Upload all models and assets for da (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 +1,576 @@
+---
+language: da
+language_name: Danish
+language_family: germanic_north
+tags:
+  - wikilangs
+  - nlp
+  - tokenizer
+  - embeddings
+  - n-gram
+  - markov
+  - wikipedia
+  - monolingual
+  - family-germanic_north
+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: 4.256
+  - name: best_isotropy
+    type: isotropy
+    value: 0.7852
+  - name: vocabulary_size
+    type: vocab
+    value: 974563
+generated: 2025-12-29
+---
+
+# Danish - Wikilangs Models
+## Comprehensive Research Report & Full Ablation Study
+
+This repository contains NLP models trained and evaluated by Wikilangs, specifically on **Danish** 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** | 3.425x | 3.39 | 0.1175% | 1,887,939 |
+| **16k** | 3.744x | 3.70 | 0.1284% | 1,727,173 |
+| **32k** | 4.025x | 3.98 | 0.1381% | 1,606,621 |
+| **64k** | 4.256x 🏆 | 4.21 | 0.1460% | 1,519,331 |
+
+### Tokenization Examples
+
+Below are sample sentences tokenized with each vocabulary size:
+
+**Sample 1:** `Se også 564 (tal)
+
+ Begivenheder 
+
+ Født 
+
+ Dødsfald 
+
+ Eksterne henvisninger 
+
+...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁se ▁også ▁ 5 6 4 ▁( tal ) ▁begivenheder ... (+7 more)` | 17 |
+| 16k | `▁se ▁også ▁ 5 6 4 ▁( tal ) ▁begivenheder ... (+7 more)` | 17 |
+| 32k | `▁se ▁også ▁ 5 6 4 ▁( tal ) ▁begivenheder ... (+7 more)` | 17 |
+| 64k | `▁se ▁også ▁ 5 6 4 ▁( tal ) ▁begivenheder ... (+7 more)` | 17 |
+
+**Sample 2:** `Rikuya Izutsu (født 10. februar 1994) er en japansk fodboldspiller.
+
+ Referencer...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁ri ku ya ▁i z ut su ▁( født ▁ ... (+23 more)` | 33 |
+| 16k | `▁ri ku ya ▁i z ut su ▁( født ▁ ... (+23 more)` | 33 |
+| 32k | `▁ri ku ya ▁iz ut su ▁( født ▁ 1 ... (+22 more)` | 32 |
+| 64k | `▁ri ku ya ▁iz utsu ▁( født ▁ 1 0 ... (+21 more)` | 31 |
+
+**Sample 3:** `Se også 598 (tal)
+
+ Begivenheder 
+
+ Født 
+
+ Dødsfald 
+
+ Eksterne henvisninger 
+
+...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁se ▁også ▁ 5 9 8 ▁( tal ) ▁begivenheder ... (+7 more)` | 17 |
+| 16k | `▁se ▁også ▁ 5 9 8 ▁( tal ) ▁begivenheder ... (+7 more)` | 17 |
+| 32k | `▁se ▁også ▁ 5 9 8 ▁( tal ) ▁begivenheder ... (+7 more)` | 17 |
+| 64k | `▁se ▁også ▁ 5 9 8 ▁( tal ) ▁begivenheder ... (+7 more)` | 17 |
+
+
+### Key Findings
+
+- **Best Compression:** 64k achieves 4.256x compression
+- **Lowest UNK Rate:** 8k with 0.1175% 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** | 147,251 🏆 | 17.17 | 1,987,899 | 9.8% | 23.4% |
+| **2-gram** | 347 🏆 | 8.44 | 19,771 | 62.6% | 98.3% |
+| **3-gram** | 860,751 | 19.72 | 4,757,337 | 4.0% | 10.6% |
+| **3-gram** | 3,240 | 11.66 | 192,309 | 23.7% | 65.1% |
+| **4-gram** | 2,463,069 | 21.23 | 8,589,777 | 2.8% | 7.0% |
+| **4-gram** | 20,967 | 14.36 | 1,299,708 | 11.6% | 35.0% |
+
+### Top 5 N-grams by Size
+
+**2-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `kategori :` | 797,923 |
+| 2 | `, og` | 388,695 |
+| 3 | `, der` | 367,530 |
+| 4 | `. i` | 304,754 |
+| 5 | `, som` | 301,325 |
+
+**3-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `henvisninger kategori :` | 91,987 |
+| 2 | `eksterne henvisninger kategori` | 84,554 |
+| 3 | `danmark kategori :` | 75,266 |
+| 4 | `referencer eksterne henvisninger` | 69,753 |
+| 5 | `) er en` | 69,674 |
+
+**4-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `eksterne henvisninger kategori :` | 84,554 |
+| 2 | `fra danmark kategori :` | 64,734 |
+| 3 | `referencer eksterne henvisninger kategori` | 49,030 |
+| 4 | `kategori : fodboldspillere fra` | 42,815 |
+| 5 | `bl . a .` | 42,727 |
+
+
+### Key Findings
+
+- **Best Perplexity:** 2-gram with 347
+- **Entropy Trend:** Decreases with larger n-grams (more predictable)
+- **Coverage:** Top-1000 patterns cover ~35% 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** | 0.7002 | 1.625 | 8.20 | 2,503,524 | 30.0% |
+| **1** | 1.4386 | 2.710 | 9.82 | 7,827 | 0.0% |
+| **2** | 0.4144 | 1.333 | 2.67 | 20,511,480 | 58.6% |
