Upload all models and assets for ckb (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 +1,574 @@
+---
+language: ckb
+language_name: CKB
+language_family: iranian_western
+tags:
+  - wikilangs
+  - nlp
+  - tokenizer
+  - embeddings
+  - n-gram
+  - markov
+  - wikipedia
+  - monolingual
+  - family-iranian_western
+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: 4.743
+  - name: best_isotropy
+    type: isotropy
+    value: 0.7972
+  - name: vocabulary_size
+    type: vocab
+    value: 267929
+generated: 2025-12-28
+---
+
+# CKB - Wikilangs Models
+## Comprehensive Research Report & Full Ablation Study
+
+This repository contains NLP models trained and evaluated by Wikilangs, specifically on **CKB** 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** | 3.683x | 3.65 | 0.0660% | 975,730 |
+| **16k** | 4.093x | 4.06 | 0.0733% | 878,011 |
+| **32k** | 4.448x | 4.41 | 0.0797% | 808,075 |
+| **64k** | 4.743x 🏆 | 4.70 | 0.0850% | 757,838 |
+
+### Tokenization Examples
+
+Below are sample sentences tokenized with each vocabulary size:
+
+**Sample 1:** `شارێکی ویلایەتی جۆرجیایە لە ویلایەتە یەکگرتووەکانی ئەمریکا.
+
+ بەستەرە دەرکییەکان`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁شارێکی ▁ویلایەتی ▁جۆر جی ایە ▁لە ▁ویلایەتە ▁یەکگرتووەکانی ▁ئەمریکا . ... (+4 more)` | 14 |
+| 16k | `▁شارێکی ▁ویلایەتی ▁جۆرجی ایە ▁لە ▁ویلایەتە ▁یەکگرتووەکانی ▁ئەمریکا . ▁بەستەرە ... (+2 more)` | 12 |
+| 32k | `▁شارێکی ▁ویلایەتی ▁جۆرجی ایە ▁لە ▁ویلایەتە ▁یەکگرتووەکانی ▁ئەمریکا . ▁بەستەرە ... (+2 more)` | 12 |
+| 64k | `▁شارێکی ▁ویلایەتی ▁جۆرجی ایە ▁لە ▁ویلایەتە ▁یەکگرتووەکانی ▁ئەمریکا . ▁بەستەرە ... (+2 more)` | 12 |
+
+**Sample 2:** `ڕووداوەکان 
+
+ لەدایکبوونەکان 
+
+ مردنەکان 
+
+ سەرچاوەکان 
+
+
+پۆل:ساڵەکان`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁ڕووداوەکان ▁لەدایکبوونەکان ▁مردنەکان ▁سەرچاوەکان ▁پۆل : ساڵەکان` | 7 |
+| 16k | `▁ڕووداوەکان ▁لەدایکبوونەکان ▁مردنەکان ▁سەرچاوەکان ▁پۆل : ساڵەکان` | 7 |
+| 32k | `▁ڕووداوەکان ▁لەدایکبوونەکان ▁مردنەکان ▁سەرچاوەکان ▁پۆل : ساڵەکان` | 7 |
+| 64k | `▁ڕووداوەکان ▁لەدایکبوونەکان ▁مردنەکان ▁سەرچاوەکان ▁پۆل : ساڵەکان` | 7 |
+
+**Sample 3:** `ڕووداوەکان 
+
+ لەدایکبوونەکان 
+
+ مردنەکان 
+
+ سەرچاوەکان 
+
+
+پۆل:ساڵەکان`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁ڕووداوەکان ▁لەدایکبوونەکان ▁مردنەکان ▁سەرچاوەکان ▁پۆل : ساڵەکان` | 7 |
+| 16k | `▁ڕووداوەکان ▁لەدایکبوونەکان ▁مردنەکان ▁سەرچاوەکان ▁پۆل : ساڵەکان` | 7 |
+| 32k | `▁ڕووداوەکان ▁لەدایکبوونەکان ▁مردنەکان ▁سەرچاوەکان ▁پۆل : ساڵەکان` | 7 |
+| 64k | `▁ڕووداوەکان ▁لەدایکبوونەکان ▁مردنەکان ▁سەرچاوەکان ▁پۆل : ساڵەکان` | 7 |
+
+
+### Key Findings
+
+- **Best Compression:** 64k achieves 4.743x compression
+- **Lowest UNK Rate:** 8k with 0.0660% 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** | 35,123 🏆 | 15.10 | 297,076 | 15.8% | 33.2% |
