Upload all models and assets for ast (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: ast
+language_name: AST
+language_family: romance_iberian
+tags:
+  - wikilangs
+  - nlp
+  - tokenizer
+  - embeddings
+  - n-gram
+  - markov
+  - wikipedia
+  - monolingual
+  - family-romance_iberian
+license: mit
+library_name: wikilangs
+pipeline_tag: feature-extraction
+datasets:
+  - omarkamali/wikipedia-monthly
+dataset_info:
+  name: wikipedia-monthly
+  description: Monthly snapshots of Wikipedia articles across 300+ languages
+metrics:
+  - name: best_compression_ratio
+    type: compression
+    value: 3.924
+  - name: best_isotropy
+    type: isotropy
+    value: 0.7692
+  - name: vocabulary_size
+    type: vocab
+    value: 654549
+generated: 2025-12-27
+---
+
+# AST - Wikilangs Models
+## Comprehensive Research Report & Full Ablation Study
+
+This repository contains NLP models trained and evaluated by Wikilangs, specifically on **AST** 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** | 3.259x | 3.22 | 0.0290% | 1,033,064 |
+| **16k** | 3.531x | 3.48 | 0.0315% | 953,475 |
+| **32k** | 3.753x | 3.70 | 0.0334% | 897,137 |
+| **64k** | 3.924x 🏆 | 3.87 | 0.0350% | 858,173 |
+
+### Tokenization Examples
+
+Below are sample sentences tokenized with each vocabulary size:
+
+**Sample 1:** `Fechos 
+
+ Personaxes importantes 
+
+ Referencies 
+
+ Enllaces esternos 
+
+Categoría...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁fechos ▁personaxes ▁importantes ▁referencies ▁enllaces ▁esternos ▁categoría : sieglu ▁viii ... (+4 more)` | 14 |
+| 16k | `▁fechos ▁personaxes ▁importantes ▁referencies ▁enllaces ▁esternos ▁categoría : sieglu ▁viii ... (+4 more)` | 14 |
+| 32k | `▁fechos ▁personaxes ▁importantes ▁referencies ▁enllaces ▁esternos ▁categoría : sieglu ▁viii ... (+4 more)` | 14 |
+| 64k | `▁fechos ▁personaxes ▁importantes ▁referencies ▁enllaces ▁esternos ▁categoría : sieglu ▁viii ... (+4 more)` | 14 |
+
+**Sample 2:** `Armental ye un llugar de la parroquia de Talarén nel conceyu asturianu de Navia....`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁ar mental ▁ye ▁un ▁llugar ▁de ▁la ▁parroquia ▁de ▁tal ... (+18 more)` | 28 |
+| 16k | `▁ar mental ▁ye ▁un ▁llugar ▁de ▁la ▁parroquia ▁de ▁tal ... (+18 more)` | 28 |
+| 32k | `▁ar mental ▁ye ▁un ▁llugar ▁de ▁la ▁parroquia ▁de ▁tal ... (+16 more)` | 26 |
+| 64k | `▁ar mental ▁ye ▁un ▁llugar ▁de ▁la ▁parroquia ▁de ▁tal ... (+16 more)` | 26 |
+
+**Sample 3:** `Fechos 
+  -
+
+ Nacencies 
+  -
+
+ Muertes 
+  -
+
+ Referencies 
+
+ Enllaces esternos 
+...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁fechos ▁- ▁nacencies ▁- ▁muertes ▁- ▁referencies ▁enllaces ▁esternos ▁categoría ... (+7 more)` | 17 |
+| 16k | `▁fechos ▁- ▁nacencies ▁- ▁muertes ▁- ▁referencies ▁enllaces ▁esternos ▁categoría ... (+7 more)` | 17 |
+| 32k | `▁fechos ▁- ▁nacencies ▁- ▁muertes ▁- ▁referencies ▁enllaces ▁esternos ▁categoría ... (+7 more)` | 17 |
+| 64k | `▁fechos ▁- ▁nacencies ▁- ▁muertes ▁- ▁referencies ▁enllaces ▁esternos ▁categoría ... (+7 more)` | 17 |
+
+
+### Key Findings
+
+- **Best Compression:** 64k achieves 3.924x compression
+- **Lowest UNK Rate:** 8k with 0.0290% 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** | 95,540 🏆 | 16.54 | 1,568,799 | 13.6% | 26.9% |
+| **2-gram** | 311 🏆 | 8.28 | 23,389 | 65.5% | 98.4% |
+| **3-gram** | 573,984 | 19.13 | 3,974,147 | 5.1% | 13.4% |
+| **3-gram** | 2,766 | 11.43 | 195,082 | 25.8% | 68.5% |
+| **4-gram** | 1,609,317 | 20.62 | 7,247,181 | 3.9% | 9.3% |
+| **4-gram** | 16,954 | 14.05 | 1,178,742 | 12.7% | 37.0% |
+
+### Top 5 N-grams by Size
+
+**2-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `d '` | 1,196,313 |
+| 2 | `de la` | 875,667 |
+| 3 | `' l` | 534,478 |
+| 4 | `| |` | 438,858 |
+| 5 | `l '` | 403,691 |
+
+**3-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `| - |` | 128,285 |
+| 2 | `referencies enllaces esternos` | 104,162 |
+| 3 | `- | |` | 89,758 |
+| 4 | `- - -` | 81,514 |
+| 5 | `d ' un` | 69,529 |
+
+**4-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `- - - -` | 69,470 |
+| 2 | `enllaces esternos categoría :` | 63,833 |
+| 3 | `referencies enllaces esternos categoría` | 60,665 |
+| 4 | `. referencies enllaces esternos` | 51,144 |
+| 5 | `| linear | -` | 50,481 |
+
+
+### Key Findings
+
+- **Best Perplexity:** 2-gram with 311
+- **Entropy Trend:** Decreases with larger n-grams (more predictable)
