Upload all models and assets for bjn (20251201)

.gitattributes

@@ -33,3 +33,8 @@ 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/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: bjn
+language_name: BJN
+language_family: austronesian_malay
+tags:
+  - wikilangs
+  - nlp
+  - tokenizer
+  - embeddings
+  - n-gram
+  - markov
+  - wikipedia
+  - monolingual
+  - family-austronesian_malay
+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: 4.777
+  - name: best_isotropy
+    type: isotropy
+    value: 0.8757
+  - name: vocabulary_size
+    type: vocab
+    value: 43482
+generated: 2025-12-28
+---
+
+# BJN - Wikilangs Models
+## Comprehensive Research Report & Full Ablation Study
+
+This repository contains NLP models trained and evaluated by Wikilangs, specifically on **BJN** 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** | 3.698x | 3.62 | 0.3416% | 403,663 |
+| **16k** | 4.081x | 4.00 | 0.3770% | 365,824 |
+| **32k** | 4.450x | 4.36 | 0.4111% | 335,441 |
+| **64k** | 4.777x 🏆 | 4.68 | 0.4413% | 312,459 |
+
+### Tokenization Examples
+
+Below are sample sentences tokenized with each vocabulary size:
+
+**Sample 1:** `Makmur yaitu sabuting kampung di Kacamatan Angsana, Kabupatin Tanah Bumbu, Prupi...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁makmur ▁yaitu ▁sabuting ▁kampung ▁di ▁kacamatan ▁ angs ana , ... (+10 more)` | 20 |
+| 16k | `▁makmur ▁yaitu ▁sabuting ▁kampung ▁di ▁kacamatan ▁angsana , ▁kabupatin ▁tanah ... (+8 more)` | 18 |
+| 32k | `▁makmur ▁yaitu ▁sabuting ▁kampung ▁di ▁kacamatan ▁angsana , ▁kabupatin ▁tanah ... (+8 more)` | 18 |
+| 64k | `▁makmur ▁yaitu ▁sabuting ▁kampung ▁di ▁kacamatan ▁angsana , ▁kabupatin ▁tanah ... (+8 more)` | 18 |
+
+**Sample 2:** `Tanjung Mangkalihat adalah sabuah kampung di Kacamatan Sandaran, Kabupatin Kutai...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁tanjung ▁mangk alihat ▁adalah ▁sabuah ▁kampung ▁di ▁kacamatan ▁sand aran ... (+11 more)` | 21 |
+| 16k | `▁tanjung ▁mangk alihat ▁adalah ▁sabuah ▁kampung ▁di ▁kacamatan ▁sandaran , ... (+10 more)` | 20 |
+| 32k | `▁tanjung ▁mangk alihat ▁adalah ▁sabuah ▁kampung ▁di ▁kacamatan ▁sandaran , ... (+10 more)` | 20 |
+| 64k | `▁tanjung ▁mangk alihat ▁adalah ▁sabuah ▁kampung ▁di ▁kacamatan ▁sandaran , ... (+10 more)` | 20 |
+
+**Sample 3:** `Sumuragung adalah sabuah kampung di Kacamatan Sumberejo, Kabupatin Bojonegoro, P...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁sum ur agung ▁adalah ▁sabuah ▁kampung ▁di ▁kacamatan ▁sumberejo , ... (+9 more)` | 19 |
+| 16k | `▁sumur agung ▁adalah ▁sabuah ▁kampung ▁di ▁kacamatan ▁sumberejo , ▁kabupatin ... (+8 more)` | 18 |
+| 32k | `▁sumur agung ▁adalah ▁sabuah ▁kampung ▁di ▁kacamatan ▁sumberejo , ▁kabupatin ... (+8 more)` | 18 |
+| 64k | `▁sumur agung ▁adalah ▁sabuah ▁kampung ▁di ▁kacamatan ▁sumberejo , ▁kabupatin ... (+8 more)` | 18 |
+
+
+### Key Findings
+
+- **Best Compression:** 64k achieves 4.777x compression
+- **Lowest UNK Rate:** 8k with 0.3416% 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** | 8,214 🏆 | 13.00 | 31,697 | 22.3% | 43.3% |
+| **2-gram** | 216 🏆 | 7.75 | 3,606 | 75.0% | 99.2% |
+| **3-gram** | 7,363 | 12.85 | 33,190 | 25.9% | 45.0% |
+| **3-gram** | 1,713 | 10.74 | 25,998 | 32.5% | 77.1% |
+| **4-gram** | 9,532 | 13.22 | 46,645 | 25.7% | 43.0% |
+| **4-gram** | 9,186 | 13.17 | 125,660 | 16.5% | 47.6% |
+
+### Top 5 N-grams by Size
+
+**2-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `indunisia .` | 7,623 |
+| 2 | `, indunisia` | 7,382 |
+| 3 | `, prupinsi` | 7,240 |
+| 4 | `, kabupatin` | 6,196 |
+| 5 | `kampung di` | 5,960 |
+
+**3-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `, indunisia .` | 7,294 |
+| 2 | `, prupinsi kalimantan` | 5,775 |
+| 3 | `kampung di kacamatan` | 5,212 |
+| 4 | `sabuah kampung di` | 3,803 |
+| 5 | `adalah sabuah kampung` | 3,803 |
+
