| # Savyasachi – AI Devotee of Lord Krishna | |
| Savyasachi is a fine-tuned AI trained to respond as a devoted follower of Lord Krishna. The model provides **philosophical, devotional, and Gita-inspired guidance** in a calm, respectful, and clear tone. | |
| --- | |
| ## Model Details | |
| - **Model name:** `Devishetty100/savyasachi` | |
| - **Base model:** Gemma 3 / LLaMA-based transformer | |
| - **Fine-tuned on:** Bhagavad Gita question-answer dialogues (`question` → `krishna_teaching`) | |
| - **Persona:** Savyasachi, the greatest devotee of Lord Krishna | |
| - **Task:** Instruction-following, conversational answering | |
| --- | |
| ## Intended Use | |
| - Chatbots for spiritual or educational purposes | |
| - Learning and understanding Bhagavad Gita teachings | |
| - Philosophical Q&A, devotional guidance | |
| **Not intended for:** | |
| - Medical, legal, or personal advice | |
| - Misrepresentation of religious texts | |
| --- | |
| ## Usage Example | |
| ```python | |
| from transformers import AutoModelForCausalLM, AutoTokenizer | |
| import torch | |
| model_name = "Devishetty100/savyasachi" | |
| tokenizer = AutoTokenizer.from_pretrained(model_name) | |
| model = AutoModelForCausalLM.from_pretrained(model_name).to("cuda" if torch.cuda.is_available() else "cpu") | |
| question = "What is the meaning of karma?" | |
| input_ids = tokenizer(question, return_tensors="pt").to(model.device).input_ids | |
| output_ids = model.generate(input_ids, max_new_tokens=150, temperature=1.0, top_p=0.95, top_k=64) | |
| answer = tokenizer.decode(output_ids[0], skip_special_tokens=True) | |
| print(answer) | |