Upload all models and assets for bar (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 +1,562 @@
+---
+language: bar
+language_name: BAR
+language_family: germanic_west_continental
+tags:
+  - wikilangs
+  - nlp
+  - tokenizer
+  - embeddings
+  - n-gram
+  - markov
+  - wikipedia
+  - monolingual
+  - family-germanic_west_continental
+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: 3.790
+  - name: best_isotropy
+    type: isotropy
+    value: 0.8361
+  - name: vocabulary_size
+    type: vocab
+    value: 225914
+generated: 2025-12-28
+---
+
+# BAR - Wikilangs Models
+## Comprehensive Research Report & Full Ablation Study
+
+This repository contains NLP models trained and evaluated by Wikilangs, specifically on **BAR** 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** | 3.051x | 3.01 | 0.0348% | 1,184,756 |
+| **16k** | 3.320x | 3.27 | 0.0378% | 1,088,767 |
+| **32k** | 3.568x | 3.52 | 0.0407% | 1,013,044 |
+| **64k** | 3.790x 🏆 | 3.74 | 0.0432% | 953,541 |
+
+### Tokenization Examples
+
+Below are sample sentences tokenized with each vocabulary size:
+
+**Sample 1:** `Platte County is a County in Nebraska in da USA.
+
+ Beleg 
+
+ Im Netz 
+
+Kategorie:...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁plat te ▁county ▁is ▁a ▁county ▁in ▁nebraska ▁in ▁da ... (+10 more)` | 20 |
+| 16k | `▁plat te ▁county ▁is ▁a ▁county ▁in ▁nebraska ▁in ▁da ... (+10 more)` | 20 |
+| 32k | `▁platte ▁county ▁is ▁a ▁county ▁in ▁nebraska ▁in ▁da ▁usa ... (+9 more)` | 19 |
+| 64k | `▁platte ▁county ▁is ▁a ▁county ▁in ▁nebraska ▁in ▁da ▁usa ... (+9 more)` | 19 |
+
+**Sample 2:** `Union County. Obgruafa am 22. Feba 2011 is a County in South Carolina in da USA....`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁union ▁county . ▁obgruafa ▁am ▁ 2 2 . ▁feba ... (+24 more)` | 34 |
+| 16k | `▁union ▁county . ▁obgruafa ▁am ▁ 2 2 . ▁feba ... (+24 more)` | 34 |
+| 32k | `▁union ▁county . ▁obgruafa ▁am ▁ 2 2 . ▁feba ... (+24 more)` | 34 |
+| 64k | `▁union ▁county . ▁obgruafa ▁am ▁ 2 2 . ▁feba ... (+24 more)` | 34 |
+
+**Sample 3:** `Des is a Iwablick iwas Joar 1561.
+
+ Im Netz`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁des ▁is ▁a ▁iwablick ▁iwas ▁joar ▁ 1 5 6 ... (+4 more)` | 14 |
+| 16k | `▁des ▁is ▁a ▁iwablick ▁iwas ▁joar ▁ 1 5 6 ... (+4 more)` | 14 |
+| 32k | `▁des ▁is ▁a ▁iwablick ▁iwas ▁joar ▁ 1 5 6 ... (+4 more)` | 14 |
+| 64k | `▁des ▁is ▁a ▁iwablick ▁iwas ▁joar ▁ 1 5 6 ... (+4 more)` | 14 |
+
+
+### Key Findings
+
+- **Best Compression:** 64k achieves 3.790x compression
+- **Lowest UNK Rate:** 8k with 0.0348% 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** | 26,677 🏆 | 14.70 | 159,756 | 14.4% | 35.4% |
+| **2-gram** | 438 🏆 | 8.77 | 9,188 | 56.6% | 97.2% |
+| **3-gram** | 61,390 | 15.91 | 246,468 | 9.0% | 25.7% |
+| **3-gram** | 4,733 | 12.21 | 83,186 | 18.8% | 56.7% |
+| **4-gram** | 99,269 | 16.60 | 386,315 | 9.1% | 23.6% |
+| **4-gram** | 33,953 | 15.05 | 473,217 | 8.7% | 26.5% |
+
+### Top 5 N-grams by Size
+
+**2-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `kategorie :` | 38,959 |
+| 2 | `vo da` | 26,615 |
+| 3 | `is a` | 23,040 |
+| 4 | `in da` | 22,485 |
+| 5 | `. de` | 21,082 |
+
+**3-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `beleg kategorie :` | 6,999 |
+| 2 | `isbn 3 -` | 5,924 |
+| 3 | `. im netz` | 5,460 |
+| 4 | `kategorie : ort` | 5,121 |
+| 5 | `| | |` | 4,937 |
+
+**4-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `, isbn 3 -` | 4,499 |
+| 2 | `| align = "` | 3,879 |
+| 3 | `align = " center` | 3,590 |
+| 4 | `= " center "` | 3,590 |
+| 5 | `kategorie : ort im` | 3,505 |
+
+
+### Key Findings
+
+- **Best Perplexity:** 2-gram with 438
+- **Entropy Trend:** Decreases with larger n-grams (more predictable)
+- **Coverage:** Top-1000 patterns cover ~27% 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** | 0.6074 | 1.523 | 4.85 | 638,594 | 39.3% |
+| **1** | 1.0706 | 2.100 | 7.75 | 3,379 | 0.0% |
+| **2** | 0.2487 | 1.188 | 1.74 | 3,093,978 | 75.1% |
+| **2** | 0.9487 | 1.930 | 6.44 | 26,199 | 5.1% |
+| **3** | 0.0996 | 1.071 | 1.21 | 5,375,695 | 90.0% |
