Upload all models and assets for btm (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 +1,554 @@
+---
+language: btm
+language_name: BTM
+language_family: austronesian_batak
+tags:
+  - wikilangs
+  - nlp
+  - tokenizer
+  - embeddings
+  - n-gram
+  - markov
+  - wikipedia
+  - monolingual
+  - family-austronesian_batak
+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: 4.785
+  - name: best_isotropy
+    type: isotropy
+    value: 0.4253
+  - name: vocabulary_size
+    type: vocab
+    value: 12175
+generated: 2025-12-28
+---
+
+# BTM - Wikilangs Models
+## Comprehensive Research Report & Full Ablation Study
+
+This repository contains NLP models trained and evaluated by Wikilangs, specifically on **BTM** 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** | 3.846x | 3.82 | 0.0837% | 246,208 |
+| **16k** | 4.225x | 4.20 | 0.0919% | 224,097 |
+| **32k** | 4.544x | 4.52 | 0.0989% | 208,361 |
+| **64k** | 4.785x 🏆 | 4.76 | 0.1041% | 197,875 |
+
+### Tokenization Examples
+
+Below are sample sentences tokenized with each vocabulary size:
+
+**Sample 1:** `23 Januari ima ari pa-23 i kalender Gregorian dohot 361 ari (sanga 362 ari i tao...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁ 2 3 ▁januari ▁ima ▁ari ▁pa - 2 3 ... (+24 more)` | 34 |
+| 16k | `▁ 2 3 ▁januari ▁ima ▁ari ▁pa - 2 3 ... (+24 more)` | 34 |
+| 32k | `▁ 2 3 ▁januari ▁ima ▁ari ▁pa - 2 3 ... (+24 more)` | 34 |
+| 64k | `▁ 2 3 ▁januari ▁ima ▁ari ▁pa - 2 3 ... (+24 more)` | 34 |
+
+**Sample 2:** `Gunung Tua Ms ima salah sada huta na adong i kecamatan Kotanopan, kabupaten Mand...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁gunung ▁tua ▁ms ▁ima ▁salah ▁sada ▁huta ▁na ▁adong ▁i ... (+13 more)` | 23 |
+| 16k | `▁gunung ▁tua ▁ms ▁ima ▁salah ▁sada ▁huta ▁na ▁adong ▁i ... (+13 more)` | 23 |
+| 32k | `▁gunung ▁tua ▁ms ▁ima ▁salah ▁sada ▁huta ▁na ▁adong ▁i ... (+13 more)` | 23 |
+| 64k | `▁gunung ▁tua ▁ms ▁ima ▁salah ▁sada ▁huta ▁na ▁adong ▁i ... (+13 more)` | 23 |
+
+**Sample 3:** `Mangalap boru ima sada karejo na ibaen i Mandailing Natal, karejo on pema jadi s...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁mangalap ▁boru ▁ima ▁sada ▁karejo ▁na ▁ibaen ▁i ▁mandailing ▁natal ... (+13 more)` | 23 |
+| 16k | `▁mangalap ▁boru ▁ima ▁sada ▁karejo ▁na ▁ibaen ▁i ▁mandailing ▁natal ... (+13 more)` | 23 |
+| 32k | `▁mangalap ▁boru ▁ima ▁sada ▁karejo ▁na ▁ibaen ▁i ▁mandailing ▁natal ... (+12 more)` | 22 |
+| 64k | `▁mangalap ▁boru ▁ima ▁sada ▁karejo ▁na ▁ibaen ▁i ▁mandailing ▁natal ... (+12 more)` | 22 |
+
+
+### Key Findings
+
+- **Best Compression:** 64k achieves 4.785x compression
+- **Lowest UNK Rate:** 8k with 0.0837% 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** | 2,793 🏆 | 11.45 | 5,965 | 25.6% | 57.6% |
+| **2-gram** | 237 🏆 | 7.89 | 1,918 | 71.1% | 99.2% |
+| **3-gram** | 2,655 | 11.37 | 5,700 | 27.3% | 55.1% |
+| **3-gram** | 1,865 | 10.87 | 12,988 | 30.0% | 75.1% |
+| **4-gram** | 3,977 | 11.96 | 8,130 | 23.7% | 46.0% |
+| **4-gram** | 8,932 | 13.12 | 49,385 | 14.7% | 44.5% |
+
+### Top 5 N-grams by Size
+
+**2-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `kategori :` | 781 |
+| 2 | `/ /` | 707 |
+| 3 | `: /` | 703 |
+| 4 | `ima sada` | 627 |
+| 5 | `, dot` | 599 |
+
+**3-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `: / /` | 702 |
+| 2 | `https : /` | 461 |
+| 3 | `. https :` | 384 |
+| 4 | `sumberna kategori :` | 348 |
+| 5 | `ari pa -` | 308 |
+
+**4-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `https : / /` | 461 |
+| 2 | `. https : /` | 384 |
+| 3 | `: / / www` | 221 |
+| 4 | `/ / www .` | 221 |
+| 5 | `ima ari pa -` | 183 |
+
+
+### Key Findings
+
+- **Best Perplexity:** 2-gram with 237
+- **Entropy Trend:** Decreases with larger n-grams (more predictable)
+- **Coverage:** Top-1000 patterns cover ~45% 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** | 0.7327 | 1.662 | 4.59 | 29,952 | 26.7% |
+| **1** | 1.0526 | 2.074 | 7.13 | 776 | 0.0% |