+| **2-gram** | 381 🏆 | 8.57 | 15,002 | 62.1% | 96.1% |
+| **3-gram** | 68,724 | 16.07 | 466,780 | 12.3% | 28.4% |
+| **3-gram** | 3,034 | 11.57 | 125,615 | 27.5% | 67.3% |
+| **4-gram** | 110,939 | 16.76 | 754,916 | 10.4% | 25.7% |
+| **4-gram** | 15,538 | 13.92 | 641,521 | 13.6% | 40.6% |
+
+### Top 5 N-grams by Size
+
+**2-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `پۆل :` | 315,394 |
+| 2 | `. لە` | 57,144 |
+| 3 | `لە ساڵی` | 47,235 |
+| 4 | `. سەرچاوەکان` | 43,826 |
+| 5 | `سەرچاوەکان پۆل` | 31,612 |
+
+**3-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `سەرچاوەکان پۆل :` | 31,612 |
+| 2 | `: / /` | 24,968 |
+| 3 | `پۆل : ئەکتەرەکانی` | 22,109 |
+| 4 | `پۆل : لەدایکبووانی` | 22,019 |
+| 5 | `ئەمریکییەکان پۆل :` | 18,941 |
+
+**4-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `. سەرچاوەکان پۆل :` | 17,589 |
+| 2 | `بەستەرە دەرەکییەکان پۆل :` | 16,063 |
+| 3 | `سەرچاوەکان بەستەرە دەرەکییەکان پۆل` | 15,316 |
+| 4 | `. سەرچاوەکان بەستەرە دەرەکییەکان` | 14,823 |
+| 5 | `http : / /` | 12,552 |
+
+
+### Key Findings
+
+- **Best Perplexity:** 2-gram with 381
+- **Entropy Trend:** Decreases with larger n-grams (more predictable)
+- **Coverage:** Top-1000 patterns cover ~41% 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** | 0.7236 | 1.651 | 6.62 | 684,303 | 27.6% |
+| **1** | 1.3292 | 2.513 | 9.41 | 4,993 | 0.0% |
+| **2** | 0.3062 | 1.236 | 1.94 | 4,525,451 | 69.4% |
+| **2** | 0.8890 | 1.852 | 6.06 | 46,978 | 11.1% |
+| **3** | 0.1145 | 1.083 | 1.24 | 8,778,035 | 88.5% |
+| **3** | 0.8191 | 1.764 | 4.27 | 284,820 | 18.1% |
+| **4** | 0.0429 🏆 | 1.030 | 1.08 | 10,844,509 | 95.7% |
+| **4** | 0.6245 🏆 | 1.542 | 2.80 | 1,215,545 | 37.5% |
+
+### Generated Text Samples
+
+Below are text samples generated from each Markov chain model:
+
+**Context Size 1:**
+
+1. `لە ١٩٩٤ وەک کانگ ھون ( ١٧٣ ڕۆژ بوو لە پارتی دیموکراتی لەو ئامێرانە دەبنەهۆی کەمکردنەوەی`
+2. `. کۆمارەکە لە ئاستی عێراق بکات وا بیر و ناودارەکان و پەرە بە گراڤ ئەوە ،`
+3. `، سوپەر فلودی ھیلێۆم وایە پێویستە بۆ ئەو بڕە داھاتێکی باشیش بوو لە تاراوگە زۆرەملێکان لە`
+
+**Context Size 2:**
+
+1. `پۆل : فیلمە بەرھەمھێنراوەکان لەلایەن بیو فلین پۆل : فیلمە ئەنیمەیشنەکانی سۆنی پیکچەرز ، ڤۆگێل لە ساڵ...`
+2. `. لە ڕێکەوتی ١٠ی کانوونی دووەمی ١٩١٩ ، بادە لە ساڵی ١٩٣٧ وەک « ئەگەر ناچار بێت`
+3. `لە ساڵی ١٩٨١ پەنای بردووەتە ویلایەتە یەکگرتووەکانی ئەمریکا . سەنتەرەکە نزیکەی ٢٥٠ کارمەندی تێدابوو ک...`
+
+**Context Size 3:**
+
+1. `سەرچاوەکان پۆل : خۆراک و ژینگە پۆل : پێشەکییەکانی ساڵی ١٩٧٢ پۆل : ئیسرائیل لە ١٩٤٨ پۆل :`
+2. `: / / kurdipedia . org / web / 20090302175610 / http : / / web . archive`
+3. `پۆل : ئەکتەرەکانی تەلەڤیزیۆنی پیاوی ئەمریکی پۆل : ئەو فیلمانەی لە نیویۆرک داندراون پۆل : فیلمە کەنەد...`
+
