Upload all models and assets for anp (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 +1,557 @@
+---
+language: anp
+language_name: ANP
+language_family: indoaryan_central
+tags:
+  - wikilangs
+  - nlp
+  - tokenizer
+  - embeddings
+  - n-gram
+  - markov
+  - wikipedia
+  - monolingual
+  - family-indoaryan_central
+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: 4.233
+  - name: best_isotropy
+    type: isotropy
+    value: 0.8322
+  - name: vocabulary_size
+    type: vocab
+    value: 9567
+generated: 2025-12-27
+---
+
+# ANP - Wikilangs Models
+## Comprehensive Research Report & Full Ablation Study
+
+This repository contains NLP models trained and evaluated by Wikilangs, specifically on **ANP** 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** | 3.431x | 3.37 | 0.0938% | 441,405 |
+| **16k** | 3.755x | 3.69 | 0.1026% | 403,338 |
+| **32k** | 4.015x | 3.95 | 0.1098% | 377,196 |
+| **64k** | 4.233x 🏆 | 4.16 | 0.1157% | 357,719 |
+
+### Tokenization Examples
+
+Below are sample sentences tokenized with each vocabulary size:
+
+**Sample 1:** `मंजूषा कला अंगप्रदेश के एक बहुचर्चित लोकगाथा बिहुला विषहरी पर आधारित छै।`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁म ंज ूष ा ▁कला ▁अंग प्र देश ▁के ▁एक ... (+16 more)` | 26 |
+| 16k | `▁मंज ूष ा ▁कला ▁अंग प्रदेश ▁के ▁एक ▁बहु च ... (+13 more)` | 23 |
+| 32k | `▁मंज ूष ा ▁कला ▁अंगप्रदेश ▁के ▁एक ▁बहु चर्चित ▁लोकगाथा ... (+7 more)` | 17 |
+| 64k | `▁मंजूषा ▁कला ▁अंगप्रदेश ▁के ▁एक ▁बहुचर्चित ▁लोकगाथा ▁बिहुला ▁विष हरी ... (+4 more)` | 14 |
+
+**Sample 2:** `कार्बन के रासायनिक तत्व छेकै। इ ठोस अवस्था मँ पैलौ जाय वाला अधातु छेकै।
+
+एकरो दे...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁कार्बन ▁के ▁रासायनिक ▁तत्व ▁छेकै । ▁इ ▁ठोस ▁अवस्था ▁मँ ... (+13 more)` | 23 |
+| 16k | `▁कार्बन ▁के ▁रासायनिक ▁तत्व ▁छेकै । ▁इ ▁ठोस ▁अवस्था ▁मँ ... (+11 more)` | 21 |
+| 32k | `▁कार्बन ▁के ▁रासायनिक ▁तत्व ▁छेकै । ▁इ ▁ठोस ▁अवस्था ▁मँ ... (+11 more)` | 21 |
+| 64k | `▁कार्बन ▁के ▁रासायनिक ▁तत्व ▁छेकै । ▁इ ▁ठोस ▁अवस्था ▁मँ ... (+11 more)` | 21 |
+
+**Sample 3:** `ब्रह्मपुत्र 
+श्रेणी:नद्दी`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁ब्रह्म पुत्र ▁श्रेणी : न द्दी` | 6 |
+| 16k | `▁ब्रह्मपुत्र ▁श्रेणी : न द्दी` | 5 |
+| 32k | `▁ब्रह्मपुत्र ▁श्रेणी : नद्दी` | 4 |
+| 64k | `▁ब्रह्मपुत्र ▁श्रेणी : नद्दी` | 4 |
+
+
+### Key Findings
+
+- **Best Compression:** 64k achieves 4.233x compression
+- **Lowest UNK Rate:** 8k with 0.0938% 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** | 1,473 🏆 | 10.52 | 19,190 | 39.0% | 79.4% |
+| **2-gram** | 611 🏆 | 9.26 | 5,088 | 50.3% | 92.6% |
+| **3-gram** | 10,511 | 13.36 | 71,690 | 14.3% | 44.9% |
+| **3-gram** | 4,774 | 12.22 | 39,446 | 20.6% | 57.0% |
+| **4-gram** | 37,995 | 15.21 | 200,757 | 8.2% | 28.1% |
+| **4-gram** | 20,752 | 14.34 | 155,528 | 10.6% | 34.2% |
+
+### Top 5 N-grams by Size
+
+**2-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `क े` | 45,157 |
+| 2 | `ा र` | 27,669 |
+| 3 | `् र` | 26,110 |
+| 4 | `ि क` | 21,854 |
+| 5 | `य ा` | 21,115 |
+
+**3-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `म े ं` | 15,079 |
+| 2 | `ि य ा` | 9,483 |
+| 3 | `छ ै ।` | 7,624 |
+| 4 | `् य ा` | 7,466 |
+| 5 | `ा क े` | 7,405 |
+
+**4-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `भ ा ष ा` | 4,716 |
+| 2 | `प ् र ा` | 3,403 |
+| 3 | `क े र ौ` | 3,367 |
+| 4 | `ा ह ै ।` | 3,140 |
+| 5 | `ै , ज े` | 3,079 |
+
+
+### Key Findings
+
+- **Best Perplexity:** 2-gram with 611
+- **Entropy Trend:** Decreases with larger n-grams (more predictable)
