Upload all models and assets for af (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 +1,566 @@
+---
+language: af
+language_name: AF
+language_family: germanic_west_anglofrisian
+tags:
+  - wikilangs
+  - nlp
+  - tokenizer
+  - embeddings
+  - n-gram
+  - markov
+  - wikipedia
+  - monolingual
+  - family-germanic_west_anglofrisian
+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: 4.193
+  - name: best_isotropy
+    type: isotropy
+    value: 0.6686
+  - name: vocabulary_size
+    type: vocab
+    value: 426702
+generated: 2025-12-27
+---
+
+# AF - Wikilangs Models
+## Comprehensive Research Report & Full Ablation Study
+
+This repository contains NLP models trained and evaluated by Wikilangs, specifically on **AF** 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** | 3.512x | 3.48 | 0.0703% | 1,403,468 |
+| **16k** | 3.805x | 3.77 | 0.0762% | 1,295,239 |
+| **32k** | 4.029x | 3.99 | 0.0807% | 1,223,476 |
+| **64k** | 4.193x 🏆 | 4.15 | 0.0840% | 1,175,389 |
+
+### Tokenization Examples
+
+Below are sample sentences tokenized with each vocabulary size:
+
+**Sample 1:** `Japan Nasionale Roete 461 is 'n nasionale snelweg in Japan.
+
+ Verwysings 
+
+Kateg...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁japan ▁nasionale ▁roete ▁ 4 6 1 ▁is ▁' n ... (+13 more)` | 23 |
+| 16k | `▁japan ▁nasionale ▁roete ▁ 4 6 1 ▁is ▁' n ... (+12 more)` | 22 |
+| 32k | `▁japan ▁nasionale ▁roete ▁ 4 6 1 ▁is ▁' n ... (+12 more)` | 22 |
+| 64k | `▁japan ▁nasionale ▁roete ▁ 4 6 1 ▁is ▁' n ... (+12 more)` | 22 |
+
+**Sample 2:** `Japan Nasionale Roete 239 is 'n nasionale snelweg in Japan.
+
+ Verwysings 
+
+Kateg...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁japan ▁nasionale ▁roete ▁ 2 3 9 ▁is ▁' n ... (+13 more)` | 23 |
+| 16k | `▁japan ▁nasionale ▁roete ▁ 2 3 9 ▁is ▁' n ... (+12 more)` | 22 |
+| 32k | `▁japan ▁nasionale ▁roete ▁ 2 3 9 ▁is ▁' n ... (+12 more)` | 22 |
+| 64k | `▁japan ▁nasionale ▁roete ▁ 2 3 9 ▁is ▁' n ... (+12 more)` | 22 |
+
+**Sample 3:** `Japan Nasionale Roete 264 is 'n nasionale snelweg in Japan.
+
+ Verwysings 
+
+Kateg...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁japan ▁nasionale ▁roete ▁ 2 6 4 ▁is ▁' n ... (+13 more)` | 23 |
+| 16k | `▁japan ▁nasionale ▁roete ▁ 2 6 4 ▁is ▁' n ... (+12 more)` | 22 |
+| 32k | `▁japan ▁nasionale ▁roete ▁ 2 6 4 ▁is ▁' n ... (+12 more)` | 22 |
+| 64k | `▁japan ▁nasionale ▁roete ▁ 2 6 4 ▁is ▁' n ... (+12 more)` | 22 |
+
+
+### Key Findings
+
+- **Best Compression:** 64k achieves 4.193x compression
+- **Lowest UNK Rate:** 8k with 0.0703% 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** | 54,654 🏆 | 15.74 | 884,766 | 16.1% | 32.6% |
+| **2-gram** | 304 🏆 | 8.25 | 15,856 | 65.2% | 98.7% |
+| **3-gram** | 293,194 | 18.16 | 2,174,878 | 7.1% | 19.5% |
+| **3-gram** | 2,671 | 11.38 | 128,050 | 26.3% | 68.2% |
+| **4-gram** | 729,597 | 19.48 | 4,032,808 | 5.3% | 15.8% |
+| **4-gram** | 15,548 | 13.92 | 732,288 | 14.1% | 38.0% |
+
+### Top 5 N-grams by Size
+
+**2-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `' n` | 602,690 |
+| 2 | `van die` | 511,614 |
+| 3 | `in die` | 336,247 |
+| 4 | `. die` | 312,198 |
+| 5 | `kategorie :` | 309,571 |
+
+**3-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `is ' n` | 99,632 |
+| 2 | `suid - afrika` | 65,103 |
+| 3 | `kategorie : geboortes` | 42,714 |
+| 4 | `: geboortes in` | 42,714 |
+| 5 | `skakels kategorie :` | 36,080 |
+
+**4-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `kategorie : geboortes in` | 42,714 |
+| 2 | `kategorie : flora van` | 35,480 |
+| 3 | `eksterne skakels kategorie :` | 34,452 |
+| 4 | `) is ' n` | 32,521 |
+| 5 | `van suid - afrika` | 27,043 |
+
+
+### Key Findings
+
+- **Best Perplexity:** 2-gram with 304
+- **Entropy Trend:** Decreases with larger n-grams (more predictable)
+- **Coverage:** Top-1000 patterns cover ~38% 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** | 0.7872 | 1.726 | 8.26 | 1,012,367 | 21.3% |
+| **1** | 1.3494 | 2.548 | 8.73 | 6,082 | 0.0% |
+| **2** | 0.4249 | 1.342 | 2.60 | 8,362,191 | 57.5% |
