Upload README.md with huggingface_hub

README.md

@@ -0,0 +1,60 @@
+---
+language: it
+license: apache-2.0
+library_name: peft
+base_model: Qwen/Qwen2.5-1.5B
+tags:
+  - lora
+  - cognitive-architecture
+  - progressive-learning
+  - magnitude-pruning
+  - math
+  - arithmetic
+datasets:
+  - custom
+pipeline_tag: text-generation
+---
+
+# Progressive Cognitive Architecture — Progressive-LoRA (Qwen 2.5 1.5B)
+
+**4-phase progressive training with magnitude pruning (the predecessor to Dream Pruning).**
+
+## What is this?
+
+This is the original Progressive-LoRA model using **magnitude pruning** (zeroing small weights) instead of SVD Dream Pruning. It was the first version of the progressive cognitive architecture.
+
+## 📊 Results
+
+| Metric | Progressive-LoRA (this) | Dream-LoRA | Flat-LoRA |
+|--------|------------------------|-----------|-----------|
+| Exact Accuracy | 37.0% ± 0.5 | 58.6% ± 2.9 | 60.6% |
+| Number Sense | 57.7% ± 0.5 | 60.0% ± 0.8 | 0.0% |
+| Metacognition | 98.5% | 100.0% | 0.0% |
+
+Dream Pruning (SVD) significantly improved upon magnitude pruning by preserving more of the learned information during compression.
+
+## 🚀 Quick Start
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from peft import PeftModel
+
+base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-1.5B", device_map="auto")
+tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2.5-1.5B")
+
+# Note: adapters are in lora_adapters/ subfolder
+model = PeftModel.from_pretrained(base_model, "dexmac/progressive-cognitive-lora", subfolder="lora_adapters")
+
+inputs = tokenizer("Calcola: 347 + 891 =", return_tensors="pt").to(model.device)
+outputs = model.generate(**inputs, max_new_tokens=20)
+print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+```
+
+## 📄 Related
+
+- **Dream-LoRA (improved)**: [dexmac/progressive-cognitive-dream-lora](https://huggingface.co/dexmac/progressive-cognitive-dream-lora)
+- **GitHub**: [dexmac221/progressive-cognitive](https://github.com/dexmac221/progressive-cognitive)
+
+## 📜 License
+
+Apache 2.0