+**4-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `sabuah kampung di kacamatan` | 3,802 |
+| 2 | `adalah sabuah kampung di` | 3,801 |
+| 3 | `kalimantan selatan , indunisia` | 2,200 |
+| 4 | `selatan , indunisia .` | 2,195 |
+| 5 | `prupinsi kalimantan selatan ,` | 2,161 |
+
+
+### Key Findings
+
+- **Best Perplexity:** 2-gram with 216
+- **Entropy Trend:** Decreases with larger n-grams (more predictable)
+- **Coverage:** Top-1000 patterns cover ~48% 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** | 0.7609 | 1.695 | 5.41 | 108,020 | 23.9% |
+| **1** | 0.9104 | 1.880 | 5.41 | 2,572 | 9.0% |
+| **2** | 0.2888 | 1.222 | 1.67 | 583,707 | 71.1% |
+| **2** | 0.7263 | 1.654 | 4.39 | 13,908 | 27.4% |
+| **3** | 0.0836 | 1.060 | 1.14 | 970,699 | 91.6% |
+| **3** | 0.7940 | 1.734 | 3.86 | 61,013 | 20.6% |
+| **4** | 0.0265 🏆 | 1.019 | 1.04 | 1,101,630 | 97.4% |
+| **4** | 0.6245 🏆 | 1.542 | 2.66 | 235,356 | 37.6% |
+
+### Generated Text Samples
+
+Below are text samples generated from each Markov chain model:
+
+**Context Size 1:**
+
+1. `, umat buddha dingarani nangkaya : kampung di kacamatan pengaron sambung makmur adalah pangalola wan...`
+2. `. jabatan dalam bintang nang malili di kacamatan tulin onsoi ngaran rasmi ; 3 . elly`
+3. `di antara ras , wan bawah tanah siang sampai tahun 1978 wan knapstad , antara 28`
+
+**Context Size 2:**
+
+1. `, indunisia . géografi watas wilayah kacamatan telaga antang ngaran kacamatan di kabupatin bengkayan...`
+2. `, prupinsi kalimantan barat , indunisia . géografi watas wilayah kampung mawar sari yaitu sabuting k...`
+3. `, kabupatin bangkalan , prupinsi kalimantan selatan , indunisia . géografi watas wilayah kampung bin...`
+
+**Context Size 3:**
+
+1. `, prupinsi kalimantan selatan , indunisia . watas wilayah watas wilayah kacamatan sepauk : pambagian...`
+2. `kampung di kacamatan selakau , kabupatin sambas , prupinsi kalimantan selatan , indunisia . watas wi...`
+3. `adalah sabuah kampung di kacamatan sungai tabuk , kabupatin banjar , prupinsi kalimantan barat , ind...`
+
+**Context Size 4:**
+
+1. `sabuah kampung di kacamatan tabang , kabupatin kutai kartanegara , prupinsi kalimantan timur , indun...`
+2. `adalah sabuah kampung di kacamatan glagah , kabupatin lamongan , prupinsi jawa timur , indunisia . g...`
+3. `kalimantan selatan , indunisia . géografi watas wilayah watas wilayah kacamatan gunungpuyuh : pambag...`
+
+
+### Key Findings
+
+- **Best Predictability:** Context-4 with 97.4% predictability
+- **Branching Factor:** Decreases with context size (more deterministic)
+- **Memory Trade-off:** Larger contexts require more storage (235,356 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 | 43,482 |
+| Total Tokens | 1,036,384 |
+| Mean Frequency | 23.83 |
+| Median Frequency | 4 |
+| Frequency Std Dev | 273.79 |
+
+### Most Common Words
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | di | 27,734 |
+| 2 | nang | 27,434 |
+| 3 | wan | 17,268 |
+| 4 | adalah | 10,732 |
+| 5 | lawan | 9,615 |
+| 6 | indunisia | 9,432 |
+| 7 | kacamatan | 9,151 |
+| 8 | kalimantan | 8,405 |
+| 9 | kampung | 7,829 |
+| 10 | matan | 7,719 |
+
+### Least Common Words (from vocabulary)
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | bepacaran | 2 |
+| 2 | beregszásziová | 2 |
+| 3 | košice | 2 |
+| 4 | satian | 2 |
+| 5 | frisna | 2 |
+| 6 | ropang | 2 |
+| 7 | caknan | 2 |
+| 8 | muktamar | 2 |
+| 9 | sandon | 2 |
+| 10 | sékuéns | 2 |
+
+### Zipf's Law Analysis
+
+| Metric | Value |
+|--------|-------|
+| Zipf Coefficient | 1.0456 |
+| R² (Goodness of Fit) | 0.994175 |
+| Adherence Quality | **excellent** |
+
+### Coverage Analysis
+
+| Top N Words | Coverage |
+|-------------|----------|
+| Top 100 | 34.5% |
+| Top 1,000 | 61.2% |
+| Top 5,000 | 80.9% |
+| Top 10,000 | 88.2% |
+
+### Key Findings
+
+- **Zipf Compliance:** R²=0.9942 indicates excellent adherence to Zipf's law
