Upload all models and assets for br (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 +1,558 @@
+---
+language: br
+language_name: BR
+language_family: celtic_brythonic
+tags:
+  - wikilangs
+  - nlp
+  - tokenizer
+  - embeddings
+  - n-gram
+  - markov
+  - wikipedia
+  - monolingual
+  - family-celtic_brythonic
+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: 3.617
+  - name: best_isotropy
+    type: isotropy
+    value: 0.8322
+  - name: vocabulary_size
+    type: vocab
+    value: 263085
+generated: 2025-12-28
+---
+
+# BR - Wikilangs Models
+## Comprehensive Research Report & Full Ablation Study
+
+This repository contains NLP models trained and evaluated by Wikilangs, specifically on **BR** 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** | 3.122x | 3.07 | 0.4020% | 881,322 |
+| **16k** | 3.329x | 3.28 | 0.4286% | 826,606 |
+| **32k** | 3.492x | 3.44 | 0.4496% | 788,002 |
+| **64k** | 3.617x 🏆 | 3.56 | 0.4657% | 760,713 |
+
+### Tokenization Examples
+
+Below are sample sentences tokenized with each vocabulary size:
+
+**Sample 1:** `Fuentes de Jiloca zo ur gumun eus Spagn e Proviñs Zaragoza, en Aragon.`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁fu ent es ▁de ▁j il oc a ▁zo ▁ur ... (+12 more)` | 22 |
+| 16k | `▁fu entes ▁de ▁j il oca ▁zo ▁ur ▁gumun ▁eus ... (+8 more)` | 18 |
+| 32k | `▁fuentes ▁de ▁jil oca ▁zo ▁ur ▁gumun ▁eus ▁spagn ▁e ... (+6 more)` | 16 |
+| 64k | `▁fuentes ▁de ▁jil oca ▁zo ▁ur ▁gumun ▁eus ▁spagn ▁e ... (+6 more)` | 16 |
+
+**Sample 2:** `Barromán zo ur gumun eus Spagn, e proviñs Ávila, en Kastilha ha León. 
+
+Rummad:K...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁bar rom án ▁zo ▁ur ▁gumun ▁eus ▁spagn , ▁e ... (+14 more)` | 24 |
+| 16k | `▁bar rom án ▁zo ▁ur ▁gumun ▁eus ▁spagn , ▁e ... (+14 more)` | 24 |
+| 32k | `▁bar rom án ▁zo ▁ur ▁gumun ▁eus ▁spagn , ▁e ... (+14 more)` | 24 |
+| 64k | `▁bar rom án ▁zo ▁ur ▁gumun ▁eus ▁spagn , ▁e ... (+14 more)` | 24 |
+
+**Sample 3:** `Tolbaños zo ur gumun eus Spagn, e proviñs Ávila, en Kastilha ha León. 
+
+Rummad:K...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁tol b añ os ▁zo ▁ur ▁gumun ▁eus ▁spagn , ... (+15 more)` | 25 |
+| 16k | `▁tol b añ os ▁zo ▁ur ▁gumun ▁eus ▁spagn , ... (+15 more)` | 25 |
+| 32k | `▁tol b años ▁zo ▁ur ▁gumun ▁eus ▁spagn , ▁e ... (+14 more)` | 24 |
+| 64k | `▁tol b años ▁zo ▁ur ▁gumun ▁eus ▁spagn , ▁e ... (+14 more)` | 24 |
+
+
+### Key Findings
+
+- **Best Compression:** 64k achieves 3.617x compression
+- **Lowest UNK Rate:** 8k with 0.4020% 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** | 37,795 🏆 | 15.21 | 396,301 | 15.5% | 32.7% |
+| **2-gram** | 361 🏆 | 8.49 | 13,774 | 60.5% | 98.2% |
+| **3-gram** | 147,885 | 17.17 | 858,386 | 7.1% | 19.7% |
+| **3-gram** | 3,461 | 11.76 | 106,981 | 21.9% | 63.7% |
+| **4-gram** | 382,971 | 18.55 | 1,581,104 | 3.6% | 13.2% |
+| **4-gram** | 22,182 | 14.44 | 580,225 | 10.3% | 33.1% |
+
+### Top 5 N-grams by Size
+
+**2-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `c '` | 168,356 |
+| 2 | `rummad :` | 166,928 |
+| 3 | `d '` | 113,851 |
+| 4 | `' h` | 99,333 |
+| 5 | `, e` | 74,128 |
+
+**3-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `ar c '` | 51,117 |
+| 2 | `d ' ar` | 44,541 |
+| 3 | `d ' an` | 33,579 |
+| 4 | `. rummad :` | 24,562 |
+| 5 | `c ' hall` | 23,162 |
+
+**4-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `bro - c '` | 16,536 |
+| 2 | `- c ' hall` | 15,614 |
+| 3 | `zo ur gumun eus` | 8,365 |
+| 4 | `ha daveennoù rummad :` | 7,182 |
+| 5 | `notennoù ha daveennoù rummad` | 7,173 |
+
+
+### Key Findings
+
+- **Best Perplexity:** 2-gram with 361
+- **Entropy Trend:** Decreases with larger n-grams (more predictable)