+**Context Size 4:**
+
+1. `. سەرچاوەکان پۆل : شوێنە ئاوەدانەکانی پارێزگای سلێمانی پۆل : گوندەکانی باشووری کوردستان پۆل : نیشتەج...`
+2. `بەستەرە دەرەکییەکان پۆل : لەدایکبووانی ١٩١٧ پۆل : مردووانی ١٩٩٧ پۆل : ئەکتەرە پیاوە ئەمریکییەکانی سە...`
+3. `سەرچاوەکان بەستەرە دەرەکییەکان پۆل : لەدایکبووانی ١٩٥٧ پۆل : نووسەرە پیاوە ئەمریکییەکانی سەدەی ٢٠ەم ...`
+
+
+### Key Findings
+
+- **Best Predictability:** Context-4 with 95.7% predictability
+- **Branching Factor:** Decreases with context size (more deterministic)
+- **Memory Trade-off:** Larger contexts require more storage (1,215,545 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 | 267,929 |
+| Total Tokens | 12,273,016 |
+| Mean Frequency | 45.81 |
+| Median Frequency | 4 |
+| Frequency Std Dev | 1797.49 |
+
+### Most Common Words
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | لە | 634,045 |
+| 2 | و | 445,997 |
+| 3 | پۆل | 315,758 |
+| 4 | بە | 217,465 |
+| 5 | کە | 180,742 |
+| 6 | بۆ | 132,391 |
+| 7 | ساڵی | 84,816 |
+| 8 | سەرچاوەکان | 65,009 |
+| 9 | بوو | 61,528 |
+| 10 | لەگەڵ | 54,424 |
+
+### Least Common Words (from vocabulary)
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | microarchitecture | 2 |
+| 2 | gigabit | 2 |
+| 3 | سوپەرکۆمپیوتەرەکە | 2 |
+| 4 | تایوانیا | 2 |
+| 5 | بایۆمۆلیکولەر | 2 |
+| 6 | gimps | 2 |
+| 7 | principatele | 2 |
+| 8 | دۆمنیتۆر | 2 |
+| 9 | باربو | 2 |
+| 10 | کاتارجیو | 2 |
+
+### Zipf's Law Analysis
+
+| Metric | Value |
+|--------|-------|
+| Zipf Coefficient | 1.0351 |
+| R² (Goodness of Fit) | 0.990570 |
+| Adherence Quality | **excellent** |
+
+### Coverage Analysis
+
+| Top N Words | Coverage |
+|-------------|----------|
+| Top 100 | 31.0% |
+| Top 1,000 | 55.5% |
+| Top 5,000 | 74.1% |
+| Top 10,000 | 80.8% |
+
+### Key Findings
+
+- **Zipf Compliance:** R²=0.9906 indicates excellent adherence to Zipf's law
+- **High Frequency Dominance:** Top 100 words cover 31.0% of corpus
+- **Long Tail:** 257,929 words needed for remaining 19.2% 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** | 129,587 | 32 | 3.467 | 1.190 | 0.7972 🏆 |
+| **mono_64d** | 129,587 | 64 | 3.957 | 1.135 | 0.7842 |
+| **mono_128d** | 129,587 | 128 | 4.525 | 1.098 | 0.7539 |
+| **embeddings_enhanced** | 0 | 0 | 0.000 | 0.000 | 0.0000 |
+
+### Key Findings
+
+- **Best Isotropy:** mono_32d with 0.7972 (more uniform distribution)
+- **Dimension Trade-off:** Higher dimensions capture more semantics but reduce isotropy
+- **Vocabulary Coverage:** All models cover 129,587 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 (4.74x) with low UNK rate |
+| N-gram | **5-gram** | Lowest perplexity (381) |
+| Markov | **Context-4** | Highest predictability (95.7%) |
+| 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 23:07:50*