+| **2** | 0.2488 | 1.188 | 1.51 | 137,080 | 75.1% |
+| **2** | 0.9494 | 1.931 | 5.09 | 5,530 | 5.1% |
+| **3** | 0.0695 | 1.049 | 1.11 | 206,158 | 93.1% |
+| **3** | 0.8132 | 1.757 | 3.45 | 28,161 | 18.7% |
+| **4** | 0.0218 🏆 | 1.015 | 1.03 | 227,434 | 97.8% |
+| **4** | 0.5358 🏆 | 1.450 | 2.24 | 97,224 | 46.4% |
+
+### Generated Text Samples
+
+Below are text samples generated from each Markov chain model:
+
+**Context Size 1:**
+
+1. `. hami sanga kamp parkarejo jerman . biasona aong ibaen anso mambuat sada tim dohot mamake`
+2. `, uhum kodrat - ira satonga tarbuka i provinsi opolskie provinsi i diafragma mambatasi bagian sian`
+3. `i kalender gregorian dohot tema na ilukis nia hagoluan nalai ima . sacara enskapitulasi tingon satio...`
+
+**Context Size 2:**
+
+1. `kategori : ilmuwan yahudi kategori : politik kategori : saro inggeris : engineering ) ima sada penya...`
+2. `/ / download / 18320 / 13475 baru nung i presepsi na padua bolasna na dua kota`
+3. `: / / www . rusdionoconsulting . com / tag / pemandian - air - terjun -`
+
+**Context Size 3:**
+
+1. `: / / onepiece . fandom . com / id / tourism / gunung - sorik - marapi`
+2. `https : / / startfmmadina . com / 40 - daftar - kumpulan - terjemahan - bahasa -`
+3. `. https : / / madina . go . id / sejarah - bulu - tangkis / tarsarupo`
+
+**Context Size 4:**
+
+1. `https : / / www . pmimedan . or . id / sahabat - nabi / songkoni juo ma`
+2. `. https : / / translate . google . com / translate ? u = https : / /`
+3. `: / / www . harianhaluan . com / news / world - middle - east - 14703476 "`
+
+
+### Key Findings
+
+- **Best Predictability:** Context-4 with 97.8% predictability
+- **Branching Factor:** Decreases with context size (more deterministic)
+- **Memory Trade-off:** Larger contexts require more storage (97,224 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 | 12,175 |
+| Total Tokens | 189,541 |
+| Mean Frequency | 15.57 |
+| Median Frequency | 3 |
+| Frequency Std Dev | 125.93 |
+
+### Most Common Words
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | i | 7,254 |
+| 2 | na | 7,126 |
+| 3 | ima | 4,033 |
+| 4 | on | 3,983 |
+| 5 | dohot | 2,990 |
+| 6 | ni | 2,684 |
+| 7 | dot | 2,484 |
+| 8 | sada | 1,835 |
+| 9 | tu | 1,710 |
+| 10 | ma | 1,482 |
+
+### Least Common Words (from vocabulary)
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | lil | 2 |
+| 2 | imah | 2 |
+| 3 | nasida | 2 |
+| 4 | sunusi | 2 |
+| 5 | nunga | 2 |
+| 6 | majmu | 2 |
+| 7 | fatawa | 2 |
+| 8 | fiqhi | 2 |
+| 9 | panjalakian | 2 |
+| 10 | martoba | 2 |
+
+### Zipf's Law Analysis
+
+| Metric | Value |
+|--------|-------|
+| Zipf Coefficient | 1.0695 |
+| R² (Goodness of Fit) | 0.987847 |
+| Adherence Quality | **excellent** |
+
+### Coverage Analysis
+
+| Top N Words | Coverage |
+|-------------|----------|
+| Top 100 | 40.1% |
+| Top 1,000 | 69.3% |
+| Top 5,000 | 90.3% |
+| Top 10,000 | 97.7% |
+
+### Key Findings
+
+- **Zipf Compliance:** R²=0.9878 indicates excellent adherence to Zipf's law
+- **High Frequency Dominance:** Top 100 words cover 40.1% of corpus
+- **Long Tail:** 2,175 words needed for remaining 2.3% 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** | 4,643 | 32 | 2.986 | 0.583 | 0.4253 🏆 |
+| **mono_64d** | 4,643 | 64 | 3.026 | 0.583 | 0.1288 |
+| **mono_128d** | 4,643 | 128 | 3.050 | 0.582 | 0.0237 |
+| **embeddings_enhanced** | 0 | 0 | 0.000 | 0.000 | 0.0000 |
+
+### Key Findings
+
+- **Best Isotropy:** mono_32d with 0.4253 (more uniform distribution)
+- **Dimension Trade-off:** Higher dimensions capture more semantics but reduce isotropy
+- **Vocabulary Coverage:** All models cover 4,643 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 (4.78x) with low UNK rate |
+| N-gram | **5-gram** | Lowest perplexity (237) |
+| Markov | **Context-4** | Highest predictability (97.8%) |
+| 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 09:18:23*