+- **Coverage:** Top-1000 patterns cover ~33% 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** | 0.7133 | 1.640 | 6.37 | 642,948 | 28.7% |
+| **1** | 1.1559 | 2.228 | 7.70 | 6,849 | 0.0% |
+| **2** | 0.3934 | 1.313 | 2.35 | 4,087,653 | 60.7% |
+| **2** | 0.7490 | 1.681 | 4.86 | 52,722 | 25.1% |
+| **3** | 0.1872 | 1.139 | 1.44 | 9,600,283 | 81.3% |
+| **3** | 0.7699 | 1.705 | 4.16 | 256,241 | 23.0% |
+| **4** | 0.0950 🏆 | 1.068 | 1.19 | 13,824,129 | 90.5% |
+| **4** | 0.6874 🏆 | 1.610 | 3.24 | 1,065,026 | 31.3% |
+
+### Generated Text Samples
+
+Below are text samples generated from each Markov chain model:
+
+**Context Size 1:**
+
+1. `, job de reims . emañ o embann levrioù gant ur yezh romanek . ramsay ha`
+2. `- kreiz kalotenn skorn . sonet gantañ cheñch anv ar - se ne em gavas youenn`
+3. `. notennoù rummad : nobiliaire et tableaux d ' orgeraie ; barnet rust e - labour`
+
+**Context Size 2:**
+
+1. `c ' hommonwealth e beredoù ar brezel en ur bolz - enor e 1785 . darn all`
+2. `rummad : geomorfologiezh rummad : pladennoù brezhonek rummad : ganedigezhioù 1958 rummad : ganedigez...`
+3. `d ' ar gonfusianegezh lakaet e voe seitek nijour , ne c ' hoar vras paula (`
+
+**Context Size 3:**
+
+1. `ar c ' hentañ derez ) ofis publik ar brezhoneg . in : studia celtica18 / 19 :`
+2. `d ' ar 26 a viz eost . an anv implij al lerc ' h pa ' z`
+3. `d ' an arabegerion eta , evit skrivañ ar sañskriteg e vez implijet ar sistem - mañ gant`
+
+**Context Size 4:**
+
+1. `bro - c ' hall ) , e - lec ' h zo anvet pentre e kembre . hen`
+2. `- c ' hall ) d ' an 3 a viz genver 1871 . krouet e oa bet gant`
+3. `zo ur gumun eus italia , e proviñs cremona , e rannvro lombardia . rummad : kumunioù lombardia rumma...`
+
+
+### Key Findings
+
+- **Best Predictability:** Context-4 with 90.5% predictability
+- **Branching Factor:** Decreases with context size (more deterministic)
+- **Memory Trade-off:** Larger contexts require more storage (1,065,026 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 | 263,085 |
+| Total Tokens | 16,823,868 |
+| Mean Frequency | 63.95 |
+| Median Frequency | 4 |
+| Frequency Std Dev | 2476.07 |
+
+### Most Common Words
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | e | 717,291 |
+| 2 | ar | 526,767 |
+| 3 | a | 475,234 |
+| 4 | an | 331,609 |
+| 5 | ha | 233,395 |
+| 6 | c | 194,241 |
+| 7 | gant | 192,785 |
+| 8 | en | 189,181 |
+| 9 | da | 173,430 |
+| 10 | rummad | 169,617 |
+
+### Least Common Words (from vocabulary)
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | maghrebonkoud | 2 |
+| 2 | fidefide | 2 |
+| 3 | 2024he | 2 |
+| 4 | ougandachess | 2 |
+| 5 | ouganda365 | 2 |
+| 6 | inmediares | 2 |
+| 7 | cytonn | 2 |
+| 8 | malinga | 2 |
+| 9 | ablainville | 2 |
+| 10 | remonter | 2 |
+
+### Zipf's Law Analysis
+
+| Metric | Value |
+|--------|-------|
+| Zipf Coefficient | 1.1027 |
+| R² (Goodness of Fit) | 0.995216 |
+| Adherence Quality | **excellent** |
+
+### Coverage Analysis
+
+| Top N Words | Coverage |
+|-------------|----------|
+| Top 100 | 40.0% |
+| Top 1,000 | 64.0% |
+| Top 5,000 | 79.7% |
+| Top 10,000 | 85.1% |
+
+### Key Findings
+
+- **Zipf Compliance:** R²=0.9952 indicates excellent adherence to Zipf's law
+- **High Frequency Dominance:** Top 100 words cover 40.0% of corpus
+- **Long Tail:** 253,085 words needed for remaining 14.9% 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** | 165,065 | 32 | 3.529 | 1.159 | 0.8322 🏆 |
+| **mono_64d** | 165,065 | 64 | 4.045 | 1.139 | 0.8224 |
+| **mono_128d** | 165,065 | 128 | 4.668 | 1.127 | 0.8005 |
+| **embeddings_enhanced** | 0 | 0 | 0.000 | 0.000 | 0.0000 |
+
+### Key Findings
+
+- **Best Isotropy:** mono_32d with 0.8322 (more uniform distribution)