+- **Coverage:** Top-1000 patterns cover ~37% 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** | 0.7150 | 1.641 | 8.69 | 1,669,949 | 28.5% |
+| **1** | 1.5193 | 2.866 | 10.68 | 8,875 | 0.0% |
+| **2** | 0.4611 | 1.377 | 2.90 | 14,499,551 | 53.9% |
+| **2** | 0.7271 | 1.655 | 4.90 | 94,766 | 27.3% |
+| **3** | 0.2234 | 1.167 | 1.58 | 42,031,886 | 77.7% |
+| **3** | 0.8068 | 1.749 | 4.59 | 464,259 | 19.3% |
+| **4** | 0.1062 🏆 | 1.076 | 1.22 | 66,322,442 | 89.4% |
+| **4** | 0.7182 🏆 | 1.645 | 3.49 | 2,131,889 | 28.2% |
+
+### Generated Text Samples
+
+Below are text samples generated from each Markov chain model:
+
+**Context Size 1:**
+
+1. `de gossip girl play ye como xenofonte en dussel , y el 7 d ' amuesa`
+2. `, pero la botánica referencies ver , collaboró en valdivia . mientres la humanidá al chinu`
+3. `. ( 2 ) - ḥḏ horusmuriu blancumennefermenfismit rahina . isbn 0 british lion , yera`
+
+**Context Size 2:**
+
+1. `d ' ellos yera detectáu polos enemigos . shiva prakash ( 1997 ) , nel conceyu sevillanu`
+2. `de la litografía y l ' ala posterior : chronica majora : una « inocente ya inconsciente`
+3. `' l xeneral prats tamién pudo ante fernando verdasco david ferrer por 6 - 2 | ríu`
+
+**Context Size 3:**
+
+1. `| - | 38378 - | | 1997 tb18 | | 4 | align = right | [`
+2. `referencies enllaces esternos categoría : montserrat`
+3. `- | | 2001 sd35 | | 16 | | 592 | | < small > 1911 <`
+
+**Context Size 4:**
+
+1. `- - - - - - - - - - - - - - - - - - -`
+2. `enllaces esternos categoría : pintores de parís categoría : sabios de la torre eiffel , los nacional...`
+3. `referencies enllaces esternos categoría : comuñes de nord`
+
+
+### Key Findings
+
+- **Best Predictability:** Context-4 with 89.4% predictability
+- **Branching Factor:** Decreases with context size (more deterministic)
+- **Memory Trade-off:** Larger contexts require more storage (2,131,889 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 | 654,549 |
+| Total Tokens | 80,184,102 |
+| Mean Frequency | 122.50 |
+| Median Frequency | 4 |
+| Frequency Std Dev | 8722.95 |
+
+### Most Common Words
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | de | 5,075,921 |
+| 2 | la | 2,521,840 |
+| 3 | y | 2,071,360 |
+| 4 | d | 1,229,266 |
+| 5 | a | 1,176,335 |
+| 6 | del | 1,090,980 |
+| 7 | en | 1,071,173 |
+| 8 | que | 1,020,518 |
+| 9 | los | 971,499 |
+| 10 | l | 968,352 |
+
+### Least Common Words (from vocabulary)
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | leptafeke | 2 |
+| 2 | haua | 2 |
+| 3 | küzdoblani | 2 |
+| 4 | contrarrellatu | 2 |
+| 5 | semilleru | 2 |
+| 6 | bisterca | 2 |
+| 7 | šafarsko | 2 |
+| 8 | vyfalu | 2 |
+| 9 | ribich | 2 |
+| 10 | lacos | 2 |
+
+### Zipf's Law Analysis
+
+| Metric | Value |
+|--------|-------|
+| Zipf Coefficient | 1.0077 |
+| R² (Goodness of Fit) | 0.995140 |
+| Adherence Quality | **excellent** |
+
+### Coverage Analysis
+
+| Top N Words | Coverage |
+|-------------|----------|
+| Top 100 | 40.0% |
+| Top 1,000 | 60.0% |
+| Top 5,000 | 76.4% |
+| Top 10,000 | 82.7% |
+
+### Key Findings
+
+- **Zipf Compliance:** R²=0.9951 indicates excellent adherence to Zipf's law
+- **High Frequency Dominance:** Top 100 words cover 40.0% of corpus
+- **Long Tail:** 644,549 words needed for remaining 17.3% 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** | 510,373 | 32 | 3.008 | 0.935 | 0.7692 🏆 |
+| **mono_64d** | 510,373 | 64 | 3.395 | 0.938 | 0.7616 |
+| **mono_128d** | 510,373 | 128 | 3.842 | 0.965 | 0.6988 |
+| **embeddings_enhanced** | 0 | 0 | 0.000 | 0.000 | 0.0000 |
+
+### Key Findings
+
+- **Best Isotropy:** mono_32d with 0.7692 (more uniform distribution)
+- **Dimension Trade-off:** Higher dimensions capture more semantics but reduce isotropy
+- **Vocabulary Coverage:** All models cover 510,373 words
+- **Recommendation:** 100d for balanced semantic capture and efficiency
+
+---
+## 6. Summary & Recommendations
+
+![Performance Dashboard](visualizations/performance_dashboard.png)
+
+### Production Recommendations
+
+| Component | Recommended | Rationale |
+|-----------|-------------|-----------|
+| Tokenizer | **32k BPE** | Best compression (3.92x) with low UNK rate |
+| N-gram | **5-gram** | Lowest perplexity (311) |
+| Markov | **Context-4** | Highest predictability (89.4%) |
+| 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 20:35:27*

