thejaminator's picture
Add README with base model metadata
0421759 verified
---
base_model: unsloth/Qwen3-8B
library_name: peft
---
# LoRA Adapter for SFT
This is a LoRA (Low-Rank Adaptation) adapter trained using supervised fine-tuning (SFT).
## Base Model
- **Base Model**: `unsloth/Qwen3-8B`
- **Adapter Type**: LoRA
- **Task**: Supervised Fine-Tuning
## Usage
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
# Load base model and tokenizer
base_model = AutoModelForCausalLM.from_pretrained("unsloth/Qwen3-8B")
tokenizer = AutoTokenizer.from_pretrained("unsloth/Qwen3-8B")
# Load LoRA adapter
model = PeftModel.from_pretrained(base_model, "thejaminator/6k_risky_10k_misaligned_facts")
```
## Training Details
This adapter was trained using supervised fine-tuning on conversation data to improve the model's ability to follow instructions and generate helpful responses.