+- **Dimension Trade-off:** Higher dimensions capture more semantics but reduce isotropy
+- **Vocabulary Coverage:** All models cover 165,065 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 (3.62x) with low UNK rate |
+| N-gram | **5-gram** | Lowest perplexity (361) |
+| Markov | **Context-4** | Highest predictability (90.5%) |
+| Embeddings | **100d** | Balanced semantic capture and isotropy |
+
+---
+## Appendix: Metrics Glossary & Interpretation Guide
+
+This section provides definitions, intuitions, and guidance for interpreting the metrics used throughout this report.
+
+### Tokenizer Metrics
+
+**Compression Ratio**
+> *Definition:* The ratio of characters to tokens (chars/token). Measures how efficiently the tokenizer represents text.
+>
+> *Intuition:* Higher compression means fewer tokens needed to represent the same text, reducing sequence lengths for downstream models. A 3x compression means ~3 characters per token on average.
+>
+> *What to seek:* Higher is generally better for efficiency, but extremely high compression may indicate overly aggressive merging that loses morphological information.
+
+**Average Token Length (Fertility)**
+> *Definition:* Mean number of characters per token produced by the tokenizer.
+>
+> *Intuition:* Reflects the granularity of tokenization. Longer tokens capture more context but may struggle with rare words; shorter tokens are more flexible but increase sequence length.
+>
+> *What to seek:* Balance between 2-5 characters for most languages. Arabic/morphologically-rich languages may benefit from slightly longer tokens.
+
+**Unknown Token Rate (OOV Rate)**
+> *Definition:* Percentage of tokens that map to the unknown/UNK token, indicating words the tokenizer cannot represent.
+>
+> *Intuition:* Lower OOV means better vocabulary coverage. High OOV indicates the tokenizer encounters many unseen character sequences.
+>
+> *What to seek:* Below 1% is excellent; below 5% is acceptable. BPE tokenizers typically achieve very low OOV due to subword fallback.
+
+### N-gram Model Metrics
+
+**Perplexity**
+> *Definition:* Measures how "surprised" the model is by test data. Mathematically: 2^(cross-entropy). Lower values indicate better prediction.
+>
+> *Intuition:* If perplexity is 100, the model is as uncertain as if choosing uniformly among 100 options at each step. A perplexity of 10 means effectively choosing among 10 equally likely options.
+>
+> *What to seek:* Lower is better. Perplexity decreases with larger n-grams (more context). Values vary widely by language and corpus size.
+
+**Entropy**
+> *Definition:* Average information content (in bits) needed to encode the next token given the context. Related to perplexity: perplexity = 2^entropy.
+>
+> *Intuition:* High entropy means high uncertainty/randomness; low entropy means predictable patterns. Natural language typically has entropy between 1-4 bits per character.
+>
+> *What to seek:* Lower entropy indicates more predictable text patterns. Entropy should decrease as n-gram size increases.
+
+**Coverage (Top-K)**
+> *Definition:* Percentage of corpus occurrences explained by the top K most frequent n-grams.
+>
+> *Intuition:* High coverage with few patterns indicates repetitive/formulaic text; low coverage suggests diverse vocabulary usage.
+>
+> *What to seek:* Depends on use case. For language modeling, moderate coverage (40-60% with top-1000) is typical for natural text.
+
+### Markov Chain Metrics
+
+**Average Entropy**
+> *Definition:* Mean entropy across all contexts, measuring average uncertainty in next-word prediction.
+>
+> *Intuition:* Lower entropy means the model is more confident about what comes next. Context-1 has high entropy (many possible next words); Context-4 has low entropy (few likely continuations).