+| **2** | 0.8337 | 1.782 | 5.41 | 53,087 | 16.6% |
+| **3** | 0.2059 | 1.153 | 1.52 | 21,746,451 | 79.4% |
+| **3** | 0.8222 | 1.768 | 4.50 | 287,186 | 17.8% |
+| **4** | 0.0978 🏆 | 1.070 | 1.19 | 33,034,213 | 90.2% |
+| **4** | 0.7160 🏆 | 1.643 | 3.45 | 1,293,510 | 28.4% |
+
+### Generated Text Samples
+
+Below are text samples generated from each Markov chain model:
+
+**Context Size 1:**
+
+1. `die ontwerp , babergh ( links | 297x297px | leigh het drie mense kategorie : eine`
+2. `. ( 1965 : " . die pos het op die vrye oefening 3 2 -`
+3. `, vacelet , belfast aangesluit . . die epi thēbas ; hierdie styl eskarp voor 1919`
+
+**Context Size 2:**
+
+1. `' n wortelkelder te bêre ) is ’ n skynbare magnitude van 3 830 in 1793 het`
+2. `van die koninklike saoedi lugmag ( usaf ) het magtiging gehad om die aanstelling van chailly se`
+3. `in die tydperk tot in 1955 het motorola die elektroplateringsmetode ontwikkel , veral teenoor die st...`
+
+**Context Size 3:**
+
+1. `is ' n aansteeklike siekte wat met die egiptiese heilige skrif , en regverdigheidsin daar nooit enig...`
+2. `suid - afrika internasionaal verteenwoordig en puik presteer . daar was ook geen ander kubistiese sk...`
+3. `: geboortes in 1917 kategorie : sterftes in 1932 kategorie : sterftes in 1547`
+
+**Context Size 4:**
+
+1. `kategorie : geboortes in 1983 kategorie : engelse aktrises van die 21ste eeu kategorie : amerikaanse...`
+2. `kategorie : flora van egipte kategorie : mense in die tweede vryheidsoorlog . bron raper , peter edm...`
+3. `eksterne skakels kategorie : lewende mense kategorie : engelse manlike akteurs van die 20ste eeu kat...`
+
+
+### Key Findings
+
+- **Best Predictability:** Context-4 with 90.2% predictability
+- **Branching Factor:** Decreases with context size (more deterministic)
+- **Memory Trade-off:** Larger contexts require more storage (1,293,510 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 | 426,702 |
+| Total Tokens | 41,728,231 |
+| Mean Frequency | 97.79 |
+| Median Frequency | 4 |
+| Frequency Std Dev | 6032.22 |
+
+### Most Common Words
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | die | 2,854,279 |
+| 2 | van | 1,329,400 |
+| 3 | in | 1,119,312 |
+| 4 | en | 1,055,574 |
+| 5 | n | 811,721 |
+| 6 | is | 769,576 |
+| 7 | het | 649,041 |
+| 8 | wat | 345,008 |
+| 9 | kategorie | 310,570 |
+| 10 | the | 296,225 |
+
+### Least Common Words (from vocabulary)
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | cdmtcs | 2 |
+| 2 | researchreports | 2 |
+| 3 | akteurskategorieë | 2 |
+| 4 | mullens | 2 |
+| 5 | grafiekstruktuur | 2 |
+| 6 | roostergrafieke | 2 |
+| 7 | sokkerbekertitels | 2 |
+| 8 | chalobah | 2 |
+| 9 | sentrumverdediger | 2 |
+| 10 | guðjohnsen | 2 |
+
+### Zipf's Law Analysis
+
+| Metric | Value |
+|--------|-------|
+| Zipf Coefficient | 1.0756 |
+| R² (Goodness of Fit) | 0.994236 |
+| Adherence Quality | **excellent** |
+
+### Coverage Analysis
+
+| Top N Words | Coverage |
+|-------------|----------|
+| Top 100 | 41.8% |
+| Top 1,000 | 63.9% |
+| Top 5,000 | 79.6% |
+| Top 10,000 | 85.2% |
+
+### Key Findings
+
+- **Zipf Compliance:** R²=0.9942 indicates excellent adherence to Zipf's law
+- **High Frequency Dominance:** Top 100 words cover 41.8% of corpus
+- **Long Tail:** 416,702 words needed for remaining 14.8% 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** | 281,289 | 32 | 3.583 | 1.675 | 0.6651 |
+| **mono_64d** | 281,289 | 64 | 3.983 | 1.616 | 0.6686 🏆 |
+| **mono_128d** | 281,289 | 128 | 4.471 | 1.595 | 0.6515 |
+| **embeddings_enhanced** | 0 | 0 | 0.000 | 0.000 | 0.0000 |
+
+### Key Findings
+
+- **Best Isotropy:** mono_64d with 0.6686 (more uniform distribution)
+- **Dimension Trade-off:** Higher dimensions capture more semantics but reduce isotropy
+- **Vocabulary Coverage:** All models cover 281,289 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 (4.19x) with low UNK rate |
+| N-gram | **5-gram** | Lowest perplexity (304) |
+| Markov | **Context-4** | Highest predictability (90.2%) |
+| 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 05:32:18*