+- **Coverage:** Top-1000 patterns cover ~34% 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** | 0.6716 | 1.593 | 6.02 | 20,822 | 32.8% |
+| **1** | 1.3840 | 2.610 | 13.82 | 755 | 0.0% |
+| **2** | 0.3861 | 1.307 | 2.68 | 125,337 | 61.4% |
+| **2** | 1.1859 | 2.275 | 7.36 | 10,432 | 0.0% |
+| **3** | 0.3331 | 1.260 | 2.02 | 335,781 | 66.7% |
+| **3** | 0.8363 | 1.785 | 3.79 | 76,706 | 16.4% |
+| **4** | 0.2418 🏆 | 1.182 | 1.55 | 678,169 | 75.8% |
+| **4** | 0.5341 🏆 | 1.448 | 2.27 | 290,862 | 46.6% |
+
+### Generated Text Samples
+
+Below are text samples generated from each Markov chain model:
+
+**Context Size 1:**
+
+1. `ा ण ी कम ी ] स ा ) घ ो क ृ त ् रभ`
+2. `् डल स ं ख ् ध ा य ु न े ण अक ा प`
+3. `े म ँ स ू प ि त च ा जन ा न ह ै ज`
+
+**Context Size 2:**
+
+1. `क े प ु स ् थ अलग करत ी थ ी , ड ा इऑक ्`
+2. `ा र अध ि न े अध ि क ा ल ा म द ि य ा`
+3. `् र ो स ि मड े ग ा न , 2006 क ो खतर े क`
+
+**Context Size 3:**
+
+1. `म े ं क े द ु मक ा स ॑ ई बड ़ ऽ क ् ष`
+2. `ि य ा च ि त ह ै और यह ी ं भ ी उनक े प ा`
+3. `् य ा 1484 छ े ल ै । २८ नवम ् बर 1889 क ो और तन`
+
+**Context Size 4:**
+
+1. `भ ा ष ा स ि न ी क स ं स ् थ ा प ि त करन`
+2. `प ् र ा प ् त छ ै । शब ् द - स ा धन इत ि`
+3. `क े र ौ व ि भ ि न ् न - भ ि न ् न प ्`
+
+
+### Key Findings
+
+- **Best Predictability:** Context-4 with 75.8% predictability
+- **Branching Factor:** Decreases with context size (more deterministic)
+- **Memory Trade-off:** Larger contexts require more storage (290,862 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 | 9,567 |
+| Total Tokens | 1,448,534 |
+| Mean Frequency | 151.41 |
+| Median Frequency | 4 |
+| Frequency Std Dev | 2528.69 |
+
+### Most Common Words
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | क | 136,096 |
+| 2 | र | 92,258 |
+| 3 | स | 77,006 |
+| 4 | म | 59,103 |
+| 5 | त | 56,261 |
+| 6 | न | 54,426 |
+| 7 | य | 53,099 |
+| 8 | ल | 48,343 |
+| 9 | प | 45,425 |
+| 10 | ह | 44,325 |
+
+### Least Common Words (from vocabulary)
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | css | 2 |
+| 2 | zeros | 2 |
+| 3 | ignored | 2 |
+| 4 | dmy | 2 |
+| 5 | mdy | 2 |
+| 6 | paren | 2 |
+| 7 | breaking | 2 |
+| 8 | inserted | 2 |
+| 9 | values | 2 |
+| 10 | separator | 2 |
+
+### Zipf's Law Analysis
+
+| Metric | Value |
+|--------|-------|
+| Zipf Coefficient | 1.4601 |
+| R² (Goodness of Fit) | 0.993188 |
+| Adherence Quality | **excellent** |
+
+### Coverage Analysis
+
+| Top N Words | Coverage |
+|-------------|----------|
+| Top 100 | 82.8% |
+| Top 1,000 | 96.2% |
+| Top 5,000 | 99.3% |
+| Top 10,000 | 0.0% |
+
+### Key Findings
+
+- **Zipf Compliance:** R²=0.9932 indicates excellent adherence to Zipf's law
+- **High Frequency Dominance:** Top 100 words cover 82.8% of corpus
+- **Long Tail:** -433 words needed for remaining 100.0% 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** | 12,167 | 32 | 3.479 | 0.971 | 0.8322 🏆 |
+| **mono_64d** | 12,167 | 64 | 3.809 | 0.905 | 0.7196 |
+| **mono_128d** | 12,167 | 128 | 3.964 | 0.866 | 0.3647 |
+| **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 12,167 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 (4.23x) with low UNK rate |
+| N-gram | **5-gram** | Lowest perplexity (611) |
+| Markov | **Context-4** | Highest predictability (75.8%) |
+| Embeddings | **100d** | Balanced semantic capture and isotropy |
+
+---
+## Appendix: Metrics Glossary & Interpretation Guide
+
+This section provides definitions, intuitions, and guidance for interpreting the metrics used throughout this report.
+
+### Tokenizer Metrics
+