+| **3** | 0.9366 | 1.914 | 4.86 | 168,731 | 6.3% |
+| **4** | 0.0401 🏆 | 1.028 | 1.07 | 6,470,090 | 96.0% |
+| **4** | 0.7672 🏆 | 1.702 | 3.38 | 820,482 | 23.3% |
+
+### Generated Text Samples
+
+Below are text samples generated from each Markov chain model:
+
+**Context Size 1:**
+
+1. `. 524 angewachsen und die fläche vo dera zoagt , par les ' n ) u`
+2. `, isbn 978 - mal 1920 bis auf des is kemnath ( " franz kafka .`
+3. `- deitscha schauspuia und a zaumgroida schdrudl is ois broad bekaunnt hans ( 2016 saha air`
+
+**Context Size 2:**
+
+1. `kategorie : artikel auf niederösterreichisch kategorie : ortsteil von wieseth kategorie : geboren 18...`
+2. `vo da mathematik , schau gorkhaland`
+3. `is a bruck ' nschlåg ; da linné aun an rio xingu . as gericht håtn zu`
+
+**Context Size 3:**
+
+1. `beleg kategorie : johann nestroy ; stücke 21 . s . highway 84 mindt . uma 32 kilometa`
+2. `isbn 3 - 417 - 20675 - 8 ( teubner - studienbücher der geographie ) . im joa`
+3. `kategorie : ort auf den färöern kategorie : streymoy`
+
+**Context Size 4:**
+
+1. `, isbn 3 - 406 - 46224 - 3 birgit zotz : destination tibet . touristisches image zwischen politik`
+2. `| align = " center " | | - | berleu | | | | | | | |`
+3. `= " center " | | | align = " center " | | | fatututa | | align`
+
+
+### Key Findings
+
+- **Best Predictability:** Context-4 with 96.0% predictability
+- **Branching Factor:** Decreases with context size (more deterministic)
+- **Memory Trade-off:** Larger contexts require more storage (820,482 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 | 225,914 |
+| Total Tokens | 5,874,699 |
+| Mean Frequency | 26.00 |
+| Median Frequency | 3 |
+| Frequency Std Dev | 709.74 |
+
+### Most Common Words
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | de | 139,734 |
+| 2 | da | 136,994 |
+| 3 | und | 120,375 |
+| 4 | in | 102,834 |
+| 5 | a | 93,585 |
+| 6 | vo | 92,275 |
+| 7 | is | 88,045 |
+| 8 | im | 71,546 |
+| 9 | kategorie | 39,103 |
+| 10 | des | 34,614 |
+
+### Least Common Words (from vocabulary)
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | mechanisches | 2 |
+| 2 | stabilisierungssystem | 2 |
+| 3 | voeffentlecht | 2 |
+| 4 | innpuls | 2 |
+| 5 | buagstej | 2 |
+| 6 | nuwenburg | 2 |
+| 7 | kulturweges | 2 |
+| 8 | spessartprojektes | 2 |
+| 9 | terrassnfermig | 2 |
+| 10 | tuamhigi | 2 |
+
+### Zipf's Law Analysis
+
+| Metric | Value |
+|--------|-------|
+| Zipf Coefficient | 0.9896 |
+| R² (Goodness of Fit) | 0.999155 |
+| Adherence Quality | **excellent** |
+
+### Coverage Analysis
+
+| Top N Words | Coverage |
+|-------------|----------|
+| Top 100 | 32.7% |
+| Top 1,000 | 54.6% |
+| Top 5,000 | 70.2% |
+| Top 10,000 | 76.9% |
+
+### Key Findings
+
+- **Zipf Compliance:** R²=0.9992 indicates excellent adherence to Zipf's law
+- **High Frequency Dominance:** Top 100 words cover 32.7% of corpus
+- **Long Tail:** 215,914 words needed for remaining 23.1% coverage
+
+---
+## 5. Word Embeddings Evaluation
+
+![Embedding Isotropy](visualizations/embedding_isotropy.png)
+
+![Similarity Matrix](visualizations/embedding_similarity.png)
+
+![t-SNE Words](visualizations/tsne_words.png)
+
+![t-SNE Sentences](visualizations/tsne_sentences.png)
+
+### Model Comparison
+
+| Model | Vocab Size | Dimension | Avg Norm | Std Norm | Isotropy |
+|-------|------------|-----------|----------|----------|----------|
+| **mono_32d** | 92,573 | 32 | 3.954 | 1.316 | 0.8131 |
+| **mono_64d** | 92,573 | 64 | 4.642 | 1.256 | 0.8361 🏆 |
+| **mono_128d** | 92,573 | 128 | 5.543 | 1.143 | 0.8310 |
+| **embeddings_enhanced** | 0 | 0 | 0.000 | 0.000 | 0.0000 |
+
+### Key Findings
+
+- **Best Isotropy:** mono_64d with 0.8361 (more uniform distribution)
+- **Dimension Trade-off:** Higher dimensions capture more semantics but reduce isotropy
+- **Vocabulary Coverage:** All models cover 92,573 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 (3.79x) with low UNK rate |
+| N-gram | **5-gram** | Lowest perplexity (438) |
+| Markov | **Context-4** | Highest predictability (96.0%) |
+| 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: 2025-12-28 00:09:41*