+- **High Frequency Dominance:** Top 100 words cover 34.5% of corpus
+- **Long Tail:** 33,482 words needed for remaining 11.8% 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** | 21,102 | 32 | 3.645 | 0.820 | 0.8757 🏆 |
+| **mono_64d** | 21,102 | 64 | 4.192 | 0.774 | 0.8364 |
+| **mono_128d** | 21,102 | 128 | 4.539 | 0.732 | 0.5631 |
+| **embeddings_enhanced** | 0 | 0 | 0.000 | 0.000 | 0.0000 |
+
+### Key Findings
+
+- **Best Isotropy:** mono_32d with 0.8757 (more uniform distribution)
+- **Dimension Trade-off:** Higher dimensions capture more semantics but reduce isotropy
+- **Vocabulary Coverage:** All models cover 21,102 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 (216) |
+| Markov | **Context-4** | Highest predictability (97.4%) |
+| Embeddings | **100d** | Balanced semantic capture and isotropy |
+
+---
+## Appendix: Metrics Glossary & Interpretation Guide
+
+This section provides definitions, intuitions, and guidance for interpreting the metrics used throughout this report.
+
+### Tokenizer Metrics
+
+**Compression Ratio**
+> *Definition:* The ratio of characters to tokens (chars/token). Measures how efficiently the tokenizer represents text.
+>
+> *Intuition:* Higher compression means fewer tokens needed to represent the same text, reducing sequence lengths for downstream models. A 3x compression means ~3 characters per token on average.
+>
+> *What to seek:* Higher is generally better for efficiency, but extremely high compression may indicate overly aggressive merging that loses morphological information.
+
+**Average Token Length (Fertility)**
+> *Definition:* Mean number of characters per token produced by the tokenizer.
+>
+> *Intuition:* Reflects the granularity of tokenization. Longer tokens capture more context but may struggle with rare words; shorter tokens are more flexible but increase sequence length.
+>
+> *What to seek:* Balance between 2-5 characters for most languages. Arabic/morphologically-rich languages may benefit from slightly longer tokens.
+
+**Unknown Token Rate (OOV Rate)**
+> *Definition:* Percentage of tokens that map to the unknown/UNK token, indicating words the tokenizer cannot represent.
+>
+> *Intuition:* Lower OOV means better vocabulary coverage. High OOV indicates the tokenizer encounters many unseen character sequences.
+>
+> *What to seek:* Below 1% is excellent; below 5% is acceptable. BPE tokenizers typically achieve very low OOV due to subword fallback.
+
+### N-gram Model Metrics
+
+**Perplexity**
+> *Definition:* Measures how "surprised" the model is by test data. Mathematically: 2^(cross-entropy). Lower values indicate better prediction.
+>
+> *Intuition:* If perplexity is 100, the model is as uncertain as if choosing uniformly among 100 options at each step. A perplexity of 10 means effectively choosing among 10 equally likely options.
+>
+> *What to seek:* Lower is better. Perplexity decreases with larger n-grams (more context). Values vary widely by language and corpus size.
+
+**Entropy**
+> *Definition:* Average information content (in bits) needed to encode the next token given the context. Related to perplexity: perplexity = 2^entropy.
+>
+> *Intuition:* High entropy means high uncertainty/randomness; low entropy means predictable patterns. Natural language typically has entropy between 1-4 bits per character.
+>
+> *What to seek:* Lower entropy indicates more predictable text patterns. Entropy should decrease as n-gram size increases.
+
+**Coverage (Top-K)**
+> *Definition:* Percentage of corpus occurrences explained by the top K most frequent n-grams.
+>
+> *Intuition:* High coverage with few patterns indicates repetitive/formulaic text; low coverage suggests diverse vocabulary usage.
+>
+> *What to seek:* Depends on use case. For language modeling, moderate coverage (40-60% with top-1000) is typical for natural text.
+
+### Markov Chain Metrics
+
