Upload all models and assets for crh (20251201)

.gitattributes

@@ -33,3 +33,9 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+visualizations/embedding_similarity.png filter=lfs diff=lfs merge=lfs -text
+visualizations/performance_dashboard.png filter=lfs diff=lfs merge=lfs -text
+visualizations/position_encoding_comparison.png filter=lfs diff=lfs merge=lfs -text
+visualizations/tsne_sentences.png filter=lfs diff=lfs merge=lfs -text
+visualizations/tsne_words.png filter=lfs diff=lfs merge=lfs -text
+visualizations/zipf_law.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,554 @@
+---
+language: crh
+language_name: CRH
+language_family: turkic_kipchak
+tags:
+  - wikilangs
+  - nlp
+  - tokenizer
+  - embeddings
+  - n-gram
+  - markov
+  - wikipedia
+  - monolingual
+  - family-turkic_kipchak
+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: 4.462
+  - name: best_isotropy
+    type: isotropy
+    value: 0.7580
+  - name: vocabulary_size
+    type: vocab
+    value: 53689
+generated: 2025-12-28
+---
+
+# CRH - Wikilangs Models
+## Comprehensive Research Report & Full Ablation Study
+
+This repository contains NLP models trained and evaluated by Wikilangs, specifically on **CRH** 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** | 3.468x | 3.42 | 0.2080% | 235,121 |
+| **16k** | 3.842x | 3.78 | 0.2305% | 212,188 |
+| **32k** | 4.188x | 4.13 | 0.2512% | 194,672 |
+| **64k** | 4.462x 🏆 | 4.39 | 0.2676% | 182,737 |
+
+### Tokenization Examples
+
+Below are sample sentences tokenized with each vocabulary size:
+
+**Sample 1:** `Novotroyevka () - Rusiyeniñ Belgorod vilâyetinde Koroça rayonında bir köy. Ealis...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁novo tr oy evka ▁() ▁- ▁rusiyeniñ ▁belgorod ▁vilâyetinde ▁koroça ... (+17 more)` | 27 |
+| 16k | `▁novotr oy evka ▁() ▁- ▁rusiyeniñ ▁belgorod ▁vilâyetinde ▁koroça ▁rayonında ... (+16 more)` | 26 |
+| 32k | `▁novotroy evka ▁() ▁- ▁rusiyeniñ ▁belgorod ▁vilâyetinde ▁koroça ▁rayonında ▁bir ... (+15 more)` | 25 |
+| 64k | `▁novotroy evka ▁() ▁- ▁rusiyeniñ ▁belgorod ▁vilâyetinde ▁koroça ▁rayonında ▁bir ... (+15 more)` | 25 |
+
+**Sample 2:** `Holodna Balka () - Ukrainanıñ Ades vilâyetinde Ades rayonında bir köy. Ealisiniñ...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁hol od na ▁balka ▁() ▁- ▁ukrainanıñ ▁ades ▁vilâyetinde ▁ades ... (+18 more)` | 28 |
+| 16k | `▁hol od na ▁balka ▁() ▁- ▁ukrainanıñ ▁ades ▁vilâyetinde ▁ades ... (+18 more)` | 28 |
+| 32k | `▁hol odna ▁balka ▁() ▁- ▁ukrainanıñ ▁ades ▁vilâyetinde ▁ades ▁rayonında ... (+17 more)` | 27 |
+| 64k | `▁holodna ▁balka ▁() ▁- ▁ukrainanıñ ▁ades ▁vilâyetinde ▁ades ▁rayonında ▁bir ... (+16 more)` | 26 |
+
+**Sample 3:** `Tereşpil () - Ukrainanıñ Vinnıtsâ vilâyetinde Hmilnık rayonında bir köy. Ealisin...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁ter eş pil ▁() ▁- ▁ukrainanıñ ▁vinnıtsâ ▁vilâyetinde ▁hmilnık ▁rayonında ... (+16 more)` | 26 |
