ ▁жум ... (+33 more)` | 43 |
+| 32k | `▁гудермес ▁( ▁) ▁— ▁россиялъул ▁буртиялъ ▁жумхӏ урият алда ▁бугеб ... (+25 more)` | 35 |
+| 64k | `▁гудермес ▁( ▁) ▁— ▁россиялъул ▁буртиялъ ▁жумхӏурият алда ▁бугеб ▁шагьар ... (+22 more)` | 32 |
+
+**Sample 3:** `Лъугьа-бахъинал 
+
+ Гьаруна 
+
+ Хвана 
+
+
+Категория:1927`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁лъугьа - бахъинал ▁гьаруна ▁хвана ▁категория : 1 9 2 ... (+1 more)` | 11 |
+| 16k | `▁лъугьа - бахъинал ▁гьаруна ▁хвана ▁категория : 1 9 2 ... (+1 more)` | 11 |
+| 32k | `▁лъугьа - бахъинал ▁гьаруна ▁хвана ▁категория : 1 9 2 ... (+1 more)` | 11 |
+| 64k | `▁лъугьа - бахъинал ▁гьаруна ▁хвана ▁категория : 1 9 2 ... (+1 more)` | 11 |
+
+
+### Key Findings
+
+- **Best Compression:** 64k achieves 4.583x compression
+- **Lowest UNK Rate:** 8k with 0.0801% 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** | 5,221 🏆 | 12.35 | 14,725 | 20.8% | 49.4% |
+| **2-gram** | 502 🏆 | 8.97 | 5,314 | 55.0% | 94.9% |
+| **3-gram** | 8,074 | 12.98 | 19,718 | 16.9% | 42.5% |
+| **3-gram** | 4,078 | 11.99 | 36,896 | 22.5% | 60.1% |
+| **4-gram** | 18,096 | 14.14 | 39,973 | 12.6% | 31.2% |
+| **4-gram** | 18,482 | 14.17 | 151,649 | 12.4% | 35.7% |
+
+### Top 5 N-grams by Size
+
+**2-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `категория :` | 6,060 |
+| 2 | `) .` | 2,431 |
+| 3 | `) ,` | 2,098 |
+| 4 | `) —` | 1,555 |
+| 5 | `. —` | 1,376 |
+
+**3-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `. география росу` | 645 |
+| 2 | `география росу буго` | 645 |
+| 3 | `. категория :` | 622 |
+| 4 | `мугъчӏваял категория :` | 614 |
+| 5 | `лъугьа - бахъинал` | 597 |
+
+**4-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `. география росу буго` | 630 |
+| 2 | `география росу буго мухъалъул` | 513 |
+| 3 | `. мугъчӏваял категория :` | 483 |
+| 4 | `лъугьа - бахъинал гьаруна` | 471 |
+| 5 | `- бахъинал гьаруна хвана` | 461 |
+
+
+### Key Findings
+
+- **Best Perplexity:** 2-gram with 502
+- **Entropy Trend:** Decreases with larger n-grams (more predictable)
+- **Coverage:** Top-1000 patterns cover ~36% 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** | 0.5741 | 1.489 | 3.65 | 105,500 | 42.6% |
+| **1** | 1.3715 | 2.587 | 12.28 | 1,091 | 0.0% |
+| **2** | 0.1898 | 1.141 | 1.41 | 384,678 | 81.0% |
+| **2** | 1.0636 | 2.090 | 6.00 | 13,391 | 0.0% |
+| **3** | 0.0614 | 1.043 | 1.11 | 542,652 | 93.9% |
+| **3** | 0.8006 | 1.742 | 3.64 | 80,309 | 19.9% |
+| **4** | 0.0249 🏆 | 1.017 | 1.04 | 599,240 | 97.5% |
+| **4** | 0.5368 🏆 | 1.451 | 2.24 | 292,181 | 46.3% |
+
+### Generated Text Samples
+
+Below are text samples generated from each Markov chain model:
+
+**Context Size 1:**
+
+1. `. хундерилифастандартгьо ̄ ⲛ ̄ л . costumes caucasus circassians caucasus . — anatidae хъизан патагӏ...`
+2. `, къагiидаби . амма руго . байрамал лъугьа - бакъбаккул кавказияб календар категория : гардарики ,`
+3. `- абилеб ) мугъчӏваял категория : « вечера на хадидже категория : пко « моноклеr »`
+
+**Context Size 2:**
+
+1. `категория : гӏанди - гӏорул жанилъуда , ралъдал гьумералдаса 1869 метраялъ тіадегіан . хіоралъул тіа...`
+2. `) . эратосфениде ( iii гӏ . байбихьи ) букіана патрикиясул титул , гьелдаса хадуб дагъистаналде .`