+**Average Entropy**
+> *Definition:* Mean entropy across all contexts, measuring average uncertainty in next-word prediction.
+>
+> *Intuition:* Lower entropy means the model is more confident about what comes next. Context-1 has high entropy (many possible next words); Context-4 has low entropy (few likely continuations).
+>
+> *What to seek:* Decreasing entropy with larger context sizes. Very low entropy (<0.1) indicates highly deterministic transitions.
+
+**Branching Factor**
+> *Definition:* Average number of unique next tokens observed for each context.
+>
+> *Intuition:* High branching = many possible continuations (flexible but uncertain); low branching = few options (predictable but potentially repetitive).
+>
+> *What to seek:* Branching factor should decrease with context size. Values near 1.0 indicate nearly deterministic chains.
+
+**Predictability**
+> *Definition:* Derived metric: (1 - normalized_entropy) × 100%. Indicates how deterministic the model's predictions are.
+>
+> *Intuition:* 100% predictability means the next word is always certain; 0% means completely random. Real text falls between these extremes.
+>
+> *What to seek:* Higher predictability for text generation quality, but too high (>98%) may produce repetitive output.
+
+### Vocabulary & Zipf's Law Metrics
+
+**Zipf's Coefficient**
+> *Definition:* The slope of the log-log plot of word frequency vs. rank. Zipf's law predicts this should be approximately -1.
+>
+> *Intuition:* A coefficient near -1 indicates the corpus follows natural language patterns where a few words are very common and most words are rare.
+>
+> *What to seek:* Values between -0.8 and -1.2 indicate healthy natural language distribution. Deviations may suggest domain-specific or artificial text.
+
+**R² (Coefficient of Determination)**
+> *Definition:* Measures how well the linear fit explains the frequency-rank relationship. Ranges from 0 to 1.
+>
+> *Intuition:* R² near 1.0 means the data closely follows Zipf's law; lower values indicate deviation from expected word frequency patterns.
+>
+> *What to seek:* R² > 0.95 is excellent; > 0.99 indicates near-perfect Zipf adherence typical of large natural corpora.
+
+**Vocabulary Coverage**
+> *Definition:* Cumulative percentage of corpus tokens accounted for by the top N words.
+>
+> *Intuition:* Shows how concentrated word usage is. If top-100 words cover 50% of text, the corpus relies heavily on common words.
+>
+> *What to seek:* Top-100 covering 30-50% is typical. Higher coverage indicates more repetitive text; lower suggests richer vocabulary.
+
+### Word Embedding Metrics
+
+**Isotropy**
+> *Definition:* Measures how uniformly distributed vectors are in the embedding space. Computed as the ratio of minimum to maximum singular values.
+>
+> *Intuition:* High isotropy (near 1.0) means vectors spread evenly in all directions; low isotropy means vectors cluster in certain directions, reducing expressiveness.
+>
+> *What to seek:* Higher isotropy generally indicates better-quality embeddings. Values > 0.1 are reasonable; > 0.3 is good. Lower-dimensional embeddings tend to have higher isotropy.
+
+**Average Norm**
+> *Definition:* Mean magnitude (L2 norm) of word vectors in the embedding space.
+>
+> *Intuition:* Indicates the typical "length" of vectors. Consistent norms suggest stable training; high variance may indicate some words are undertrained.
+>
+> *What to seek:* Relatively consistent norms across models. The absolute value matters less than consistency (low std deviation).
+
+**Cosine Similarity**
+> *Definition:* Measures angular similarity between vectors, ranging from -1 (opposite) to 1 (identical direction).
+>
+> *Intuition:* Words with similar meanings should have high cosine similarity. This is the standard metric for semantic relatedness in embeddings.
+>