models/embeddings/monolingual/btm_128d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de188f03b192c2ca078f88738ce73449fcbf4320cd395bd332ecef48590d80a8
+size 1028830510

models/embeddings/monolingual/btm_128d.meta.json

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

models/embeddings/monolingual/btm_128d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "btm",
+  "dimension": 128,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 128,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 4643
+}

models/embeddings/monolingual/btm_32d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13216c4fed3a766eb9ae09c6f6a9636ab84897b0ed1d8023ee3f84bfc09c60fe
+size 257264686

models/embeddings/monolingual/btm_32d.meta.json

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

models/embeddings/monolingual/btm_32d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "btm",
+  "dimension": 32,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 32,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 4643
+}

models/embeddings/monolingual/btm_64d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51848e61c1fca01ab96c42576eb6e62eefed815eb2acc39e03fea244cb0d7a55
+size 514453294

models/embeddings/monolingual/btm_64d.meta.json

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

models/embeddings/monolingual/btm_64d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "btm",
+  "dimension": 64,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 64,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 4643
+}

models/subword_markov/btm_markov_ctx1_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9fb8732e5d0b088e8cb27dcdd5b2f86c532d02ef8d60b6a276973da326e2489
+size 44542

models/subword_markov/btm_markov_ctx1_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "subword",
+  "language": "btm",
+  "unique_contexts": 776,
+  "total_transitions": 1324161
+}

models/subword_markov/btm_markov_ctx2_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b32d6be400f0fe5ba28845e329de73b2eca4c57b1ecb34e7b271b0a81000cd5
+size 221609

models/subword_markov/btm_markov_ctx2_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "subword",
+  "language": "btm",
+  "unique_contexts": 5530,
+  "total_transitions": 1322799
+}

models/subword_markov/btm_markov_ctx3_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:baf72e1c86d729f7636eb3dd63f352dcd015fc51d995b57fe2a4a2ae73643c47
+size 752323

models/subword_markov/btm_markov_ctx3_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "subword",
+  "language": "btm",
+  "unique_contexts": 28161,
+  "total_transitions": 1321437
+}

models/subword_markov/btm_markov_ctx4_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f0a69fd846d53a6e55464bb04597ad44e2cd78771e025aad882c45ac3903096d
+size 1709839

models/subword_markov/btm_markov_ctx4_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "subword",
+  "language": "btm",
+  "unique_contexts": 97224,
+  "total_transitions": 1320075
+}

models/subword_ngram/btm_2gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86dfdf6bd3e46f0af84e22f9dfc755a0a319afbfb427a853d7447f54aa36b813
+size 25120

models/subword_ngram/btm_2gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "subword",
+  "language": "btm",
+  "unique_ngrams": 1918,
+  "total_ngrams": 1324161
+}