models/embeddings/monolingual/ast_128d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39f5f134a23b0ae5b3bba7a4864ecc3599e28dd6a5c44748c538e6eb50542a1b
+size 1555994683

models/embeddings/monolingual/ast_128d.meta.json

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

models/embeddings/monolingual/ast_128d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "ast",
+  "dimension": 128,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 128,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 510373
+}

models/embeddings/monolingual/ast_32d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aecd461d0dd299f160b9fefcb4d6bbc47ffe601253410b6bf66aa59353683649
+size 396028219

models/embeddings/monolingual/ast_32d.meta.json

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

models/embeddings/monolingual/ast_32d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "ast",
+  "dimension": 32,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 32,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 510373
+}

models/embeddings/monolingual/ast_64d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7552caa3e9f8ef8607c05afb5b9266e4c9a3e70fdba272075a7b9912e995045
+size 782683707

models/embeddings/monolingual/ast_64d.meta.json

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

models/embeddings/monolingual/ast_64d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "ast",
+  "dimension": 64,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 64,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 510373
+}

models/subword_markov/ast_markov_ctx1_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:797c1dfcb6c516214d2dc02ed941e64e840e49d291e2cc758d5482d86ae04640
+size 602821

models/subword_markov/ast_markov_ctx1_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "subword",
+  "language": "ast",
+  "unique_contexts": 8875,
+  "total_transitions": 508763835
+}

models/subword_markov/ast_markov_ctx2_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0ee540fb4354bf9779b55ea96cde4252defd9f562e14d445ac6b6c9b457c3f1
+size 3577755

models/subword_markov/ast_markov_ctx2_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "subword",
+  "language": "ast",
+  "unique_contexts": 94766,
+  "total_transitions": 508625606
+}

models/subword_markov/ast_markov_ctx3_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db3f45e2155b76125f075309cbec6b285d5ee2401fd063d4b596638cb534193e
+size 16073938

models/subword_markov/ast_markov_ctx3_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "subword",
+  "language": "ast",
+  "unique_contexts": 464259,
+  "total_transitions": 508487377
+}

models/subword_markov/ast_markov_ctx4_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd0b9d4e9db8c19eeb444ba13e49b7a48ec4b84512f949ef315285aff0822a0e
+size 57292008

models/subword_markov/ast_markov_ctx4_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "subword",
+  "language": "ast",
+  "unique_contexts": 2131889,
+  "total_transitions": 508349148
+}

models/subword_ngram/ast_2gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:417e9408d0c904ed21f23501e55082a7494276c2cfce4650855bdde090aa9e15
+size 314961

