|
|
--- |
|
|
license: apache-2.0 |
|
|
tags: |
|
|
- pruned |
|
|
- python |
|
|
- optimized |
|
|
- wanda |
|
|
- activation-pruning |
|
|
base_model: Qwen/Qwen2.5-3B-Instruct |
|
|
pipeline_tag: text-generation |
|
|
--- |
|
|
|
|
|
# Qwen2.5-3B-Instruct-python-aggressive |
|
|
|
|
|
> 🎯 **PYTHON-optimized** | 📦 **Aggressive** pruning | ⚡ **20% weights pruned** |
|
|
|
|
|
This model is a **aggressively pruned** version of [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct), specialized for **PYTHON** tasks using activation-aware weight pruning (Wanda-style). |
|
|
|
|
|
## ✨ Key Features |
|
|
|
|
|
- **Specialization**: Optimized for Python tasks |
|
|
- **Pruning Method**: Wanda-style (|W| × |activation|) importance scoring |
|
|
- **Size Reduction**: 20% weights pruned |
|
|
- **Use Case**: Maximum compression for edge deployment |
|
|
|
|
|
## 📊 Performance Comparison |
|
|
|
|
|
| Category | Original | Pruned | Change | |
|
|
|----------|----------|--------|--------| |
|
|
| **Python** | 100.0% | 40.0% ⭐ | ↓ 60.0% | |
|
|
| Html | 6.7% | 0.0% | ↓ 6.7% | |
|
|
| Trivia | 66.7% | 60.0% | ↓ 6.7% | |
|
|
| Math | 60.0% | 66.7% | ↑ 6.7% | |
|
|
| Reasoning | 100.0% | 86.7% | ↓ 13.3% | |
|
|
| Medical | 86.7% | 80.0% | ↓ 6.7% | |
|
|
| Linux | 100.0% | 100.0% | → | |
|
|
| Writing | 73.3% | 73.3% | → | |
|
|
|
|
|
**Average**: 74.2% → 63.3% (-10.8%) |
|
|
|
|
|
**Python Retention**: 40.0% of original performance |
|
|
|
|
|
 |
|
|
|
|
|
## 🚀 Quick Start |
|
|
|
|
|
```python |
|
|
from transformers import AutoModelForCausalLM, AutoTokenizer |
|
|
|
|
|
model = AutoModelForCausalLM.from_pretrained("CompactAI/Qwen2.5-3B-Instruct-python-aggressive") |
|
|
tokenizer = AutoTokenizer.from_pretrained("CompactAI/Qwen2.5-3B-Instruct-python-aggressive") |
|
|
|
|
|
# Example usage |
|
|
inputs = tokenizer("Your prompt here", return_tensors="pt") |
|
|
outputs = model.generate(**inputs, max_new_tokens=100) |
|
|
print(tokenizer.decode(outputs[0], skip_special_tokens=True)) |
|
|
``` |
|
|
|
|
|
## 📋 Technical Details |
|
|
|
|
|
| Property | Value | |
|
|
|----------|-------| |
|
|
| Base Model | [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct) | |
|
|
| Specialization | Python | |
|
|
| Prune Mode | Aggressive | |
|
|
| Pruning Method | Activation-based weight pruning (Wanda) | |
|
|
| Weight Reduction | 20% weights pruned | |
|
|
|
|
|
## 🔗 Related Models |
|
|
|
|
|
This model is part of the **Qwen2.5-3B-Instruct** pruned model collection. Variants: |
|
|
- **Safe** - Conservative pruning (~10-20%), high accuracy retention |
|
|
- **Aggressive** - Maximum compression (~40-50%), best for edge deployment |
|
|
|
|
|
## 📜 License |
|
|
|
|
|
This model inherits the license from the base model [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct). |
|
|
|
|
|
--- |
|
|
*Generated by ZANNPS [Zeto Automatic Neural Network Pruning System]* |
|
|
|