+3. `) , лачен ( falco peregrinus ) , продолжительность 2 ч . ii – i гіасрабазул гіорхъода`
+
+**Context Size 3:**
+
+1. `география росу буго мухъалъул центер уркарахъалдаса 50 км - лъ жанубияб бакътӏерхьудехун . демографи...`
+2. `. география росу буго ралъдал гьурматӏама 606 метралъ борхалъуда , мухъалъул центер хунзахъа шималия...`
+3. `. категория : ираналъул останал * категория : азиялъул исламиял хіаракатчагіи категория : тіалибан`
+
+**Context Size 4:**
+
+1. `. география росу буго мухъалъул марказ лъаратӏаса 11 км - алъ шималалдехун . демография ккола моноэт...`
+2. `география росу буго мухъалъул центер дешлахӏаралдаса 13 км - лъ рикӏкӏад . история 1886 соналъул бая...`
+3. `. мугъчӏваял категория : гӏандалазул бол чагӏи категория : кавказалъул имамзаби`
+
+
+### Key Findings
+
+- **Best Predictability:** Context-4 with 97.5% predictability
+- **Branching Factor:** Decreases with context size (more deterministic)
+- **Memory Trade-off:** Larger contexts require more storage (292,181 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 | 38,576 |
+| Total Tokens | 474,364 |
+| Mean Frequency | 12.30 |
+| Median Frequency | 3 |
+| Frequency Std Dev | 81.10 |
+
+### Most Common Words
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | ва | 7,190 |
+| 2 | категория | 6,086 |
+| 3 | буго | 5,703 |
+| 4 | бугеб | 2,911 |
+| 5 | ккола | 2,903 |
+| 6 | росу | 2,847 |
+| 7 | мухъалъул | 2,671 |
+| 8 | гьеб | 2,187 |
+| 9 | дагъистаналъул | 1,923 |
+| 10 | росдал | 1,903 |
+
+### Least Common Words (from vocabulary)
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | уркутамахьи | 2 |
+| 2 | континуумалде | 2 |
+| 3 | къулецӏмаги | 2 |
+| 4 | гьаркӏасуниб | 2 |
+| 5 | махӏарги | 2 |
+| 6 | пилибхиталъул | 2 |
+| 7 | заповедникалда | 2 |
+| 8 | пилибхит | 2 |
+| 9 | лъалъадул | 2 |
+| 10 | хӏанчӏи | 2 |
+
+### Zipf's Law Analysis
+
+| Metric | Value |
+|--------|-------|
+| Zipf Coefficient | 0.9487 |
+| R² (Goodness of Fit) | 0.992879 |
+| Adherence Quality | **excellent** |
+
+### Coverage Analysis
+
+| Top N Words | Coverage |
+|-------------|----------|
+| Top 100 | 22.2% |
+| Top 1,000 | 49.8% |
+| Top 5,000 | 72.6% |
+| Top 10,000 | 82.2% |
+
+### Key Findings
+
+- **Zipf Compliance:** R²=0.9929 indicates excellent adherence to Zipf's law
+- **High Frequency Dominance:** Top 100 words cover 22.2% of corpus
+- **Long Tail:** 28,576 words needed for remaining 17.8% 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** | 12,900 | 32 | 4.114 | 0.854 | 0.8716 🏆 |
+| **mono_64d** | 12,900 | 64 | 4.625 | 0.771 | 0.7752 |
+| **mono_128d** | 12,900 | 128 | 4.775 | 0.759 | 0.3123 |
+| **embeddings_enhanced** | 0 | 0 | 0.000 | 0.000 | 0.0000 |
+
+### Key Findings
+
+- **Best Isotropy:** mono_32d with 0.8716 (more uniform distribution)
+- **Dimension Trade-off:** Higher dimensions capture more semantics but reduce isotropy
+- **Vocabulary Coverage:** All models cover 12,900 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 (4.58x) with low UNK rate |
+| N-gram | **5-gram** | Lowest perplexity (502) |
+| Markov | **Context-4** | Highest predictability (97.5%) |
+| 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 20:39:38*