models/subword_ngram/btm_3gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ac8d488fb34cbcad5cee074bab2aadc311377c61f106fdc7d15893ac59de7ac
+size 146419

models/subword_ngram/btm_3gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "subword",
+  "language": "btm",
+  "unique_ngrams": 12988,
+  "total_ngrams": 1322799
+}

models/subword_ngram/btm_4gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a86ef1d182c41d4ec45dc51eae1271722efec2f5a35604893699f66328234dfd
+size 594616

models/subword_ngram/btm_4gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "subword",
+  "language": "btm",
+  "unique_ngrams": 49385,
+  "total_ngrams": 1321437
+}

models/tokenizer/btm_tokenizer_16k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddd3c2f8af55f1ab6fddfd062122811cc700509a098060ad39d467f214f86732
+size 505236

models/tokenizer/btm_tokenizer_32k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2a3d5dac68f0879797fe727d4d2b0cbc7840fdedeee4e86c6bb03f5adc7f0da
+size 788854

models/tokenizer/btm_tokenizer_64k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07075eace5ffa2fc3001cd72bdd62fa08486ac53b2331193d7cd888c21fbbbf1
+size 1322499

models/tokenizer/btm_tokenizer_8k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:842451ea10488cf3aa06059e02fdc805910e7ac01548035e26320a5ec1c675ff
+size 372071

models/vocabulary/btm_vocabulary.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2e1669d9cb891748049f7b0f79cddae67d9c1d88f9aacfd042c76bc0023e258
+size 194790

models/vocabulary/btm_vocabulary_metadata.json

@@ -0,0 +1,16 @@
+{
+  "language": "btm",
+  "vocabulary_size": 12175,
+  "statistics": {
+    "type_token_ratio": 0.1439864864864865,
+    "coverage": {
+      "top_100": 0.36650096525096526,
+      "top_1000": 0.6339671814671815,
+      "top_5000": 0.825945945945946,
+      "top_10000": 0.8937789575289575
+    },
+    "hapax_count": 17659,
+    "hapax_ratio": 0.5919085607025542,
+    "total_documents": 1362
+  }
+}

models/word_markov/btm_markov_ctx1_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c217e1a408565da0c01ce1be6c3a82b8d05ba2eb243a350de04d9a1d0c26215
+size 1027194

models/word_markov/btm_markov_ctx1_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "word",
+  "language": "btm",
+  "unique_contexts": 29952,
+  "total_transitions": 252626
+}

models/word_markov/btm_markov_ctx2_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:669e28490302d21e70f64127f665d428c1484dca09fc3c62a516bd2cce61c8e0
+size 2302244

models/word_markov/btm_markov_ctx2_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "word",
+  "language": "btm",
+  "unique_contexts": 137080,
+  "total_transitions": 251264
+}

models/word_markov/btm_markov_ctx3_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ca7a37098cceada0a3b495bf357855b55ae3c07593991310fc65a3987345d93
+size 3187577

models/word_markov/btm_markov_ctx3_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "word",
+  "language": "btm",
+  "unique_contexts": 206158,
+  "total_transitions": 249909
+}

models/word_markov/btm_markov_ctx4_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3644395ecf865c63d7aad1a53f542d746ec0774d12d6dae591b4e5db701c596a
+size 3651246

models/word_markov/btm_markov_ctx4_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "word",
+  "language": "btm",
+  "unique_contexts": 227434,
+  "total_transitions": 248555
+}

models/word_ngram/btm_2gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45947c8329577e1dfdfbb73f89b03e03aff959f4f2a1ebbc65fabc83ca7fc373
+size 76262

models/word_ngram/btm_2gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "word",
+  "language": "btm",
+  "unique_ngrams": 5965,
+  "total_ngrams": 252626
+}

models/word_ngram/btm_3gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71ff57ab29c1ba5d6f5a7953a92d53d02c269b0162697be67aee03a490046ccb
+size 84350

models/word_ngram/btm_3gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "word",
+  "language": "btm",
+  "unique_ngrams": 5700,
+  "total_ngrams": 251264
+}

models/word_ngram/btm_4gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80d28cd8c28607e55ee503e11b2b708cd6502cdec2a7c34455ba7d5777b38646
+size 129368

models/word_ngram/btm_4gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "word",
+  "language": "btm",
+  "unique_ngrams": 8130,
+  "total_ngrams": 249909
+}