models/embeddings/monolingual/bar_128d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b67e002458c6ffe3851456e1d214d989436508fafb3940aa0a7aeac7f9501600
+size 1120431691

models/embeddings/monolingual/bar_128d.meta.json

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

models/embeddings/monolingual/bar_128d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "bar",
+  "dimension": 128,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 128,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 92573
+}

models/embeddings/monolingual/bar_32d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d60a4fc93329fd2886ca8100a062412d4d07afea28e09e264dc800ac247d496b
+size 281335627

models/embeddings/monolingual/bar_32d.meta.json

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

models/embeddings/monolingual/bar_32d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "bar",
+  "dimension": 32,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 32,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 92573
+}

models/embeddings/monolingual/bar_64d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8669bc90711c772638673e6ddeca2a5ceb915402131cac9b7c3628c941baa702
+size 561034315

models/embeddings/monolingual/bar_64d.meta.json

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

models/embeddings/monolingual/bar_64d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "bar",
+  "dimension": 64,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 64,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 92573
+}

models/subword_markov/bar_markov_ctx1_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60e1fcddc267b2ddb4763a4f04458f5fc0a3864d4d3b65b068a66791c6d11ca9
+size 199512

models/subword_markov/bar_markov_ctx1_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "subword",
+  "language": "bar",
+  "unique_contexts": 3379,
+  "total_transitions": 41526016
+}

models/subword_markov/bar_markov_ctx2_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44ac0dfee6ebc508b0694bd88f8515a3ecd00e7b531cab5055a6e18d776b7900
+size 1274970

models/subword_markov/bar_markov_ctx2_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "subword",
+  "language": "bar",
+  "unique_contexts": 26199,
+  "total_transitions": 41498726
+}

models/subword_markov/bar_markov_ctx3_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dab3f88d8a597024d6c2d45205f49335376cff21c6acc22705b1aeb5a3d6d9b7
+size 5898144

models/subword_markov/bar_markov_ctx3_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "subword",
+  "language": "bar",
+  "unique_contexts": 168731,
+  "total_transitions": 41471436
+}

models/subword_markov/bar_markov_ctx4_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7d35e1ed6aa219268b066af5ab218034a029fe871fbf27a8621d574e9ea9bf2
+size 20162408

models/subword_markov/bar_markov_ctx4_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "subword",
+  "language": "bar",
+  "unique_contexts": 820482,
+  "total_transitions": 41444146
+}

models/subword_ngram/bar_2gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98fc1ea2db56ddb0c1e4d55f5df47bce76f4c454fad62a3186181d6e59bf7819
+size 120210