models/embeddings/monolingual/av_128d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9b6638db57121afe5ec40e5d975a427a9f37c786498e2a2d32874648d6e940b
+size 1037515926

models/embeddings/monolingual/av_128d.meta.json

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

models/embeddings/monolingual/av_128d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "av",
+  "dimension": 128,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 128,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 12900
+}

models/embeddings/monolingual/av_32d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a659bfb895d802a760b82e4bfa4c8d06af1ca7e7e48c9edcbe7eea312922c4c5
+size 259608726

models/embeddings/monolingual/av_32d.meta.json

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

models/embeddings/monolingual/av_32d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "av",
+  "dimension": 32,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 32,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 12900
+}

models/embeddings/monolingual/av_64d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42a486c86184075e6fbcad6c04646238858ba9785c971dda95529c259785636b
+size 518911126

models/embeddings/monolingual/av_64d.meta.json

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

models/embeddings/monolingual/av_64d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "av",
+  "dimension": 64,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 64,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 12900
+}

models/subword_markov/av_markov_ctx1_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80d9f829d3520985c3ec2e555c4f56f88029ba50c0f8e0a8d143025208979f1b
+size 96598

models/subword_markov/av_markov_ctx1_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "subword",
+  "language": "av",
+  "unique_contexts": 1091,
+  "total_transitions": 4387638
+}

models/subword_markov/av_markov_ctx2_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e31801ab66f88e5b5e45a7c74584663475f8a158e42122c638b42657aca01fe
+size 605158

models/subword_markov/av_markov_ctx2_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "subword",
+  "language": "av",
+  "unique_contexts": 13391,
+  "total_transitions": 4383567
+}

models/subword_markov/av_markov_ctx3_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e9124d975e5792b36fc393927806d32e658235b480893ccef7d0dde1592724b
+size 2084996

models/subword_markov/av_markov_ctx3_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "subword",
+  "language": "av",
+  "unique_contexts": 80309,
+  "total_transitions": 4379496
+}

models/subword_markov/av_markov_ctx4_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e309e2636831f32a2161c4e5b0ad24463660acaea5b1b23d507678e2b7e207f
+size 5912238

models/subword_markov/av_markov_ctx4_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "subword",
+  "language": "av",
+  "unique_contexts": 292181,
+  "total_transitions": 4375425
+}

models/subword_ngram/av_2gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4efe6ef223dd1f6472674537d1c596680156ddacece7718273c5a6229338ee6f
+size 67795

models/subword_ngram/av_2gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "subword",
+  "language": "av",
+  "unique_ngrams": 5314,
+  "total_ngrams": 4387638
+}

models/subword_ngram/av_3gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe60524de25890ebbb4579aeb1d0dc4e16f60cf7c9a21aadae938fe7d8a651c0
+size 482417

models/subword_ngram/av_3gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "subword",
+  "language": "av",
+  "unique_ngrams": 36896,
+  "total_ngrams": 4383567
+}