+| **2** | 0.7435 | 1.674 | 5.32 | 76,803 | 25.6% |
+| **3** | 0.1972 | 1.146 | 1.49 | 54,702,359 | 80.3% |
+| **3** | 0.9118 | 1.881 | 5.38 | 408,247 | 8.8% |
+| **4** | 0.0910 🏆 | 1.065 | 1.18 | 81,699,919 | 90.9% |
+| **4** | 0.7874 🏆 | 1.726 | 3.84 | 2,194,966 | 21.3% |
+
+### Generated Text Samples
+
+Below are text samples generated from each Markov chain model:
+
+**Context Size 1:**
+
+1. `. ali var således at lukke alle ved nogle indfødte var de fleste insektarter fundet sted`
+2. `, fra 1979 fuldmægtig ved opførelsen af en metrostation nedenunder , men palle von gyldenskiold (`
+3. `i fjendskab mellem køge bugt , som ejendommen var blevet beskrevet som har stenz . von`
+
+**Context Size 2:**
+
+1. `kategori : arveret kategori : sovjetiske film fra 2009 kategori : museer i oslo kategori : modtagere`
+2. `, og det sydlige hellas og bliver dræbt af en anden organisation , der stod udenfor eu`
+3. `, der indsamler penge til sin karriere og er det dominerende træ . gunnar carlquist ( født`
+
+**Context Size 3:**
+
+1. `henvisninger kategori : ideologier kategori : etnicitet kategori : immaterialret kategori : straffel...`
+2. `eksterne henvisninger kategori : kenyas provinser`
+3. `danmark kategori : fløjtenister fra danmark kategori : maskinfabrikker i danmark kategori : det dans...`
+
+**Context Size 4:**
+
+1. `eksterne henvisninger kategori : fugle fra vestasien kategori : ræve`
+2. `fra danmark kategori : personer i dansk biografisk leksikon kategori : tyskere i 1900 - tallet kateg...`
+3. `referencer eksterne henvisninger kategori : elvis presley - sange kategori : singler fra 1963 katego...`
+
+
+### Key Findings
+
+- **Best Predictability:** Context-4 with 90.9% predictability
+- **Branching Factor:** Decreases with context size (more deterministic)
+- **Memory Trade-off:** Larger contexts require more storage (2,194,966 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 | 974,563 |
+| Total Tokens | 94,217,021 |
+| Mean Frequency | 96.68 |
+| Median Frequency | 4 |
+| Frequency Std Dev | 6265.69 |
+
+### Most Common Words
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | i | 3,414,957 |
+| 2 | og | 2,580,527 |
+| 3 | af | 1,724,110 |
+| 4 | en | 1,367,373 |
+| 5 | til | 1,141,339 |
+| 6 | er | 1,091,325 |
+| 7 | den | 1,045,804 |
+| 8 | at | 983,539 |
+| 9 | på | 952,118 |
+| 10 | som | 942,026 |
+
+### Least Common Words (from vocabulary)
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | sinoefloden | 2 |
+| 2 | deathconsciousness | 2 |
+| 3 | folkedanseforeninger | 2 |
+| 4 | affranchi | 2 |
+| 5 | superfilmen | 2 |
+| 6 | kettletoft | 2 |
+| 7 | sandays | 2 |
+| 8 | crummack | 2 |
+| 9 | rousays | 2 |
+| 10 | 2025919140 | 2 |
+
+### Zipf's Law Analysis
+
+| Metric | Value |
+|--------|-------|
+| Zipf Coefficient | 1.0148 |
+| R² (Goodness of Fit) | 0.997060 |
+| Adherence Quality | **excellent** |
+
+### Coverage Analysis
+
+| Top N Words | Coverage |
+|-------------|----------|
+| Top 100 | 36.6% |
+| Top 1,000 | 57.4% |
+| Top 5,000 | 73.3% |
+| Top 10,000 | 79.6% |
+
+### Key Findings
+
+- **Zipf Compliance:** R²=0.9971 indicates excellent adherence to Zipf's law
+- **High Frequency Dominance:** Top 100 words cover 36.6% of corpus
+- **Long Tail:** 964,563 words needed for remaining 20.4% 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** | 679,813 | 32 | 3.066 | 0.860 | 0.7852 🏆 |
+| **mono_64d** | 679,813 | 64 | 3.485 | 0.868 | 0.7644 |
+| **mono_128d** | 679,813 | 128 | 3.895 | 0.896 | 0.7063 |
+| **embeddings_enhanced** | 0 | 0 | 0.000 | 0.000 | 0.0000 |
+
+### Key Findings
+
+- **Best Isotropy:** mono_32d with 0.7852 (more uniform distribution)
+- **Dimension Trade-off:** Higher dimensions capture more semantics but reduce isotropy
+- **Vocabulary Coverage:** All models cover 679,813 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 (4.26x) with low UNK rate |
+| N-gram | **5-gram** | Lowest perplexity (347) |
+| Markov | **Context-4** | Highest predictability (90.9%) |
+| 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: 2025-12-29 09:01:56*