models/subword_ngram/bar_2gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "subword",
+  "language": "bar",
+  "unique_ngrams": 9188,
+  "total_ngrams": 41526016
+}

models/subword_ngram/bar_3gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb5ad1ec42d5688187fa2244bcea28e012a96c3954cbff00e1919cc637e39cfa
+size 980179

models/subword_ngram/bar_3gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "subword",
+  "language": "bar",
+  "unique_ngrams": 83186,
+  "total_ngrams": 41498726
+}

models/subword_ngram/bar_4gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74cb0c9a29de3bf166be1817974bf03a79e844a9172dca886d1e6c11444fe105
+size 5250784

models/subword_ngram/bar_4gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "subword",
+  "language": "bar",
+  "unique_ngrams": 473217,
+  "total_ngrams": 41471436
+}

models/tokenizer/bar_tokenizer_16k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86990145267f2f23bece3cf170063df25b610b0172bb06154c137114776c8277
+size 497891

models/tokenizer/bar_tokenizer_32k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75f629bdc6954475d208780fe096b578aa8ab68a13993884c783d727d30b36c9
+size 769242

models/tokenizer/bar_tokenizer_64k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3817f1c29c6409fc0d09a5641df6d546b4d73373e90aff314bd676d06f34afe9
+size 1329778

models/tokenizer/bar_tokenizer_8k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc3653e3a2ae044e86fc99a6e9f56a602c7345188f2a36c7e517e330131174fa
+size 367058

models/vocabulary/bar_vocabulary.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c50b70ce0ccf85aa73d5c98c50c1637e81c02c9079bae2cfca6036b4a7fd444
+size 3610908

models/vocabulary/bar_vocabulary_metadata.json

@@ -0,0 +1,16 @@
+{
+  "language": "bar",
+  "vocabulary_size": 225914,
+  "statistics": {
+    "type_token_ratio": 0.1016273267091797,
+    "coverage": {
+      "top_100": 0.3056598360284214,
+      "top_1000": 0.5100686297601529,
+      "top_5000": 0.6556657283942213,
+      "top_10000": 0.7183725874101852
+    },
+    "hapax_count": 413098,
+    "hapax_ratio": 0.6464636031874206,
+    "total_documents": 27290
+  }
+}

models/word_markov/bar_markov_ctx1_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b3dd31c7a802cfbef574568e2ba6b615e18144bf29db409393fb10850929bd7d
+size 30129994

models/word_markov/bar_markov_ctx1_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "word",
+  "language": "bar",
+  "unique_contexts": 638594,
+  "total_transitions": 8127466
+}

models/word_markov/bar_markov_ctx2_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3db5d683eb8fa5679d0baab234482104310aed5348832ff37632b96f81bac464
+size 67578257

models/word_markov/bar_markov_ctx2_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "word",
+  "language": "bar",
+  "unique_contexts": 3093978,
+  "total_transitions": 8100177
+}

models/word_markov/bar_markov_ctx3_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9253b3e64a7a3aa9bd84e29a7bf91e310187acd7b81345ac3362887b3c6a9f7b
+size 96618918

models/word_markov/bar_markov_ctx3_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "word",
+  "language": "bar",
+  "unique_contexts": 5375695,
+  "total_transitions": 8072889
+}

models/word_markov/bar_markov_ctx4_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab88f182fc703c53bba76e38edc370a4c8ed1ad8a0ec15d58155beab90c2d334
+size 113834013

models/word_markov/bar_markov_ctx4_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "word",
+  "language": "bar",
+  "unique_contexts": 6470090,
+  "total_transitions": 8045602
+}

models/word_ngram/bar_2gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:305be2b4a731fc13d78d8a31b51d838be655ee3899d4a3479c5d81c5bc114a10
+size 2084252

models/word_ngram/bar_2gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "word",
+  "language": "bar",
+  "unique_ngrams": 159756,
+  "total_ngrams": 8127466
+}

models/word_ngram/bar_3gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:817288d320b289c5f49b63cb4084a8d5c1cfaf5535e133621f6117c6ac86bce6
+size 3391005

models/word_ngram/bar_3gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "word",
+  "language": "bar",
+  "unique_ngrams": 246468,
+  "total_ngrams": 8100177
+}

models/word_ngram/bar_4gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70d5bc187fee7c4a76f7ca6f6f46bdc686870760f287cface360e35199b8986c
+size 5697675

models/word_ngram/bar_4gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "word",
+  "language": "bar",
+  "unique_ngrams": 386315,
+  "total_ngrams": 8072889
+}