117 lines
2.6 KiB
Markdown
117 lines
2.6 KiB
Markdown
---
|
|
library_name: transformers
|
|
tags:
|
|
- text-generation
|
|
- causal-lm
|
|
- conversational
|
|
- llama
|
|
- dpo
|
|
- safetensors
|
|
- study-assistant
|
|
---
|
|
|
|
# Model Card for Master Buddy AI v1.0
|
|
|
|
Master Buddy AI is a fine-tuned conversational language model designed as a computer science and engineering study assistant. It is built to help users understand programming concepts, algorithms, and general technical topics in a structured and helpful manner.
|
|
|
|
---
|
|
|
|
## Model Details
|
|
|
|
### Model Description
|
|
|
|
Master Buddy AI v1.0 is a decoder-only transformer-based language model fine-tuned using:
|
|
|
|
- Supervised Fine-Tuning (SFT)
|
|
- Direct Preference Optimization (DPO)
|
|
|
|
The goal of this model is to improve response helpfulness, reasoning ability, and instruction-following quality.
|
|
|
|
- **Developed by:** Seran Vishwa
|
|
- **Model type:** Causal Language Model (Decoder-only Transformer)
|
|
- **Language(s):** English
|
|
- **License:** Apache-2.0
|
|
- **Finetuned from model:** LLaMA-based architecture or equivalent base model
|
|
|
|
---
|
|
|
|
## Model Sources
|
|
|
|
- **Repository:** https://huggingface.co/SeranVishwa/master-buddy-v1.0
|
|
- **Demo:** Add deployment link when available
|
|
|
|
---
|
|
|
|
## Uses
|
|
|
|
### Direct Use
|
|
|
|
This model is intended for:
|
|
|
|
- Programming assistance
|
|
- Data structures and algorithms explanations
|
|
- Computer science concepts (OS, DBMS, Networks)
|
|
- Educational and study support
|
|
- General question answering
|
|
|
|
---
|
|
|
|
### Downstream Use
|
|
|
|
The model can be integrated into:
|
|
|
|
- AI tutoring systems
|
|
- Chatbot applications
|
|
- Educational platforms
|
|
- Developer tools
|
|
- Web-based AI assistants
|
|
|
|
---
|
|
|
|
### Out-of-Scope Use
|
|
|
|
This model is not intended for:
|
|
|
|
- Medical advice
|
|
- Legal advice
|
|
- Financial decision-making
|
|
- Safety-critical systems
|
|
- Harmful or malicious use cases
|
|
|
|
---
|
|
|
|
## Bias, Risks, and Limitations
|
|
|
|
- The model may produce incorrect or hallucinated responses
|
|
- It reflects biases present in training data
|
|
- It should not be used as a sole source of truth
|
|
- Outputs should be verified for critical use cases
|
|
|
|
---
|
|
|
|
### Recommendations
|
|
|
|
Users should:
|
|
|
|
- Verify important technical outputs
|
|
- Use the model as a learning assistant, not an authority
|
|
- Be aware of possible reasoning errors
|
|
|
|
---
|
|
|
|
## How to Get Started
|
|
|
|
### Transformers
|
|
|
|
```python
|
|
from transformers import AutoModelForCausalLM, AutoTokenizer
|
|
|
|
model = AutoModelForCausalLM.from_pretrained("SeranVishwa/master-buddy-v1.0")
|
|
tokenizer = AutoTokenizer.from_pretrained("SeranVishwa/master-buddy-v1.0")
|
|
|
|
prompt = "Explain Dijkstra's algorithm"
|
|
|
|
inputs = tokenizer(prompt, return_tensors="pt")
|
|
outputs = model.generate(**inputs, max_new_tokens=200)
|
|
|
|
print(tokenizer.decode(outputs[0], skip_special_tokens=True)) |