+| 16k | `▁tereş pil ▁() ▁- ▁ukrainanıñ ▁vinnıtsâ ▁vilâyetinde ▁hmilnık ▁rayonında ▁bir ... (+15 more)` | 25 |
+| 32k | `▁tereş pil ▁() ▁- ▁ukrainanıñ ▁vinnıtsâ ▁vilâyetinde ▁hmilnık ▁rayonında ▁bir ... (+15 more)` | 25 |
+| 64k | `▁tereşpil ▁() ▁- ▁ukrainanıñ ▁vinnıtsâ ▁vilâyetinde ▁hmilnık ▁rayonında ▁bir ▁köy ... (+14 more)` | 24 |
+
+
+### Key Findings
+
+- **Best Compression:** 64k achieves 4.462x compression
+- **Lowest UNK Rate:** 8k with 0.2080% 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** | 1,152 🏆 | 10.17 | 18,186 | 53.8% | 71.2% |
+| **2-gram** | 405 🏆 | 8.66 | 4,814 | 59.9% | 97.2% |
+| **3-gram** | 1,983 | 10.95 | 27,457 | 46.4% | 65.5% |
+| **3-gram** | 2,478 | 11.28 | 35,169 | 32.1% | 69.8% |
+| **4-gram** | 4,721 | 12.21 | 55,382 | 36.7% | 54.6% |
+| **4-gram** | 8,182 | 13.00 | 157,163 | 26.4% | 52.7% |
+
+### Top 5 N-grams by Size
+
+**2-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `kategoriya :` | 32,150 |
+| 2 | `( )` | 23,218 |
+| 3 | `) -` | 21,379 |
+| 4 | `ealisiniñ sayısı` | 20,740 |
+| 5 | `. ealisiniñ` | 20,734 |
+
+**3-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `. ealisiniñ sayısı` | 20,734 |
+| 2 | `( ) -` | 19,732 |
+| 3 | `. kategoriya :` | 16,456 |
+| 4 | `kişi . kategoriya` | 14,755 |
+| 5 | `bir köy .` | 10,054 |
+
+**4-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `kişi . kategoriya :` | 14,755 |
+| 2 | `rayonında bir köy .` | 9,313 |
+| 3 | `( ) - rusiyeniñ` | 9,196 |
+| 4 | `bir köy . ealisiniñ` | 9,139 |
+| 5 | `köy . ealisiniñ sayısı` | 9,139 |
+
+
+### Key Findings
+
+- **Best Perplexity:** 2-gram with 405
+- **Entropy Trend:** Decreases with larger n-grams (more predictable)
+- **Coverage:** Top-1000 patterns cover ~53% 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** | 0.5665 | 1.481 | 3.09 | 136,749 | 43.3% |
+| **1** | 1.1866 | 2.276 | 9.27 | 1,393 | 0.0% |
+| **2** | 0.1665 | 1.122 | 1.39 | 422,458 | 83.3% |
+| **2** | 0.9666 | 1.954 | 5.70 | 12,902 | 3.3% |
+| **3** | 0.0665 | 1.047 | 1.14 | 585,200 | 93.4% |
+| **3** | 0.8194 | 1.765 | 3.81 | 73,541 | 18.1% |
+| **4** | 0.0328 🏆 | 1.023 | 1.07 | 663,395 | 96.7% |
+| **4** | 0.5748 🏆 | 1.489 | 2.40 | 280,480 | 42.5% |
+
+### Generated Text Samples
+
+Below are text samples generated from each Markov chain model:
+
+**Context Size 1:**
+
+1. `. menbalar türkiye i ̇ zmalkovo krasnoye rayonında bir qasaba . ealisiniñ sayısı 27 , başqırtistan`
+2. `- rusiyeniñ belgorod vilâyetinde podilsk rayonı ( , cin æmæ jyn kad ! bu ağnıñ esas`
+3. `, latin elifbesiniñ 19 ( ) , cıyınlarda cıyınlarnıñ cıyıntıgı cıyıntıq alına kelgen medinet es herma...`
+
+**Context Size 2:**
+
+1. `kategoriya : troitskoye rayonındaki meskün yerler kategoriya : primorye ülkesindeki meskün yerler ka...`
+2. `( ) - rusiyeniñ altay ülkesinde şelaboliha rayonında bir köy . ealisiniñ sayısı 47 kişi . kategoriya`
+3. `) - rusiyede , başqırtistan cumhuriyetiniñ miyeke rayonında bir hutor . ealisiniñ sayısı 177 kişi . ...`