+**Compression Ratio**
+> *Definition:* The ratio of characters to tokens (chars/token). Measures how efficiently the tokenizer represents text.
+>
+> *Intuition:* Higher compression means fewer tokens needed to represent the same text, reducing sequence lengths for downstream models. A 3x compression means ~3 characters per token on average.
+>
+> *What to seek:* Higher is generally better for efficiency, but extremely high compression may indicate overly aggressive merging that loses morphological information.
+
+**Average Token Length (Fertility)**
+> *Definition:* Mean number of characters per token produced by the tokenizer.
+>
+> *Intuition:* Reflects the granularity of tokenization. Longer tokens capture more context but may struggle with rare words; shorter tokens are more flexible but increase sequence length.
+>
+> *What to seek:* Balance between 2-5 characters for most languages. Arabic/morphologically-rich languages may benefit from slightly longer tokens.
+
+**Unknown Token Rate (OOV Rate)**
+> *Definition:* Percentage of tokens that map to the unknown/UNK token, indicating words the tokenizer cannot represent.
+>
+> *Intuition:* Lower OOV means better vocabulary coverage. High OOV indicates the tokenizer encounters many unseen character sequences.
+>
+> *What to seek:* Below 1% is excellent; below 5% is acceptable. BPE tokenizers typically achieve very low OOV due to subword fallback.
+
+### N-gram Model Metrics
+
+**Perplexity**
+> *Definition:* Measures how "surprised" the model is by test data. Mathematically: 2^(cross-entropy). Lower values indicate better prediction.
+>
+> *Intuition:* If perplexity is 100, the model is as uncertain as if choosing uniformly among 100 options at each step. A perplexity of 10 means effectively choosing among 10 equally likely options.
+>
+> *What to seek:* Lower is better. Perplexity decreases with larger n-grams (more context). Values vary widely by language and corpus size.
+
+**Entropy**
+> *Definition:* Average information content (in bits) needed to encode the next token given the context. Related to perplexity: perplexity = 2^entropy.
+>
+> *Intuition:* High entropy means high uncertainty/randomness; low entropy means predictable patterns. Natural language typically has entropy between 1-4 bits per character.
+>
+> *What to seek:* Lower entropy indicates more predictable text patterns. Entropy should decrease as n-gram size increases.
+
+**Coverage (Top-K)**
+> *Definition:* Percentage of corpus occurrences explained by the top K most frequent n-grams.
+>
+> *Intuition:* High coverage with few patterns indicates repetitive/formulaic text; low coverage suggests diverse vocabulary usage.
+>
+> *What to seek:* Depends on use case. For language modeling, moderate coverage (40-60% with top-1000) is typical for natural text.
+
+### Markov Chain Metrics
+
+**Average Entropy**
+> *Definition:* Mean entropy across all contexts, measuring average uncertainty in next-word prediction.
+>
+> *Intuition:* Lower entropy means the model is more confident about what comes next. Context-1 has high entropy (many possible next words); Context-4 has low entropy (few likely continuations).
+>
+> *What to seek:* Decreasing entropy with larger context sizes. Very low entropy (<0.1) indicates highly deterministic transitions.
+
+**Branching Factor**
+> *Definition:* Average number of unique next tokens observed for each context.
+>
+> *Intuition:* High branching = many possible continuations (flexible but uncertain); low branching = few options (predictable but potentially repetitive).