+>
+> *What to seek:* Decreasing entropy with larger context sizes. Very low entropy (<0.1) indicates highly deterministic transitions.
+
+**Branching Factor**
+> *Definition:* Average number of unique next tokens observed for each context.
+>
+> *Intuition:* High branching = many possible continuations (flexible but uncertain); low branching = few options (predictable but potentially repetitive).
+>
+> *What to seek:* Branching factor should decrease with context size. Values near 1.0 indicate nearly deterministic chains.
+
+**Predictability**
+> *Definition:* Derived metric: (1 - normalized_entropy) × 100%. Indicates how deterministic the model's predictions are.
+>
+> *Intuition:* 100% predictability means the next word is always certain; 0% means completely random. Real text falls between these extremes.
+>
+> *What to seek:* Higher predictability for text generation quality, but too high (>98%) may produce repetitive output.
+
+### Vocabulary & Zipf's Law Metrics
+
+**Zipf's Coefficient**
+> *Definition:* The slope of the log-log plot of word frequency vs. rank. Zipf's law predicts this should be approximately -1.
+>
+> *Intuition:* A coefficient near -1 indicates the corpus follows natural language patterns where a few words are very common and most words are rare.
+>
+> *What to seek:* Values between -0.8 and -1.2 indicate healthy natural language distribution. Deviations may suggest domain-specific or artificial text.
+
+**R² (Coefficient of Determination)**
+> *Definition:* Measures how well the linear fit explains the frequency-rank relationship. Ranges from 0 to 1.
+>
+> *Intuition:* R² near 1.0 means the data closely follows Zipf's law; lower values indicate deviation from expected word frequency patterns.
+>
+> *What to seek:* R² > 0.95 is excellent; > 0.99 indicates near-perfect Zipf adherence typical of large natural corpora.
+
+**Vocabulary Coverage**
+> *Definition:* Cumulative percentage of corpus tokens accounted for by the top N words.
+>
+> *Intuition:* Shows how concentrated word usage is. If top-100 words cover 50% of text, the corpus relies heavily on common words.
+>
+> *What to seek:* Top-100 covering 30-50% is typical. Higher coverage indicates more repetitive text; lower suggests richer vocabulary.
+
+### Word Embedding Metrics
+
+**Isotropy**
+> *Definition:* Measures how uniformly distributed vectors are in the embedding space. Computed as the ratio of minimum to maximum singular values.
+>
+> *Intuition:* High isotropy (near 1.0) means vectors spread evenly in all directions; low isotropy means vectors cluster in certain directions, reducing expressiveness.
+>
+> *What to seek:* Higher isotropy generally indicates better-quality embeddings. Values > 0.1 are reasonable; > 0.3 is good. Lower-dimensional embeddings tend to have higher isotropy.
+
+**Average Norm**
+> *Definition:* Mean magnitude (L2 norm) of word vectors in the embedding space.
+>
+> *Intuition:* Indicates the typical "length" of vectors. Consistent norms suggest stable training; high variance may indicate some words are undertrained.
+>
+> *What to seek:* Relatively consistent norms across models. The absolute value matters less than consistency (low std deviation).
+
+**Cosine Similarity**
+> *Definition:* Measures angular similarity between vectors, ranging from -1 (opposite) to 1 (identical direction).
+>
+> *Intuition:* Words with similar meanings should have high cosine similarity. This is the standard metric for semantic relatedness in embeddings.
+>
+> *What to seek:* Semantically related words should score > 0.5; unrelated words should be near 0. Synonyms often score > 0.7.
+
+**t-SNE Visualization**