+
+**Context Size 3:**
+
+1. `. ealisiniñ sayısı 485 kişi . kategoriya : herson vilâyeti`
+2. `( ) - rusiyeniñ brânsk vilâyetinde karaçev rayonında bir köy . ealisiniñ sayısı 423 kişi . kategoriy...`
+3. `. kategoriya : başqırtistandaki meskün yerler`
+
+**Context Size 4:**
+
+1. `kişi . kategoriya : ades vilâyetindeki köyler`
+2. `rayonında bir köy . ealisiniñ sayısı 448 kişi . i ̇ htar kategoriya : tahtamukay rayonındaki meskün ...`
+3. `( ) - rusiyeniñ amur vilâyetinde şimanovsk rayonında bir köy . ealisiniñ sayısı 654 kişi . kategoriy...`
+
+
+### Key Findings
+
+- **Best Predictability:** Context-4 with 96.7% predictability
+- **Branching Factor:** Decreases with context size (more deterministic)
+- **Memory Trade-off:** Larger contexts require more storage (280,480 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 | 53,689 |
+| Total Tokens | 889,654 |
+| Mean Frequency | 16.57 |
+| Median Frequency | 3 |
+| Frequency Std Dev | 308.97 |
+
+### Most Common Words
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | kategoriya | 32,152 |
+| 2 | bir | 27,919 |
+| 3 | kişi | 20,861 |
+| 4 | sayısı | 20,822 |
+| 5 | ealisiniñ | 20,770 |
+| 6 | rayonında | 17,392 |
+| 7 | i | 13,962 |
+| 8 | meskün | 13,507 |
+| 9 | yerler | 12,928 |
+| 10 | vilâyetinde | 12,440 |
+
+### Least Common Words (from vocabulary)
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | зияде | 2 |
+| 2 | atalarnıñ | 2 |
+| 3 | kotsubınskıylar | 2 |
+| 4 | yüneskonıñ | 2 |
+| 5 | اورمودا | 2 |
+| 6 | دیللر | 2 |
+| 7 | ازبری | 2 |
+| 8 | اولان | 2 |
+| 9 | قیزی | 2 |
+| 10 | samançı | 2 |
+
+### Zipf's Law Analysis
+
+| Metric | Value |
+|--------|-------|
+| Zipf Coefficient | 1.0203 |
+| R² (Goodness of Fit) | 0.996904 |
+| Adherence Quality | **excellent** |
+
+### Coverage Analysis
+
+| Top N Words | Coverage |
+|-------------|----------|
+| Top 100 | 46.7% |
+| Top 1,000 | 65.0% |
+| Top 5,000 | 79.6% |
+| Top 10,000 | 85.3% |
+
+### Key Findings
+
+- **Zipf Compliance:** R²=0.9969 indicates excellent adherence to Zipf's law
+- **High Frequency Dominance:** Top 100 words cover 46.7% of corpus
+- **Long Tail:** 43,689 words needed for remaining 14.7% 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** | 16,090 | 32 | 4.513 | 0.777 | 0.7580 🏆 |
+| **mono_64d** | 16,090 | 64 | 4.759 | 0.736 | 0.5447 |
+| **mono_128d** | 16,090 | 128 | 4.802 | 0.733 | 0.1564 |
+| **embeddings_enhanced** | 0 | 0 | 0.000 | 0.000 | 0.0000 |
+
+### Key Findings
+
+- **Best Isotropy:** mono_32d with 0.7580 (more uniform distribution)
+- **Dimension Trade-off:** Higher dimensions capture more semantics but reduce isotropy
+- **Vocabulary Coverage:** All models cover 16,090 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 (4.46x) with low UNK rate |
+| N-gram | **5-gram** | Lowest perplexity (405) |
+| Markov | **Context-4** | Highest predictability (96.7%) |
+| 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 23:15:40*