+>
+> *What to seek:* Branching factor should decrease with context size. Values near 1.0 indicate nearly deterministic chains.
+
+**Predictability**
+> *Definition:* Derived metric: (1 - normalized_entropy) × 100%. Indicates how deterministic the model's predictions are.
+>
+> *Intuition:* 100% predictability means the next word is always certain; 0% means completely random. Real text falls between these extremes.
+>
+> *What to seek:* Higher predictability for text generation quality, but too high (>98%) may produce repetitive output.
+
+### Vocabulary & Zipf's Law Metrics
+
+**Zipf's Coefficient**
+> *Definition:* The slope of the log-log plot of word frequency vs. rank. Zipf's law predicts this should be approximately -1.
+>
+> *Intuition:* A coefficient near -1 indicates the corpus follows natural language patterns where a few words are very common and most words are rare.
+>
+> *What to seek:* Values between -0.8 and -1.2 indicate healthy natural language distribution. Deviations may suggest domain-specific or artificial text.
+
+**R² (Coefficient of Determination)**
+> *Definition:* Measures how well the linear fit explains the frequency-rank relationship. Ranges from 0 to 1.
+>
+> *Intuition:* R² near 1.0 means the data closely follows Zipf's law; lower values indicate deviation from expected word frequency patterns.
+>
+> *What to seek:* R² > 0.95 is excellent; > 0.99 indicates near-perfect Zipf adherence typical of large natural corpora.
+
+**Vocabulary Coverage**
+> *Definition:* Cumulative percentage of corpus tokens accounted for by the top N words.
+>
+> *Intuition:* Shows how concentrated word usage is. If top-100 words cover 50% of text, the corpus relies heavily on common words.
+>
+> *What to seek:* Top-100 covering 30-50% is typical. Higher coverage indicates more repetitive text; lower suggests richer vocabulary.
+
+### Word Embedding Metrics
+
+**Isotropy**
+> *Definition:* Measures how uniformly distributed vectors are in the embedding space. Computed as the ratio of minimum to maximum singular values.
+>
+> *Intuition:* High isotropy (near 1.0) means vectors spread evenly in all directions; low isotropy means vectors cluster in certain directions, reducing expressiveness.
+>
+> *What to seek:* Higher isotropy generally indicates better-quality embeddings. Values > 0.1 are reasonable; > 0.3 is good. Lower-dimensional embeddings tend to have higher isotropy.
+
+**Average Norm**
+> *Definition:* Mean magnitude (L2 norm) of word vectors in the embedding space.
+>
+> *Intuition:* Indicates the typical "length" of vectors. Consistent norms suggest stable training; high variance may indicate some words are undertrained.
+>
+> *What to seek:* Relatively consistent norms across models. The absolute value matters less than consistency (low std deviation).
+
+**Cosine Similarity**
+> *Definition:* Measures angular similarity between vectors, ranging from -1 (opposite) to 1 (identical direction).
+>
+> *Intuition:* Words with similar meanings should have high cosine similarity. This is the standard metric for semantic relatedness in embeddings.
+>
+> *What to seek:* Semantically related words should score > 0.5; unrelated words should be near 0. Synonyms often score > 0.7.
+
+**t-SNE Visualization**
+> *Definition:* t-Distributed Stochastic Neighbor Embedding - a dimensionality reduction technique that preserves local structure for visualization.
+>