+> *Definition:* t-Distributed Stochastic Neighbor Embedding - a dimensionality reduction technique that preserves local structure for visualization.
+>
+> *Intuition:* Clusters in t-SNE plots indicate groups of semantically related words. Spread indicates vocabulary diversity; tight clusters suggest semantic coherence.
+>
+> *What to seek:* Meaningful clusters (e.g., numbers together, verbs together). Avoid over-interpreting distances - t-SNE preserves local, not global, structure.
+
+### General Interpretation Guidelines
+
+1. **Compare within model families:** Metrics are most meaningful when comparing models of the same type (e.g., 8k vs 64k tokenizer).
+2. **Consider trade-offs:** Better performance on one metric often comes at the cost of another (e.g., compression vs. OOV rate).
+3. **Context matters:** Optimal values depend on downstream tasks. Text generation may prioritize different metrics than classification.
+4. **Corpus influence:** All metrics are influenced by corpus characteristics. Wikipedia text differs from social media or literature.
+5. **Language-specific patterns:** Morphologically rich languages (like Arabic) may show different optimal ranges than analytic languages.
+
+
+### Visualizations Index
+
+| Visualization | Description |
+|---------------|-------------|
+| Tokenizer Compression | Compression ratios by vocabulary size |
+| Tokenizer Fertility | Average token length by vocabulary |
+| Tokenizer OOV | Unknown token rates |
+| Tokenizer Total Tokens | Total tokens by vocabulary |
+| N-gram Perplexity | Perplexity by n-gram size |
+| N-gram Entropy | Entropy by n-gram size |
+| N-gram Coverage | Top pattern coverage |
+| N-gram Unique | Unique n-gram counts |
+| Markov Entropy | Entropy by context size |
+| Markov Branching | Branching factor by context |
+| Markov Contexts | Unique context counts |
+| Zipf's Law | Frequency-rank distribution with fit |
+| Vocab Frequency | Word frequency distribution |
+| Top 20 Words | Most frequent words |
+| Vocab Coverage | Cumulative coverage curve |
+| Embedding Isotropy | Vector space uniformity |
+| Embedding Norms | Vector magnitude distribution |
+| Embedding Similarity | Word similarity heatmap |
+| Nearest Neighbors | Similar words for key terms |
+| t-SNE Words | 2D word embedding visualization |
+| t-SNE Sentences | 2D sentence embedding visualization |
+| Position Encoding | Encoding method comparison |
+| Model Sizes | Storage requirements |
+| Performance Dashboard | Comprehensive performance overview |
+
+---
+## About This Project
+
+### Data Source
+
+Models trained on [wikipedia-monthly](https://huggingface.co/datasets/omarkamali/wikipedia-monthly) - a monthly snapshot of Wikipedia articles across 300+ languages.
+
+### Project
+
+A project by **[Wikilangs](https://wikilangs.org)** - Open-source NLP models for every Wikipedia language.
+
+### Maintainer
+
+[Omar Kamali](https://omarkamali.com) - [Omneity Labs](https://omneitylabs.com)
+
+### Citation
+
+If you use these models in your research, please cite:
+
+```bibtex
+@misc{wikilangs2025,
+  author = {Kamali, Omar},
+  title = {Wikilangs: Open NLP Models for Wikipedia Languages},
+  year = {2025},
+  publisher = {HuggingFace},
+  url = {https://huggingface.co/wikilangs}
+  institution = {Omneity Labs}
+}
+```
+
+### License
+
+MIT License - Free for academic and commercial use.
+
+### Links
+
+- 🌐 Website: [wikilangs.org](https://wikilangs.org)
+- 🤗 Models: [huggingface.co/wikilangs](https://huggingface.co/wikilangs)
+- 📊 Data: [wikipedia-monthly](https://huggingface.co/datasets/omarkamali/wikipedia-monthly)
+- 👤 Author: [Omar Kamali](https://huggingface.co/omarkamali)
+---
+*Generated by Wikilangs Models Pipeline*
+
+*Report Date: 2025-12-28 08:17:06*