models/embeddings/monolingual/ckb_128d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:41b65febdda7187f16f63077ef252551994a6e3947afcc78bbee22f7792d2c3b
+size 1159811328

models/embeddings/monolingual/ckb_128d.meta.json

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

models/embeddings/monolingual/ckb_128d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "ckb",
+  "dimension": 128,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 128,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 129587
+}

models/embeddings/monolingual/ckb_32d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa322184e03fc86097b753549864e28479c19a0f7f851a311fc6c9b07ca61784
+size 292288512

models/embeddings/monolingual/ckb_32d.meta.json

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

models/embeddings/monolingual/ckb_32d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "ckb",
+  "dimension": 32,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 32,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 129587
+}

models/embeddings/monolingual/ckb_64d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64aa558ec3f57c14e06b75aa214fae009c21695a65c7e5e450a744a7bdf362fe
+size 581462784

models/embeddings/monolingual/ckb_64d.meta.json

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

models/embeddings/monolingual/ckb_64d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "ckb",
+  "dimension": 64,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 64,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 129587
+}

models/subword_markov/ckb_markov_ctx1_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03d0443702a80a2957e258cfa45a0b3470f1b7c07853a411f35b6d93f4afb2c8
+size 328092

models/subword_markov/ckb_markov_ctx1_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "subword",
+  "language": "ckb",
+  "unique_contexts": 4993,
+  "total_transitions": 85527017
+}

models/subword_markov/ckb_markov_ctx2_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f1ea381df82c1b0e26b22165542771e484be712c855c85606ab8a53a4f60aa4
+size 2060794

models/subword_markov/ckb_markov_ctx2_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "subword",
+  "language": "ckb",
+  "unique_contexts": 46978,
+  "total_transitions": 85446235
+}

models/subword_markov/ckb_markov_ctx3_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6a716a5f7d117bbb45a7edd9f6a801c8bf99e17c4641490db0e5f6918e64f57
+size 9594491

models/subword_markov/ckb_markov_ctx3_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "subword",
+  "language": "ckb",
+  "unique_contexts": 284820,
+  "total_transitions": 85365453
+}

models/subword_markov/ckb_markov_ctx4_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:146b6fa63bf73ebf6bcb8bd0dbfd02d14ab02b74ec8a0f67e6f8278666bb2770
+size 30314193

models/subword_markov/ckb_markov_ctx4_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "subword",
+  "language": "ckb",
+  "unique_contexts": 1215545,
+  "total_transitions": 85284671
+}

models/subword_ngram/ckb_2gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a8d5ed23796744064159155b6a8eb09b9e7ccc31fd3789bc489bcc890fd3950
+size 202131

models/subword_ngram/ckb_2gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "subword",
+  "language": "ckb",
+  "unique_ngrams": 15002,
+  "total_ngrams": 85527017
+}

models/subword_ngram/ckb_3gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ae1d7e56ffe951fbc16e1b96e4b9213628ff12fce9be06bef038bd418284ba4
+size 1566915

models/subword_ngram/ckb_3gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "subword",
+  "language": "ckb",
+  "unique_ngrams": 125615,
+  "total_ngrams": 85446235
+}

models/subword_ngram/ckb_4gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a64971d7e7e32eed12ccb8129e7d8457f094683217e6d22fd974784a62f0c1b1
+size 8256147