models/subword_ngram/ast_2gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "subword",
+  "language": "ast",
+  "unique_ngrams": 23389,
+  "total_ngrams": 508763835
+}

models/subword_ngram/ast_3gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:730e803a7b1373e12874f3e3ee787c6229e54c4f714b7b5b45ecbc90e7e1f2ef
+size 2302607

models/subword_ngram/ast_3gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "subword",
+  "language": "ast",
+  "unique_ngrams": 195082,
+  "total_ngrams": 508625606
+}

models/subword_ngram/ast_4gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b418657cc08b4f91e0b2c2eda00aef5c4302d754fe9dccc5feb84ff1b24a6291
+size 13600667

models/subword_ngram/ast_4gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "subword",
+  "language": "ast",
+  "unique_ngrams": 1178742,
+  "total_ngrams": 508487377
+}

models/tokenizer/ast_tokenizer_16k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:907f4bece802b1d3656fdf65b960935cdd809ca5733b7b73b0823ce18af2c113
+size 513109

models/tokenizer/ast_tokenizer_32k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88d4563db9be715e9225ff77615f89d344e1f2db1b32e1119ffcb1adf873068a
+size 795952

models/tokenizer/ast_tokenizer_64k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bab637cbda657ea59afedbac3a722b8a7d18839ff031e3a67f1d79f914cb1edd
+size 1371599

models/tokenizer/ast_tokenizer_8k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53538ed204ba1e90b9a3ec72f89f3a9639ddf5f7e06a5d91daf2182e1dbacf32
+size 374583

models/vocabulary/ast_vocabulary.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e95a025d0a2e0d5c5fc9e67b2724543af5eaadf84e83944b5601b9e113deee95
+size 9493041

models/vocabulary/ast_vocabulary_metadata.json

@@ -0,0 +1,16 @@
+{
+  "language": "ast",
+  "vocabulary_size": 654549,
+  "statistics": {
+    "type_token_ratio": 0.020582308161438793,
+    "coverage": {
+      "top_100": 0.39513983436676664,
+      "top_1000": 0.5923739422740554,
+      "top_5000": 0.7541024014353347,
+      "top_10000": 0.8168048946874377
+    },
+    "hapax_count": 1016752,
+    "hapax_ratio": 0.608359595309283,
+    "total_documents": 138229
+  }
+}

models/word_markov/ast_markov_ctx1_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e50511b1eedd319fddb7babd82679e20996cdc8c40bb7414b7729fa744061d38
+size 125168928

models/word_markov/ast_markov_ctx1_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "word",
+  "language": "ast",
+  "unique_contexts": 1669949,
+  "total_transitions": 101266716
+}

models/word_markov/ast_markov_ctx2_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e97c5fda98a744b9ecdbc923910e1c4de28e3c699d3c7a3519ea2d5826e515a
+size 436569711

models/word_markov/ast_markov_ctx2_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "word",
+  "language": "ast",
+  "unique_contexts": 14499551,
+  "total_transitions": 101128488
+}

models/word_markov/ast_markov_ctx3_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d70d8996da08da859f83f17208f24ec5fe8a079a8cdc33b458965bdf29e6a65
+size 863814981

models/word_markov/ast_markov_ctx3_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "word",
+  "language": "ast",
+  "unique_contexts": 42031886,
+  "total_transitions": 100990260
+}

models/word_markov/ast_markov_ctx4_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:edd90f355e0db9ed093dd1b20ea832c072fe6c4eb8eda9897b82053102ed9e22
+size 1203044127

models/word_markov/ast_markov_ctx4_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "word",
+  "language": "ast",
+  "unique_contexts": 66322442,
+  "total_transitions": 100852034
+}

models/word_ngram/ast_2gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4498d3f067003db47603a9cd549742083283922e1e3bcfac7e995862e92c13a0
+size 22263161

models/word_ngram/ast_2gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "word",
+  "language": "ast",
+  "unique_ngrams": 1568799,
+  "total_ngrams": 101266716
+}

models/word_ngram/ast_3gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77d87442f951bfee562611f31c08d2afebd188296c7dc1f3fc4691abc873c580
+size 61024149

models/word_ngram/ast_3gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "word",
+  "language": "ast",
+  "unique_ngrams": 3974147,
+  "total_ngrams": 101128488
+}

models/word_ngram/ast_4gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a09cc0505f23b8e2addb0fd5ea158db1477037a7668eedab235820dfbcb3df5
+size 117588970

models/word_ngram/ast_4gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "word",
+  "language": "ast",
+  "unique_ngrams": 7247181,
+  "total_ngrams": 100990260
+}