| # Word2Vec Model | |
| This is a Word2Vec model trained on text data. | |
| ## Model Details | |
| - **Vocabulary size**: 3 | |
| - **Embedding dimension**: 64 | |
| - **Total words**: 35174 | |
| ## Usage | |
| ```python | |
| from transformers import AutoTokenizer | |
| import torch | |
| # Load tokenizer | |
| tokenizer = AutoTokenizer.from_pretrained("roshbeed/hn-word2vec") | |
| # Load embeddings | |
| embeddings = torch.load("word_embeddings.pt") | |
| ``` | |
| ## Training Data | |
| Trained on text8 dataset with synthetic upvote scores. | |