Fix model sizes (32B/8B), add thinking tag, add all datasets and eval results

Upload index.html with huggingface_hub

Upload index.html with huggingface_hub

initial commit