+> *Intuition:* Clusters in t-SNE plots indicate groups of semantically related words. Spread indicates vocabulary diversity; tight clusters suggest semantic coherence.
+>
+> *What to seek:* Meaningful clusters (e.g., numbers together, verbs together). Avoid over-interpreting distances - t-SNE preserves local, not global, structure.
+
+### General Interpretation Guidelines
+
+1. **Compare within model families:** Metrics are most meaningful when comparing models of the same type (e.g., 8k vs 64k tokenizer).
+2. **Consider trade-offs:** Better performance on one metric often comes at the cost of another (e.g., compression vs. OOV rate).
+3. **Context matters:** Optimal values depend on downstream tasks. Text generation may prioritize different metrics than classification.
+4. **Corpus influence:** All metrics are influenced by corpus characteristics. Wikipedia text differs from social media or literature.
+5. **Language-specific patterns:** Morphologically rich languages (like Arabic) may show different optimal ranges than analytic languages.
+
+
+### Visualizations Index
+
+| Visualization | Description |
+|---------------|-------------|
+| Tokenizer Compression | Compression ratios by vocabulary size |
+| Tokenizer Fertility | Average token length by vocabulary |
+| Tokenizer OOV | Unknown token rates |
+| Tokenizer Total Tokens | Total tokens by vocabulary |
+| N-gram Perplexity | Perplexity by n-gram size |
+| N-gram Entropy | Entropy by n-gram size |
+| N-gram Coverage | Top pattern coverage |
+| N-gram Unique | Unique n-gram counts |
+| Markov Entropy | Entropy by context size |
+| Markov Branching | Branching factor by context |
+| Markov Contexts | Unique context counts |
+| Zipf's Law | Frequency-rank distribution with fit |
+| Vocab Frequency | Word frequency distribution |
+| Top 20 Words | Most frequent words |
+| Vocab Coverage | Cumulative coverage curve |
+| Embedding Isotropy | Vector space uniformity |
+| Embedding Norms | Vector magnitude distribution |
+| Embedding Similarity | Word similarity heatmap |
+| Nearest Neighbors | Similar words for key terms |
+| t-SNE Words | 2D word embedding visualization |
+| t-SNE Sentences | 2D sentence embedding visualization |
+| Position Encoding | Encoding method comparison |
+| Model Sizes | Storage requirements |
+| Performance Dashboard | Comprehensive performance overview |
+
+---
+## About This Project
+
+### Data Source
+
+Models trained on [wikipedia-monthly](https://huggingface.co/datasets/omarkamali/wikipedia-monthly) - a monthly snapshot of Wikipedia articles across 300+ languages.
+
+### Project
+
+A project by **[Wikilangs](https://wikilangs.org)** - Open-source NLP models for every Wikipedia language.
+
+### Maintainer
+
+[Omar Kamali](https://omarkamali.com) - [Omneity Labs](https://omneitylabs.com)
+
+### Citation
+
+If you use these models in your research, please cite:
+
+```bibtex
+@misc{wikilangs2025,
+  author = {Kamali, Omar},
+  title = {Wikilangs: Open NLP Models for Wikipedia Languages},
+  year = {2025},
+  publisher = {HuggingFace},
+  url = {https://huggingface.co/wikilangs}
+  institution = {Omneity Labs}
+}
+```
+
+### License
+
+MIT License - Free for academic and commercial use.
+
+### Links
+
+- 🌐 Website: [wikilangs.org](https://wikilangs.org)
+- 🤗 Models: [huggingface.co/wikilangs](https://huggingface.co/wikilangs)
+- 📊 Data: [wikipedia-monthly](https://huggingface.co/datasets/omarkamali/wikipedia-monthly)
+- 👤 Author: [Omar Kamali](https://huggingface.co/omarkamali)
+---
+*Generated by Wikilangs Models Pipeline*
+
+*Report Date: 2025-12-27 06:08:15*