+> *What to seek:* Semantically related words should score > 0.5; unrelated words should be near 0. Synonyms often score > 0.7.
+
+**t-SNE Visualization**
+> *Definition:* t-Distributed Stochastic Neighbor Embedding - a dimensionality reduction technique that preserves local structure for visualization.
+>
+> *Intuition:* Clusters in t-SNE plots indicate groups of semantically related words. Spread indicates vocabulary diversity; tight clusters suggest semantic coherence.
+>
+> *What to seek:* Meaningful clusters (e.g., numbers together, verbs together). Avoid over-interpreting distances - t-SNE preserves local, not global, structure.
+
+### General Interpretation Guidelines
+
+1. **Compare within model families:** Metrics are most meaningful when comparing models of the same type (e.g., 8k vs 64k tokenizer).
+2. **Consider trade-offs:** Better performance on one metric often comes at the cost of another (e.g., compression vs. OOV rate).
+3. **Context matters:** Optimal values depend on downstream tasks. Text generation may prioritize different metrics than classification.
+4. **Corpus influence:** All metrics are influenced by corpus characteristics. Wikipedia text differs from social media or literature.
+5. **Language-specific patterns:** Morphologically rich languages (like Arabic) may show different optimal ranges than analytic languages.
+
+
+### Visualizations Index
+
+| Visualization | Description |
+|---------------|-------------|
+| Tokenizer Compression | Compression ratios by vocabulary size |
+| Tokenizer Fertility | Average token length by vocabulary |
+| Tokenizer OOV | Unknown token rates |
+| Tokenizer Total Tokens | Total tokens by vocabulary |
+| N-gram Perplexity | Perplexity by n-gram size |
+| N-gram Entropy | Entropy by n-gram size |
+| N-gram Coverage | Top pattern coverage |
+| N-gram Unique | Unique n-gram counts |
+| Markov Entropy | Entropy by context size |
+| Markov Branching | Branching factor by context |
+| Markov Contexts | Unique context counts |
+| Zipf's Law | Frequency-rank distribution with fit |
+| Vocab Frequency | Word frequency distribution |
+| Top 20 Words | Most frequent words |
+| Vocab Coverage | Cumulative coverage curve |
+| Embedding Isotropy | Vector space uniformity |
+| Embedding Norms | Vector magnitude distribution |
+| Embedding Similarity | Word similarity heatmap |
+| Nearest Neighbors | Similar words for key terms |
+| t-SNE Words | 2D word embedding visualization |
+| t-SNE Sentences | 2D sentence embedding visualization |
+| Position Encoding | Encoding method comparison |
+| Model Sizes | Storage requirements |
+| Performance Dashboard | Comprehensive performance overview |
+
+---
+## About This Project
+
+### Data Source
+
+Models trained on [wikipedia-monthly](https://huggingface.co/datasets/omarkamali/wikipedia-monthly) - a monthly snapshot of Wikipedia articles across 300+ languages.
+
+### Project
+
+A project by **[Wikilangs](https://wikilangs.org)** - Open-source NLP models for every Wikipedia language.
+
+### Maintainer
+
+[Omar Kamali](https://omarkamali.com) - [Omneity Labs](https://omneitylabs.com)
+
+### Citation
+
+If you use these models in your research, please cite:
+
+```bibtex
+@misc{wikilangs2025,
+  author = {Kamali, Omar},
+  title = {Wikilangs: Open NLP Models for Wikipedia Languages},
+  year = {2025},
+  publisher = {HuggingFace},
+  url = {https://huggingface.co/wikilangs}
+  institution = {Omneity Labs}
+}
+```
+
+### License
+
+MIT License - Free for academic and commercial use.
+
+### Links
+
+- 🌐 Website: [wikilangs.org](https://wikilangs.org)
+- 🤗 Models: [huggingface.co/wikilangs](https://huggingface.co/wikilangs)
+- 📊 Data: [wikipedia-monthly](https://huggingface.co/datasets/omarkamali/wikipedia-monthly)
+- 👤 Author: [Omar Kamali](https://huggingface.co/omarkamali)
+---
+*Generated by Wikilangs Models Pipeline*
+
+*Report Date: 2025-12-28 05:18:09*