models/embeddings/monolingual/da_128d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68c3acea0b3acc83f64903aac63268a5886819a3724537b20264e8742bbe0389
+size 1733202257

models/embeddings/monolingual/da_128d.meta.json

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

models/embeddings/monolingual/da_128d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "da",
+  "dimension": 128,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 128,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 679813
+}

models/embeddings/monolingual/da_32d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de1aa27bfa4123e9f8a42dedeb8f7bc963d4dce71c567da730cd9d71333333e2
+size 443105873

models/embeddings/monolingual/da_32d.meta.json

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

models/embeddings/monolingual/da_32d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "da",
+  "dimension": 32,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 32,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 679813
+}

models/embeddings/monolingual/da_64d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:465fb7807ee1c1b7cc0cbc92c8b94765964d007aaca0e89095d66ba847a12af5
+size 873138001

models/embeddings/monolingual/da_64d.meta.json

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

models/embeddings/monolingual/da_64d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "da",
+  "dimension": 64,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 64,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 679813
+}

models/subword_markov/da_markov_ctx1_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36e2282c95edc33f735242830727d3cb38b457f0d60410f8fc958ae54a853cee
+size 501415

models/subword_markov/da_markov_ctx1_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "subword",
+  "language": "da",
+  "unique_contexts": 7827,
+  "total_transitions": 630161089
+}

models/subword_markov/da_markov_ctx2_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f3e3b34698c83d39fb848b9ed450e9d333640fb38b7af701911a7bf5b6c4470
+size 3244857

models/subword_markov/da_markov_ctx2_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "subword",
+  "language": "da",
+  "unique_contexts": 76803,
+  "total_transitions": 629849363
+}

models/subword_markov/da_markov_ctx3_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1a5503d267e8fad2c6f3d8d57b043b8b39f15768b2399cf0cf521aba952aa5d
+size 16479657

models/subword_markov/da_markov_ctx3_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "subword",
+  "language": "da",
+  "unique_contexts": 408247,
+  "total_transitions": 629537637
+}

models/subword_markov/da_markov_ctx4_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07015bd1fa896217f9ece588bbcfab316cebe27c1f5353d7bd0afb2b0f4effb4
+size 64259074

models/subword_markov/da_markov_ctx4_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "subword",
+  "language": "da",
+  "unique_contexts": 2194966,
+  "total_transitions": 629225911
+}

models/subword_ngram/da_2gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8ca2619f57b1ac7bbefcabc250a3418b8e0a026afbf506eb2d9ec32d51d4dc8
+size 273435

models/subword_ngram/da_2gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "subword",
+  "language": "da",
+  "unique_ngrams": 19771,
+  "total_ngrams": 630161089
+}