models/embeddings/monolingual/crh_128d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2711de686fbbdccec9b4c232aeb1a2ff6589a0a5ab35d03b8eed6ce02bd7f456
+size 1040761703

models/embeddings/monolingual/crh_128d.meta.json

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

models/embeddings/monolingual/crh_128d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "crh",
+  "dimension": 128,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 128,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 16090
+}

models/embeddings/monolingual/crh_32d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66a6e8733bef4321f51d9516ccf447352157f4a62af7a17ee0b3077db4528dcf
+size 260404583

models/embeddings/monolingual/crh_32d.meta.json

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

models/embeddings/monolingual/crh_32d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "crh",
+  "dimension": 32,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 32,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 16090
+}

models/embeddings/monolingual/crh_64d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d958dcba834b7704a2fa91467c099cd2ee01f0d1d5fed91d7a38303939710c0
+size 520523623

models/embeddings/monolingual/crh_64d.meta.json

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

models/embeddings/monolingual/crh_64d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "crh",
+  "dimension": 64,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 64,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 16090
+}

models/subword_markov/crh_markov_ctx1_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9de08c51e7f35ddea586397377dd1a690a335be3deafe5034997cfcfdc8fb563
+size 98281

models/subword_markov/crh_markov_ctx1_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "subword",
+  "language": "crh",
+  "unique_contexts": 1393,
+  "total_transitions": 7898659
+}

models/subword_markov/crh_markov_ctx2_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9bb8639872bba7668e3be3688a341970c45c509563b7d7eff3fddaa1b857968
+size 581277

models/subword_markov/crh_markov_ctx2_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "subword",
+  "language": "crh",
+  "unique_contexts": 12902,
+  "total_transitions": 7868965
+}

models/subword_markov/crh_markov_ctx3_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ab59e75e6595f11a4437338aa658f32767c561497c79e161cf3d4148cdd2c79
+size 2047334

models/subword_markov/crh_markov_ctx3_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "subword",
+  "language": "crh",
+  "unique_contexts": 73541,
+  "total_transitions": 7839271
+}

models/subword_markov/crh_markov_ctx4_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4fcb0e55ae35489db7ded118cd7a233f52f8dd657681364cb8e3372187263e04
+size 5676695

models/subword_markov/crh_markov_ctx4_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "subword",
+  "language": "crh",
+  "unique_contexts": 280480,
+  "total_transitions": 7809577
+}

models/subword_ngram/crh_2gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1d24c9778380ff73f830ff74e1323c5c9a67b1ad255aca792619d77b25d5912
+size 64028

models/subword_ngram/crh_2gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "subword",
+  "language": "crh",
+  "unique_ngrams": 4814,
+  "total_ngrams": 7898659
+}

models/subword_ngram/crh_3gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4fd9693cd6180b77894216a45c8557781e0405949f934ac0db671694231e38f
+size 451805

models/subword_ngram/crh_3gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "subword",
+  "language": "crh",
+  "unique_ngrams": 35169,
+  "total_ngrams": 7868965
+}

models/subword_ngram/crh_4gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45d6866f9f3500b9daf768646045ee8c250bd213b574f0a8fce2c8ef66523e4e
+size 1836873