models/embeddings/monolingual/af_128d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1664aff332c7a349b9940cce2ba0d366d41d2cd3ceeddc6b04037c36e8b7ab30
+size 1317292760

models/embeddings/monolingual/af_128d.meta.json

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

models/embeddings/monolingual/af_128d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "af",
+  "dimension": 128,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 128,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 281289
+}

models/embeddings/monolingual/af_32d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0e08a6e56419bc15cf41c317b15d8f38d74f34682f19a2349f9305f60573c26
+size 333262808

models/embeddings/monolingual/af_32d.meta.json

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

models/embeddings/monolingual/af_32d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "af",
+  "dimension": 32,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 32,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 281289
+}

models/embeddings/monolingual/af_64d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f17d4f4c541f08173c83987cd4b8539b48d2fa16f305ad30acc059e9b365735a
+size 661272792

models/embeddings/monolingual/af_64d.meta.json

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

models/embeddings/monolingual/af_64d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "af",
+  "dimension": 64,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 64,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 281289
+}

models/subword_markov/af_markov_ctx1_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f42ef48ee495f26f7a3063da48c9ee906a4205528b9506ab33ed9fc982d9111c
+size 366862

models/subword_markov/af_markov_ctx1_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "subword",
+  "language": "af",
+  "unique_contexts": 6082,
+  "total_transitions": 268485278
+}

models/subword_markov/af_markov_ctx2_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c91a87c97f14e34aeb8427bc36ed9584e434a40332a3f97910ffc78c5417d6b7
+size 2262231

models/subword_markov/af_markov_ctx2_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "subword",
+  "language": "af",
+  "unique_contexts": 53087,
+  "total_transitions": 268357949
+}

models/subword_markov/af_markov_ctx3_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:994eb0d42a5e555f306278b50da3fd9f4347eccf479e465f5537de98e4429a31
+size 10525742

models/subword_markov/af_markov_ctx3_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "subword",
+  "language": "af",
+  "unique_contexts": 287186,
+  "total_transitions": 268230620
+}

models/subword_markov/af_markov_ctx4_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1cda0c6adb1b4b36a5836b96ba23dc2c9f880706859d0799461ae0a2a1d05e1a
+size 35196938

models/subword_markov/af_markov_ctx4_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "subword",
+  "language": "af",
+  "unique_contexts": 1293510,
+  "total_transitions": 268103291
+}

models/subword_ngram/af_2gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a314440c3a6118bd2558c0f5ae77b9b7567e26e683b45b28da4d7b61c5985153
+size 212926

models/subword_ngram/af_2gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "subword",
+  "language": "af",
+  "unique_ngrams": 15856,
+  "total_ngrams": 268485278
+}