models/embeddings/monolingual/anp_128d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e5037926f606a5261779717c539c9dbbd1ae5b5239b116508b41fe3616febae
+size 1036770452

models/embeddings/monolingual/anp_128d.meta.json

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

models/embeddings/monolingual/anp_128d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "anp",
+  "dimension": 128,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 128,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 12167
+}

models/embeddings/monolingual/anp_32d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd35312939967d31d7d904a9ca1a76cbbdcc4c44fad15183039efe9bf32cbdd2
+size 259426196

models/embeddings/monolingual/anp_32d.meta.json

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

models/embeddings/monolingual/anp_32d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "anp",
+  "dimension": 32,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 32,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 12167
+}

models/embeddings/monolingual/anp_64d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95dc1f4810dddc464023dcea010decc1b484ac058bc2a8c9392076ea90c65544
+size 518540948

models/embeddings/monolingual/anp_64d.meta.json

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

models/embeddings/monolingual/anp_64d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "anp",
+  "dimension": 64,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 64,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 12167
+}

models/subword_markov/anp_markov_ctx1_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4297aac5947b461b82ded3981b487a038458065f66a84607b9d1d55bb71f1718
+size 83720

models/subword_markov/anp_markov_ctx1_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "subword",
+  "language": "anp",
+  "unique_contexts": 755,
+  "total_transitions": 4171640
+}

models/subword_markov/anp_markov_ctx2_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0e556496b318b18ecfe9c97c36c2d23a696de1824af13d8e624e9d9e2c53bad
+size 561964

models/subword_markov/anp_markov_ctx2_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "subword",
+  "language": "anp",
+  "unique_contexts": 10432,
+  "total_transitions": 4168473
+}

models/subword_markov/anp_markov_ctx3_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c16a8be91a6fe7953eef4ac6eccf82dca4b99b8816d61fea6fe3718b7600bb34
+size 2057035

models/subword_markov/anp_markov_ctx3_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "subword",
+  "language": "anp",
+  "unique_contexts": 76706,
+  "total_transitions": 4165306
+}

models/subword_markov/anp_markov_ctx4_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e54d19d4e7dafad81e1f50c98d138e50a189da919f32e24193b667682fa96a8
+size 5952951

models/subword_markov/anp_markov_ctx4_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "subword",
+  "language": "anp",
+  "unique_contexts": 290862,
+  "total_transitions": 4162139
+}

models/subword_ngram/anp_2gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:355f1e9ea12646e5552194569897d90f61b54d55a6596d37ad576bac53b4c18a
+size 66307

models/subword_ngram/anp_2gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "subword",
+  "language": "anp",
+  "unique_ngrams": 5088,
+  "total_ngrams": 4171640
+}

models/subword_ngram/anp_3gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b14727440da7b6e77cb354d37b0c27476ed4374fb0258851694ef1e8116220f5
+size 516525

models/subword_ngram/anp_3gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "subword",
+  "language": "anp",
+  "unique_ngrams": 39446,
+  "total_ngrams": 4168473
+}

models/subword_ngram/anp_4gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e2aa49d6568ec1b8bf85049449b1f7efb6e4ddd46688f5f01f50be945751e4cd
+size 1939235