models/subword_ngram/av_4gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0969deb3aaf25140d6a8f6829e34559fe563aa08e2863a2d91328f85f8f16ac0
+size 1846055

models/subword_ngram/av_4gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "subword",
+  "language": "av",
+  "unique_ngrams": 151649,
+  "total_ngrams": 4379496
+}

models/tokenizer/av_tokenizer_16k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37ab57b7217643d6cfb077a937974ffd4dcebcfabf852aa82bfdfe4ce2a0500d
+size 576619

models/tokenizer/av_tokenizer_32k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28d374a06b6b7aa8a4d3941bcd4695e5533544ad636d5d0bb610fb7793bc4354
+size 936914

models/tokenizer/av_tokenizer_64k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5521eccdbb9f46291a027fc56f120b592ff6dfdafcd09218747f5dfebcdc505
+size 1716704

models/tokenizer/av_tokenizer_8k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0fe4689b0c6b4beca83bfdeb9649a3570f36538f43ca7cf4f02c432b94214c16
+size 403337

models/vocabulary/av_vocabulary.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3288037a303bb285d697f0b1691c0479ba6d9e29db6034c70012c05716d329f1
+size 738845

models/vocabulary/av_vocabulary_metadata.json

@@ -0,0 +1,16 @@
+{
+  "language": "av",
+  "vocabulary_size": 38576,
+  "statistics": {
+    "type_token_ratio": 0.19482218346291424,
+    "coverage": {
+      "top_100": 0.1948757649215124,
+      "top_1000": 0.43634707482188784,
+      "top_5000": 0.6366234812427942,
+      "top_10000": 0.7208073432465191
+    },
+    "hapax_count": 66868,
+    "hapax_ratio": 0.6341565191001859,
+    "total_documents": 4071
+  }
+}

models/word_markov/av_markov_ctx1_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63f373412e062fd6835774cb1266eaa3e245850dba38a8e6151d4ac7a4759ede
+size 5274705

models/word_markov/av_markov_ctx1_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "word",
+  "language": "av",
+  "unique_contexts": 105500,
+  "total_transitions": 728256
+}

models/word_markov/av_markov_ctx2_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb88d7a9d6079d49efd1ad1b30db00e9d3c96aab81c84a0f8d0ef679e10fe3d4
+size 11187735

models/word_markov/av_markov_ctx2_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "word",
+  "language": "av",
+  "unique_contexts": 384678,
+  "total_transitions": 724185
+}

models/word_markov/av_markov_ctx3_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:907faf6f55ecd04d74aa3d7f21d0111dda47ad2d999e00d289b0665dfe132da5
+size 14841527

models/word_markov/av_markov_ctx3_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "word",
+  "language": "av",
+  "unique_contexts": 542652,
+  "total_transitions": 720116
+}

models/word_markov/av_markov_ctx4_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:910fed69ecfd672367e2d66b7147fecc45ddedd0acda63f49e00c5b1add64c18
+size 17509163

models/word_markov/av_markov_ctx4_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "word",
+  "language": "av",
+  "unique_contexts": 599240,
+  "total_transitions": 716050
+}

models/word_ngram/av_2gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6cb354b1f97cbdfc6d757a9c19d07428b4d51d46a086e2f6d4633877e66a449f
+size 309626

models/word_ngram/av_2gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "word",
+  "language": "av",
+  "unique_ngrams": 14725,
+  "total_ngrams": 728256
+}

models/word_ngram/av_3gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2a68230c75c38fad71bc4e8c7c054c54a5ac1bae080c287811284115b39aee8
+size 461245

models/word_ngram/av_3gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "word",
+  "language": "av",
+  "unique_ngrams": 19718,
+  "total_ngrams": 724185
+}

models/word_ngram/av_4gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a2c985bdc681c13eadcce7267033132b7282125a774fd8af76143f2674a46d5
+size 1007425

models/word_ngram/av_4gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "word",
+  "language": "av",
+  "unique_ngrams": 39973,
+  "total_ngrams": 720116
+}