models/subword_ngram/af_3gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b94502700c29b4beab32c042c6cbfb0121c688512ef843b491c7da0e3591e6a
+size 1557780

models/subword_ngram/af_3gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "subword",
+  "language": "af",
+  "unique_ngrams": 128050,
+  "total_ngrams": 268357949
+}

models/subword_ngram/af_4gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7e9ec21d12eea1d385fb08e137b0390323ad38d5af55693ca5ec808d63eeb8e
+size 8462003

models/subword_ngram/af_4gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "subword",
+  "language": "af",
+  "unique_ngrams": 732288,
+  "total_ngrams": 268230620
+}

models/tokenizer/af_tokenizer_16k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f743e4753c9efba2d99338cf7ad44c6a473a342a46e4df9671acf4191ba7afd
+size 508391

models/tokenizer/af_tokenizer_32k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf99915b4dc311a913ef83532e1b5909f83560e280ce3c1ec65b0482070c3bdd
+size 789416

models/tokenizer/af_tokenizer_64k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ae7229d1b022d86ec54a1b3e2c2739e76aa8b7c1327e3cd4f47c0530c99afff
+size 1366342

models/tokenizer/af_tokenizer_8k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ebbd1727f82be8bbf58a61f934983bfa3f23d2f65c59c9c6d62ab68017db7ba
+size 372400

models/vocabulary/af_vocabulary.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:abad63eeefcdc78640a7f186a0aff7fc6bad429ed352ee2a07643f9ccbe0379e
+size 6727061

models/vocabulary/af_vocabulary_metadata.json

@@ -0,0 +1,16 @@
+{
+  "language": "af",
+  "vocabulary_size": 426702,
+  "statistics": {
+    "type_token_ratio": 0.023913563254074188,
+    "coverage": {
+      "top_100": 0.4125284635056328,
+      "top_1000": 0.6301073515896018,
+      "top_5000": 0.7848384789184834,
+      "top_10000": 0.8399447900573702
+    },
+    "hapax_count": 585162,
+    "hapax_ratio": 0.5783010365029293,
+    "total_documents": 127329
+  }
+}

models/word_markov/af_markov_ctx1_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9712c1db6a1f679bed53a9525fc4d37129d4817df0e1662701aba5781263ec1d
+size 74330320

models/word_markov/af_markov_ctx1_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "word",
+  "language": "af",
+  "unique_contexts": 1012367,
+  "total_transitions": 52392040
+}

models/word_markov/af_markov_ctx2_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:465d77d2f538e56b8c138515df12efeb4c4dfda8fbb0f2bd4cf5cf04ed0d532a
+size 231993381

models/word_markov/af_markov_ctx2_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "word",
+  "language": "af",
+  "unique_contexts": 8362191,
+  "total_transitions": 52264712
+}

models/word_markov/af_markov_ctx3_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d8d8d6ec490aff768a6e7268a3a3a2ecb9e17e4e3a136bb8ba504fef57669f3
+size 430211054

models/word_markov/af_markov_ctx3_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "word",
+  "language": "af",
+  "unique_contexts": 21746451,
+  "total_transitions": 52137387
+}

models/word_markov/af_markov_ctx4_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:375d27128bfbbc7da4eb573e0dc99e9765bb9855979e6b5b65f4711973a8564f
+size 585301615

models/word_markov/af_markov_ctx4_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "word",
+  "language": "af",
+  "unique_contexts": 33034213,
+  "total_transitions": 52010063
+}

models/word_ngram/af_2gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13c0ceecb17348926c4a01efeeca2e5b09313fb6b9ebcb6509bf450f38b1768b
+size 12049266

models/word_ngram/af_2gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "word",
+  "language": "af",
+  "unique_ngrams": 884766,
+  "total_ngrams": 52392040
+}

models/word_ngram/af_3gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1bea1ba20ff8534a9328f94766681494a3d52c03a132b5961ecec14dcdfa5f4e
+size 31811393

models/word_ngram/af_3gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "word",
+  "language": "af",
+  "unique_ngrams": 2174878,
+  "total_ngrams": 52264712
+}

models/word_ngram/af_4gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e84236aa01f40edbdaf74357b179a706c758462c276c3af1e0755f191f11a1f
+size 62258431

models/word_ngram/af_4gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "word",
+  "language": "af",
+  "unique_ngrams": 4032808,
+  "total_ngrams": 52137387
+}