models/embeddings/monolingual/br_128d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3454154976810c137904a12ad207abedf1ec5887e8aca2f5a7122ad0f8d5930
+size 1195958360

models/embeddings/monolingual/br_128d.meta.json

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

models/embeddings/monolingual/br_128d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "br",
+  "dimension": 128,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 128,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 165065
+}

models/embeddings/monolingual/br_32d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64e33fc4be67c9fbddd42b666550e7082d827aabcc323efa5fa8f0d43bbce29b
+size 301188440

models/embeddings/monolingual/br_32d.meta.json

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

models/embeddings/monolingual/br_32d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "br",
+  "dimension": 32,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 32,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 165065
+}

models/embeddings/monolingual/br_64d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9f750fa84e5a98167cab2dbbbb436fa61f2fe773cee5d410fe13aeafa89b674
+size 599445080

models/embeddings/monolingual/br_64d.meta.json

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

models/embeddings/monolingual/br_64d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "br",
+  "dimension": 64,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 64,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 165065
+}

models/subword_markov/br_markov_ctx1_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7589832008594b2044eb3a660f9c0ee1c38b0a71dc2cf76a5e11e054aa0709c2
+size 359681

models/subword_markov/br_markov_ctx1_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "subword",
+  "language": "br",
+  "unique_contexts": 6849,
+  "total_transitions": 100507380
+}

models/subword_markov/br_markov_ctx2_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b80fd5521997f26184b281b5d6b8bbdead1d92da1152a2bc63ff0293af7aaeaa
+size 1941406

models/subword_markov/br_markov_ctx2_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "subword",
+  "language": "br",
+  "unique_contexts": 52722,
+  "total_transitions": 100417214
+}

models/subword_markov/br_markov_ctx3_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:909df4c05b66b69687d1d8f920db593ab05009de50d4a603df49a4228042cb35
+size 8952059

models/subword_markov/br_markov_ctx3_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "subword",
+  "language": "br",
+  "unique_contexts": 256241,
+  "total_transitions": 100327048
+}

models/subword_markov/br_markov_ctx4_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2615bccc86f4696fdc966eac1f934f561672e8ed319db55e0b79b9044123108
+size 27438080

models/subword_markov/br_markov_ctx4_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "subword",
+  "language": "br",
+  "unique_contexts": 1065026,
+  "total_transitions": 100236882
+}

models/subword_ngram/br_2gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97148ef1780fce9d7c84e24a99d7d18050738bc639866e47145c2650f4d24549
+size 180245

models/subword_ngram/br_2gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "subword",
+  "language": "br",
+  "unique_ngrams": 13774,
+  "total_ngrams": 100507380
+}

models/subword_ngram/br_3gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5896217e8144736c8569249112eb653f0c7159070b9cdcc4fab60365ffc6f9ea
+size 1306949

models/subword_ngram/br_3gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "subword",
+  "language": "br",
+  "unique_ngrams": 106981,
+  "total_ngrams": 100417214
+}