models/embeddings/monolingual/bjn_128d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e65409ff41e4182c5ab1696db0a4dd291168ff75e7a3d84800d52d65677827b8
+size 1045964232

models/embeddings/monolingual/bjn_128d.meta.json

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

models/embeddings/monolingual/bjn_128d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "bjn",
+  "dimension": 128,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 128,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 21102
+}

models/embeddings/monolingual/bjn_32d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c4ba0d5d7cc97005bb29d178bcc33da604ec2ff54b23f1f552000768cabdbf8
+size 261757896

models/embeddings/monolingual/bjn_32d.meta.json

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

models/embeddings/monolingual/bjn_32d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "bjn",
+  "dimension": 32,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 32,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 21102
+}

models/embeddings/monolingual/bjn_64d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:200944aec2429d1617154426e2e8a0393267c3a2b5366e032a99e04000daf227
+size 523160008

models/embeddings/monolingual/bjn_64d.meta.json

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

models/embeddings/monolingual/bjn_64d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "bjn",
+  "dimension": 64,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 64,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 21102
+}

models/subword_markov/bjn_markov_ctx1_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e9986331e278cde13c4581c7a4b37a46d7a9c5a6972ad0119dec3127a4a9468
+size 105165

models/subword_markov/bjn_markov_ctx1_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "subword",
+  "language": "bjn",
+  "unique_contexts": 2572,
+  "total_transitions": 7876088
+}

models/subword_markov/bjn_markov_ctx2_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62714b996e21cfc6c73dffa6ddadff23b1cb4ebbb512ea7979b9607d639f9e52
+size 495942

models/subword_markov/bjn_markov_ctx2_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "subword",
+  "language": "bjn",
+  "unique_contexts": 13908,
+  "total_transitions": 7864568
+}

models/subword_markov/bjn_markov_ctx3_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21487466bfac2e926216be93e702aaf0111e80998a59a47c1d5f6d60bb3e55ae
+size 1728073

models/subword_markov/bjn_markov_ctx3_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "subword",
+  "language": "bjn",
+  "unique_contexts": 61013,
+  "total_transitions": 7853048
+}

models/subword_markov/bjn_markov_ctx4_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40577f595c4773cf05fd06e7676f078a7def769251fd22df537a94333ef0d960
+size 4742991

models/subword_markov/bjn_markov_ctx4_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "subword",
+  "language": "bjn",
+  "unique_contexts": 235356,
+  "total_transitions": 7841528
+}

models/subword_ngram/bjn_2gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:947ebba2eea6709955c942ae6a45f98916212ba270e6d663b77823d2c8442b1d
+size 46094

models/subword_ngram/bjn_2gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "subword",
+  "language": "bjn",
+  "unique_ngrams": 3606,
+  "total_ngrams": 7876088
+}

models/subword_ngram/bjn_3gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d3d72e0319456f87568ec49b946d626c3ff9f845502be1f8d3d69d8c4dac8db
+size 311919

models/subword_ngram/bjn_3gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "subword",
+  "language": "bjn",
+  "unique_ngrams": 25998,
+  "total_ngrams": 7864568
+}