models/subword_ngram/crh_4gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "subword",
+  "language": "crh",
+  "unique_ngrams": 157163,
+  "total_ngrams": 7839271
+}

models/tokenizer/crh_tokenizer_16k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8bf313aa3863643a3d3a48927e34d71e36c34b5c866db4ec55ff481f2c42cf78
+size 511102

models/tokenizer/crh_tokenizer_32k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a90dac68538f5086942dbcf27b4787b28e5927dcb62fb4edbfd0d3f645b170cd
+size 798788

models/tokenizer/crh_tokenizer_64k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ca674ae3850ca528644e4e9d7d41feaaf7f45a78b69b9eccd17204ad57d458a
+size 1388807

models/tokenizer/crh_tokenizer_8k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd97a27e642560427bbf6bc2e1ec3da514ec6c83ec093a16660de476f4797974
+size 373198

models/vocabulary/crh_vocabulary.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37995711df1e3c6e65e7b604e037154a533bfd0c17bbed09013a0b12d6d0cc3a
+size 925242

models/vocabulary/crh_vocabulary_metadata.json

@@ -0,0 +1,16 @@
+{
+  "language": "crh",
+  "vocabulary_size": 53689,
+  "statistics": {
+    "type_token_ratio": 0.14047543157959674,
+    "coverage": {
+      "top_100": 0.42677284364428997,
+      "top_1000": 0.594827213933929,
+      "top_5000": 0.727782518841444,
+      "top_10000": 0.7806146474876361
+    },
+    "hapax_count": 82936,
+    "hapax_ratio": 0.6070338517840805,
+    "total_documents": 29694
+  }
+}

models/word_markov/crh_markov_ctx1_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35f93040f44b20143ca6995aa27d0e51be65a8aed5ec256feed368bc396e7b39
+size 4448665

models/word_markov/crh_markov_ctx1_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "word",
+  "language": "crh",
+  "unique_contexts": 136749,
+  "total_transitions": 1289772
+}

models/word_markov/crh_markov_ctx2_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1186f4b8afeb522aea0d59da82796161e681cf8812db7eafe8dce2d594b4cff3
+size 8722410

models/word_markov/crh_markov_ctx2_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "word",
+  "language": "crh",
+  "unique_contexts": 422458,
+  "total_transitions": 1260079
+}

models/word_markov/crh_markov_ctx3_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70418ae3723bd86e8fa05bb025daf404c3ba9978b2435ea847cff6c13a4929e8
+size 11645146

models/word_markov/crh_markov_ctx3_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "word",
+  "language": "crh",
+  "unique_contexts": 585200,
+  "total_transitions": 1230387
+}

models/word_markov/crh_markov_ctx4_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca323a9f7ec0e2f8d1a95747ea890278db7668783f6c23a6e5fb60385dc416b1
+size 13583842

models/word_markov/crh_markov_ctx4_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "word",
+  "language": "crh",
+  "unique_contexts": 663395,
+  "total_transitions": 1200699
+}

models/word_ngram/crh_2gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d99c79cd54a7e225c135d72d01faebb42fc513cea6d991424a374f1c1f37d7b3
+size 271118

models/word_ngram/crh_2gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "word",
+  "language": "crh",
+  "unique_ngrams": 18186,
+  "total_ngrams": 1289772
+}

models/word_ngram/crh_3gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5273683d72465d19ccff790d2d7b74fa2a21bc70f48fb303660846dc838afdf4
+size 450204

models/word_ngram/crh_3gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "word",
+  "language": "crh",
+  "unique_ngrams": 27457,
+  "total_ngrams": 1260079
+}

models/word_ngram/crh_4gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dfb14a5090afc83bba2389cd24b93178e9858ededac8c4e8505df8fb240e9d5f
+size 905910

models/word_ngram/crh_4gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "word",
+  "language": "crh",
+  "unique_ngrams": 55382,
+  "total_ngrams": 1230387
+}