models/subword_ngram/br_4gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:018393f222505ae1b2a7dc78a37fb2b916100fe4fd5ae7b93b4852d77884fc8d
+size 6695773

models/subword_ngram/br_4gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "subword",
+  "language": "br",
+  "unique_ngrams": 580225,
+  "total_ngrams": 100327048
+}

models/tokenizer/br_tokenizer_16k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a752f92ace2ce968d9f9080803babfaec23f0f941b90aeaef31fda3dfce785ad
+size 501248

models/tokenizer/br_tokenizer_32k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1de8e69da6c92acfc056827b6f959e3a09c7f5e3833dbf06d735ef89d1a4415
+size 772253

models/tokenizer/br_tokenizer_64k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:952d3588c7c165d850f099355da1f2fb472e90d978c3fca8d4064f335f551dce
+size 1327800

models/tokenizer/br_tokenizer_8k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae0f58e053910435e1f0c3469037b7e403142c364ac16080087a71a6a1757cb9
+size 369734

models/vocabulary/br_vocabulary.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e743e0213a6e24e64e5597ee9208bbcc913f68ffe7dfd9358fa210925a8f67c
+size 4050849

models/vocabulary/br_vocabulary_metadata.json

@@ -0,0 +1,16 @@
+{
+  "language": "br",
+  "vocabulary_size": 263085,
+  "statistics": {
+    "type_token_ratio": 0.03735837819296646,
+    "coverage": {
+      "top_100": 0.3910641436030635,
+      "top_1000": 0.6260291451547847,
+      "top_5000": 0.779592055722224,
+      "top_10000": 0.8327078880623957
+    },
+    "hapax_count": 379609,
+    "hapax_ratio": 0.5906527834397085,
+    "total_documents": 90166
+  }
+}

models/word_markov/br_markov_ctx1_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29d6439c7c56bab6e2e37e1f57fd518d0ba61a6913f7d0eb2a4fea4b55dbef56
+size 35882245

models/word_markov/br_markov_ctx1_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "word",
+  "language": "br",
+  "unique_contexts": 642948,
+  "total_transitions": 22080722
+}

models/word_markov/br_markov_ctx2_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ceff4f844e237ead4ea08a340d17ef9f3e7fe62d5bc6d7e8c3d3b31b59d2cf3d
+size 102417911

models/word_markov/br_markov_ctx2_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "word",
+  "language": "br",
+  "unique_contexts": 4087653,
+  "total_transitions": 21990556
+}

models/word_markov/br_markov_ctx3_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a8a1170156b1dcf6b52e7667e73eb5c87efc4eed7afa9a113410a6fc6268928
+size 178712506

models/word_markov/br_markov_ctx3_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "word",
+  "language": "br",
+  "unique_contexts": 9600283,
+  "total_transitions": 21900429
+}

models/word_markov/br_markov_ctx4_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4dc51b5cdaf1e6d6f00bfc342106c2a491d891dca552edee6c7dc09f3e43da76
+size 237588645

models/word_markov/br_markov_ctx4_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "word",
+  "language": "br",
+  "unique_contexts": 13824129,
+  "total_transitions": 21810331
+}

models/word_ngram/br_2gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0378b53defe0eedb7989c2010d751a14d99928386aba3ced76bad82ec402ec36
+size 5232515

models/word_ngram/br_2gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "word",
+  "language": "br",
+  "unique_ngrams": 396301,
+  "total_ngrams": 22080722
+}

models/word_ngram/br_3gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:513c163c0b3b0ac784029384146f1f680f6310d92cb835eb7aad27b794f28240
+size 12329602

models/word_ngram/br_3gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "word",
+  "language": "br",
+  "unique_ngrams": 858386,
+  "total_ngrams": 21990556
+}

models/word_ngram/br_4gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e17780562a29c8720acf9f11550b417c291cd4606856491e09f594c95e9fc88
+size 23878363

models/word_ngram/br_4gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "word",
+  "language": "br",
+  "unique_ngrams": 1581104,
+  "total_ngrams": 21900429
+}