models/subword_ngram/da_3gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14e01ea2fde43d22f7708e4abd070d10512956496ee36e85a20a8a92ce15358f
+size 2327975

models/subword_ngram/da_3gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "subword",
+  "language": "da",
+  "unique_ngrams": 192309,
+  "total_ngrams": 629849363
+}

models/subword_ngram/da_4gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13389e44c39a59c594db7e4f570fdfddecb9c53507210fcd712e5a2e80816e26
+size 15313345

models/subword_ngram/da_4gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "subword",
+  "language": "da",
+  "unique_ngrams": 1299708,
+  "total_ngrams": 629537637
+}

models/tokenizer/da_tokenizer_16k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5257f8c08d312560577f764125bf86dee3ad20060303d5d179efc751e6926aaa
+size 509881

models/tokenizer/da_tokenizer_32k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72f637e73e0c17b40cdde4b080377b234f3d7ddb88a0836e27d0124aec823468
+size 792817

models/tokenizer/da_tokenizer_64k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f34a8719a8206abf56c31b11ccd067be63136835dddf9ee6648468ac9b749ed
+size 1376877

models/tokenizer/da_tokenizer_8k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a9341fa8df03ff9f7d47d5875e47073097d21830ea7a2b2b07ddda14141fb9e
+size 372662

models/vocabulary/da_vocabulary.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5eaf29d24437f2bdfcf35e6f6e0f105f1b298b3da7b0c1e2899592ab05ed482
+size 15264314

models/vocabulary/da_vocabulary_metadata.json

@@ -0,0 +1,16 @@
+{
+  "language": "da",
+  "vocabulary_size": 974563,
+  "statistics": {
+    "type_token_ratio": 0.02614497849971959,
+    "coverage": {
+      "top_100": 0.36016111340236506,
+      "top_1000": 0.5648239365833638,
+      "top_5000": 0.7214206987908641,
+      "top_10000": 0.7832362713874816
+    },
+    "hapax_count": 1528707,
+    "hapax_ratio": 0.6106840252949143,
+    "total_documents": 311726
+  }
+}

models/word_markov/da_markov_ctx1_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd504f360cbe5a90b1c7b013522b8db902e17e2e16ba3ba9c36f60150cd26922
+size 194239107

models/word_markov/da_markov_ctx1_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "word",
+  "language": "da",
+  "unique_contexts": 2503524,
+  "total_transitions": 118622822
+}

models/word_markov/da_markov_ctx2_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:594706495f332b8802a7647db03b3acd0fc22282c1fbc02ec68f34b4f84095d8
+size 631690789

models/word_markov/da_markov_ctx2_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "word",
+  "language": "da",
+  "unique_contexts": 20511480,
+  "total_transitions": 118311096
+}

models/word_markov/da_markov_ctx3_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4bfb0119561556428450fed5e0734bc4b4fe8465a3c95ab3d8ff351e5a77222
+size 1177486879

models/word_markov/da_markov_ctx3_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "word",
+  "language": "da",
+  "unique_contexts": 54702359,
+  "total_transitions": 117999399
+}

models/word_markov/da_markov_ctx4_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1fddfa27a0ab301ae8109bf1efff0d813130fcf435f4d8ad93ff5de65d5668b
+size 1562558497

models/word_markov/da_markov_ctx4_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "word",
+  "language": "da",
+  "unique_contexts": 81699919,
+  "total_transitions": 117687712
+}

models/word_ngram/da_2gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd7b6757d9fb0a62ac18504d10e1d1a654a45cf4d054563749c97771dce306b9
+size 28868531

models/word_ngram/da_2gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "word",
+  "language": "da",
+  "unique_ngrams": 1987899,
+  "total_ngrams": 118622822
+}

models/word_ngram/da_3gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6efb68ff7c45ace7310c437470ae9af5d2a7d0fe7e569a00a80cbcc2d6d02de1
+size 73998481

models/word_ngram/da_3gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "word",
+  "language": "da",
+  "unique_ngrams": 4757337,
+  "total_ngrams": 118311096
+}

models/word_ngram/da_4gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25bf8faa87743209fd95d073a46772678a45830bccee581ce6568264bab97912
+size 141207105

models/word_ngram/da_4gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "word",
+  "language": "da",
+  "unique_ngrams": 8589777,
+  "total_ngrams": 117999399
+}