models/subword_ngram/anp_4gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "subword",
+  "language": "anp",
+  "unique_ngrams": 155528,
+  "total_ngrams": 4165306
+}

models/tokenizer/anp_tokenizer_16k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efb5cb1039a55c7f7bb857ffb86e33f81c6dc4d0c89b00b0b2bf254abbf18eaa
+size 617718

models/tokenizer/anp_tokenizer_32k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b0c4178d49ec2b0885bec0eaad33f6a5f7167c07accfe726ef9bfc8acb39865
+size 1020495

models/tokenizer/anp_tokenizer_64k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:688f4ca6efb65501b43b539b7fd3cb2a1fd1b2ecc0be1dc318fb516b5c44a990
+size 1853030

models/tokenizer/anp_tokenizer_8k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:326640bb83ed996d173f81797fc5a30047a8b1b52c4a054dda0f50dacd58e6d7
+size 424492

models/vocabulary/anp_vocabulary.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7846de396d06f632490fe25bb62f459573bd28942c91c218251658557d166439
+size 154433

models/vocabulary/anp_vocabulary_metadata.json

@@ -0,0 +1,16 @@
+{
+  "language": "anp",
+  "vocabulary_size": 9567,
+  "statistics": {
+    "type_token_ratio": 0.014131336612329602,
+    "coverage": {
+      "top_100": 0.8221627535895281,
+      "top_1000": 0.954931271936766,
+      "top_5000": 0.9849848553672154,
+      "top_10000": 0.992719888352438
+    },
+    "hapax_count": 11059,
+    "hapax_ratio": 0.5361679433724426,
+    "total_documents": 3167
+  }
+}

models/word_markov/anp_markov_ctx1_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:192ca60a300173cfabd8b1fb3660b244c3be732254c68e60bf1ee02dd07b598e
+size 886162

models/word_markov/anp_markov_ctx1_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "word",
+  "language": "anp",
+  "unique_contexts": 20822,
+  "total_transitions": 2733532
+}

models/word_markov/anp_markov_ctx2_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a4a530af52e6c9f3eaab3f9243da477a38d9854b21682a25de5c7b9f6c8f144
+size 2833179

models/word_markov/anp_markov_ctx2_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "word",
+  "language": "anp",
+  "unique_contexts": 125337,
+  "total_transitions": 2730365
+}

models/word_markov/anp_markov_ctx3_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7e40a5b87677577d689aa253f08740d9ad689107604462b8e1e4df9ad5a7cba
+size 6746500

models/word_markov/anp_markov_ctx3_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "word",
+  "language": "anp",
+  "unique_contexts": 335781,
+  "total_transitions": 2727198
+}

models/word_markov/anp_markov_ctx4_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df1d380e4e3dbf318266422fddd74c8e3c3c5bbb43bcf44df36feb99255fc67d
+size 12362534

models/word_markov/anp_markov_ctx4_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "word",
+  "language": "anp",
+  "unique_contexts": 678169,
+  "total_transitions": 2724031
+}

models/word_ngram/anp_2gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:380fd982ea2b330e8e2918f8c25bfd63e9d5582ff2d103f96917453a9eec5bcd
+size 261069

models/word_ngram/anp_2gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "word",
+  "language": "anp",
+  "unique_ngrams": 19190,
+  "total_ngrams": 2733532
+}

models/word_ngram/anp_3gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f8e4b8e3220800bfc9f062abf2dce74ef08616fa574c66f309c8d9d202644d1
+size 955234

models/word_ngram/anp_3gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "word",
+  "language": "anp",
+  "unique_ngrams": 71690,
+  "total_ngrams": 2730365
+}

models/word_ngram/anp_4gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9711dcecd0f23b0bdaba6380c42f3dd115a1ded144cc810366c614694e28520a
+size 2713081

models/word_ngram/anp_4gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "word",
+  "language": "anp",
+  "unique_ngrams": 200757,
+  "total_ngrams": 2727198
+}