Update README.md

README.md

@@ -7,42 +7,88 @@ tags:
   - philosophy
   - conversational-ai
   - religious-text
+  - self-realization
 license: apache-2.0
 pipeline_tag: text-generation
 ---
 
-# AtmasiddhiGPT
+# AtmasiddhiGPTv9-gguf
 
-AtmasiddhiGPT is a language model trained to reflect and provide insights based on the teachings of the **Atmasiddhi Shastra**, a Jain text written by Shrimad Rajchandra. This model can be used to generate responses that capture themes of self-realization, soul, and philosophical principles in Jainism, making it a valuable tool for those exploring Jain spirituality and philosophy.
+**AtmasiddhiGPTv9-gguf** is a fine-tuned version of the **LLaMA 3.2 3B Instruct** model, designed to provide conversational insights and spiritual guidance based on the teachings of the *Atmasiddhi Shastra*, a revered Jain text by Shrimad Rajchandra. This model is specially aligned with contemporary interpretations by Shri Gurudevshri Rakeshbhai, making it a valuable tool for exploring the themes of self-realization, soul, and liberation in Jain philosophy.
 
 ## Model Details
 
-- **Model Type**: Language Model
+- **Model Name**: AtmasiddhiGPTv9-gguf
+- **Base Model**: LLaMA 3.2 3B Instruct (Meta)
+- **Model Type**: Language Model (GGUF format)
 - **Language**: English
-- **Training Data**: Selected spiritual texts, with an emphasis on **Atmasiddhi Shastra**
-- **Intended Use**: Conversational AI, philosophical inquiry, spirituality
+- **Intended Use**: Spiritual guidance, philosophical inquiry, Jain studies, self-reflection
+- **Alignment**: Based on the recent commentaries and teachings of Shri Gurudevshri Rakeshbhai on the *Atmasiddhi Shastra*
+- **Recommended Platforms**: LM Studio, Jan (support GGUF models)
 - **License**: Apache 2.0
-- **Framework**: PyTorch, Transformers
+- **Framework**: GGUF-compatible
 
-## How to Use
+## Model Scope and Purpose
 
-To use AtmasiddhiGPT for generating responses in a conversational style or to interpret spiritual themes, you can load it directly with the Hugging Face `transformers` library. Here’s an example of how to set it up:
+**AtmasiddhiGPTv9-gguf** is designed to serve as an interactive tool for individuals seeking a deeper understanding of Jain spiritual concepts, guided by the most recent teachings of Shri Gurudevshri Rakeshbhai. This model uses the philosophical foundation of the *Atmasiddhi Shastra* while adopting the conversational style of the LLaMA 3.2 3B Instruct model, ensuring responses are both spiritually aligned and easily understandable.
+
+### Key Philosophical Themes
+
+The model focuses on interpreting key themes of the *Atmasiddhi Shastra*, particularly as presented in Shri Gurudevshri Rakeshbhai’s teachings. These include:
+
+1. **The Nature of the Soul (Atma)**: Exploring the soul's inherent qualities, permanence, and its distinction from physical existence.
+2. **Path to Liberation (Moksha)**: Insights into the steps and virtues needed to achieve liberation from the cycle of birth and death.
+3. **Karma and Its Impact**: Explanations of karmic law, the effects of accumulated karma, and how it shapes the soul’s journey.
+4. **Self-Realization**: Encouraging self-inquiry to unveil true self-identity and transcend ego-driven life.
+5. **Discernment and Detachment (Vairagya)**: Offering practical advice on embracing detachment, renouncing material attachments, and cultivating spiritual insight.
+
+The model seeks to convey these themes with the depth and clarity characteristic of Shri Gurudevshri’s teachings, while maintaining the conversational ease provided by the LLaMA 3.2 3B Instruct model architecture.
+
+## Recommended Platforms: LM Studio and Jan
+
+AtmasiddhiGPTv9-gguf is optimized for use with GGUF-compatible applications like LM Studio and Jan, which allow local, offline interactions with the model.
+
+### [LM Studio](https://lmstudio.ai/download)
+
+LM Studio is a free application supporting GGUF-formatted models, ideal for downloading and running large language models offline.
+
+**How to Use AtmasiddhiGPTv9-gguf with LM Studio**:
+1. **Download LM Studio**: Visit the [LM Studio download page](https://lmstudio.ai/download) and choose your operating system.
+2. **Install and Launch**: Follow the installation instructions provided.
+3. **Load the Model**:
+   - Search for "AtmasiddhiGPTv9-gguf" in the model catalog, or import it manually if previously downloaded.
+   - Interact with the model via LM Studio’s chat interface or set up a local API server for integration into applications.
+
+For additional guidance, refer to the [LM Studio Documentation](https://lmstudio.ai/docs).
+
+### [Jan](https://jan.ai/download)
+
+Jan is an open-source application that supports GGUF models, allowing users to interact with models entirely offline.
+
+**How to Use AtmasiddhiGPTv9-gguf with Jan**:
+1. **Download Jan**: Access the [Jan download page](https://jan.ai/download).
+2. **Install and Launch Jan**: Follow the setup instructions.
+3. **Import the Model**:
+   - Use Jan’s model management section to add the AtmasiddhiGPTv9-gguf model.
+   - Engage with the model via Jan’s conversational interface.
+
+Refer to [Jan Documentation](https://jan.ai/docs) for more details.
+
+## Example Code for Local Use
+
+To load AtmasiddhiGPTv9-gguf with compatible libraries (if supported) or GGUF-compatible applications, you can use this sample code:
 
 ```python
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
-# Load the model and tokenizer
-model_name = "your-username/atmasiddhi-gpt"  # replace with actual model path
+# Model path
+model_name = "HappyAIUser/AtmasiddhiGPTv9-gguf"
 tokenizer = AutoTokenizer.from_pretrained(model_name)
 model = AutoModelForCausalLM.from_pretrained(model_name)
 
-# Define input text (e.g., a question or a spiritual query)
-input_text = "What does Atmasiddhi say about self-realization?"
-
-# Tokenize and generate
+# Sample input
+input_text = "What insights does Atmasiddhi offer about liberation?"
 inputs = tokenizer(input_text, return_tensors="pt")
 outputs = model.generate(**inputs, max_length=50)
-
-# Decode and print the generated response
 response = tokenizer.decode(outputs[0], skip_special_tokens=True)
 print(response)