v3: improved model with hard negatives (F1 74%->95%)

v3: improved model with hard negatives (F1 74%->95%)

Upload README.md with huggingface_hub

Upload vocab.txt with huggingface_hub

Upload special_tokens_map.json with huggingface_hub

Upload tokenizer_config.json with huggingface_hub

Upload tokenizer.json with huggingface_hub

Upload model.safetensors with huggingface_hub

Upload config.json with huggingface_hub

initial commit