models/subword_ngram/ckb_4gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "subword",
+  "language": "ckb",
+  "unique_ngrams": 641521,
+  "total_ngrams": 85365453
+}

models/tokenizer/ckb_tokenizer_16k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5818932355f70a312071fdf012ffcc1f50c4e1451abc5e72acc73d3be72b0626
+size 577885

models/tokenizer/ckb_tokenizer_32k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab68f2fd4bd5528d04b467e28ccd3f89089b8d2fc9a5bda6c8dac38fff5e4996
+size 934428

models/tokenizer/ckb_tokenizer_64k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7225ca8ff935c1a6affb6e0ac59d01a87f70c6a2205c6f3fe8db054411b7ee5
+size 1659196

models/tokenizer/ckb_tokenizer_8k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e9166b87186603a55e723d61f0f46f77b7f7d38a03ee0e8d993f1e91703dd00
+size 404312

models/vocabulary/ckb_vocabulary.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ddb813ea1d647c688855dc38fd806dfef2594dd35cdd646090b3888aff0fcbe
+size 4435068

models/vocabulary/ckb_vocabulary_metadata.json

@@ -0,0 +1,16 @@
+{
+  "language": "ckb",
+  "vocabulary_size": 267929,
+  "statistics": {
+    "type_token_ratio": 0.053893850732878223,
+    "coverage": {
+      "top_100": 0.29993262638188223,
+      "top_1000": 0.5369777451124179,
+      "top_5000": 0.7163147474143433,
+      "top_10000": 0.7817037242581987
+    },
+    "hapax_count": 415927,
+    "hapax_ratio": 0.6082084532416181,
+    "total_documents": 80782
+  }
+}

models/word_markov/ckb_markov_ctx1_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dcdc9ceae2045b65dda575029e9fa539662c1e44e10fe9fc75a489868b61c334
+size 52948370

models/word_markov/ckb_markov_ctx1_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "word",
+  "language": "ckb",
+  "unique_contexts": 684303,
+  "total_transitions": 15213811
+}

models/word_markov/ckb_markov_ctx2_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19f5e168313ff46635e3fd74d85d9b056bc8a862821538f049e04df57d27bc90
+size 163638338

models/word_markov/ckb_markov_ctx2_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "word",
+  "language": "ckb",
+  "unique_contexts": 4525451,
+  "total_transitions": 15133047
+}

models/word_markov/ckb_markov_ctx3_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e83caf11d902e66eb835337b0e123b8093f39e7f58e1c65b1f6a606f82a67bf
+size 257707250

models/word_markov/ckb_markov_ctx3_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "word",
+  "language": "ckb",
+  "unique_contexts": 8778035,
+  "total_transitions": 15053786
+}

models/word_markov/ckb_markov_ctx4_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e75f9bc70629fc67dd0c2bbea4fa0773a92883a136f775dd0b74f2e313bd4c5
+size 319480235

models/word_markov/ckb_markov_ctx4_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "word",
+  "language": "ckb",
+  "unique_contexts": 10844509,
+  "total_transitions": 14974526
+}

models/word_ngram/ckb_2gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5daa7fc9d2e32bb6868e61d734cffe27224cae7f8ec5fbc738564107aaf679b
+size 6437308

models/word_ngram/ckb_2gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "word",
+  "language": "ckb",
+  "unique_ngrams": 297076,
+  "total_ngrams": 15213811
+}

models/word_ngram/ckb_3gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7d0e3c3bf3c8488d5daa5fc51eb6e5fe82deacb144fe125c31a762e405111ddf
+size 11543522

models/word_ngram/ckb_3gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "word",
+  "language": "ckb",
+  "unique_ngrams": 466780,
+  "total_ngrams": 15133047
+}

models/word_ngram/ckb_4gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0f51843388b5c6b1409743a19137c90f12baca602af40ec707d368f0da1f9e7
+size 20077575

models/word_ngram/ckb_4gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "word",
+  "language": "ckb",
+  "unique_ngrams": 754916,
+  "total_ngrams": 15053786
+}