models/subword_ngram/bjn_4gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78caf0f27c30dc7aaf8637610d8d16e7e8bf3bb39788eb3d95c3ad4faae93613
+size 1385663

models/subword_ngram/bjn_4gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "subword",
+  "language": "bjn",
+  "unique_ngrams": 125660,
+  "total_ngrams": 7853048
+}

models/tokenizer/bjn_tokenizer_16k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0789678d80af8c0732774eb4b6bb1e41146fd09072f82089a4bb2aa3a90517c8
+size 505456

models/tokenizer/bjn_tokenizer_32k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4919663d1e73201b9de368a024ef7e7fb2b833b887eb7c39a545e5b90b0f1b6d
+size 780920

models/tokenizer/bjn_tokenizer_64k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84987346017d033ed1a7339bbb8ad76c2fc1c9f7824e8d1912cfa72d31384a80
+size 1331984

models/tokenizer/bjn_tokenizer_8k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f08a90f9e0b51994122ed4bb0a18d36e598dfea67df86f18ae6cc765079db22e
+size 371550

models/vocabulary/bjn_vocabulary.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:176907dc0990fd1f1a9773b5c9d088fb3e3e76c41eb4e4335673c1e3550bc153
+size 741453

models/vocabulary/bjn_vocabulary_metadata.json

@@ -0,0 +1,16 @@
+{
+  "language": "bjn",
+  "vocabulary_size": 43482,
+  "statistics": {
+    "type_token_ratio": 0.09796208341358296,
+    "coverage": {
+      "top_100": 0.32512364499260493,
+      "top_1000": 0.576528164893907,
+      "top_5000": 0.761968399210707,
+      "top_10000": 0.8305046096597537
+    },
+    "hapax_count": 64348,
+    "hapax_ratio": 0.5967541500510062,
+    "total_documents": 11520
+  }
+}

models/word_markov/bjn_markov_ctx1_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e1e490ec45c03263bbee5b0cb5d14a9309b3856b41d89553247da0fad241778
+size 4355355

models/word_markov/bjn_markov_ctx1_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "word",
+  "language": "bjn",
+  "unique_contexts": 108020,
+  "total_transitions": 1341605
+}

models/word_markov/bjn_markov_ctx2_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:31270f3260dd0112d37d7fc2f7d4231f5a1ee74904fee4b7d4ed309f3add4d7f
+size 11014789

models/word_markov/bjn_markov_ctx2_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "word",
+  "language": "bjn",
+  "unique_contexts": 583707,
+  "total_transitions": 1330085
+}

models/word_markov/bjn_markov_ctx3_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6dbbda53445631ccecd1be92001ce4f423233d3fe041ffb12afe0a356c91a9ef
+size 16006739

models/word_markov/bjn_markov_ctx3_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "word",
+  "language": "bjn",
+  "unique_contexts": 970699,
+  "total_transitions": 1318565
+}

models/word_markov/bjn_markov_ctx4_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1bb5f64ee3e9ecd1c820d8a4acba751e1066a6106871f181251a4debf90ddba
+size 18389438

models/word_markov/bjn_markov_ctx4_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "word",
+  "language": "bjn",
+  "unique_contexts": 1101630,
+  "total_transitions": 1307048
+}

models/word_ngram/bjn_2gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd6e87d662bfb47bba380d18db5125e72bf837e8638bf8005e5b38bea2e82aab
+size 438250

models/word_ngram/bjn_2gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "word",
+  "language": "bjn",
+  "unique_ngrams": 31697,
+  "total_ngrams": 1341605
+}

models/word_ngram/bjn_3gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef08ff903e4ce470871df54fd8f6d35ed89c905fe3c1fcfd357809fa41aa7266
+size 523283

models/word_ngram/bjn_3gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "word",
+  "language": "bjn",
+  "unique_ngrams": 33190,
+  "total_ngrams": 1330085
+}

models/word_ngram/bjn_4gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f632f15f1987aec2e7d1af2a418ac4cf9f9f426b023424d33b864115d7ae130
+size 732377

models/word_ngram/bjn_4gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "word",
+  "language": "bjn",
+  "unique_ngrams": 46645,
+  "total_ngrams": 1318565
+}