Update README.md

README.md

@@ -1,6 +1,48 @@
+---
+license: apache-2.0
+language:
+  - en
+library_name: transformers
+tags:
+  - llama
+  - llama3
+  - causal-lm
+  - instruction-tuned
+  - hf-internal-testing
+pipeline_tag: text-generation
+---
+
+# 🦙 LLaMA3.2-1B-Instruct
+
+`pAce576/llama3.2-1b-Instruct` is a 1.2 billion parameter language model based on Meta's LLaMA3 architecture. This model has been instruction-tuned for conversational and general-purpose natural language generation tasks.
+
+## 🧠 Model Details
+
+- **Architecture**: LLaMA3.2 (custom 1.2B variant)
+- **Base Model**: LLaMA3-like Transformer
+- **Instruction Tuning**: Yes
+- **Parameters**: ~1.2 billion
+- **Layers**: Custom, designed for efficient inference on resource-constrained environments
+- **Precision**: fp16 supported (also tested with int8/4-bit via quantization)
+
+## 📚 Intended Use
+
+This model is intended for:
+
+- Dialogue generation  
+- Instruction following  
+- Story writing  
+- Light reasoning tasks
+
+**Example usage:**
+
+```python
 from transformers import AutoTokenizer, AutoModelForCausalLM
 
-# Load tokenizer and model
-model_name = "pAce576/llama3.2-1b-Instruct"
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForCausalLM.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained("pAce576/llama3.2-1b-Instruct")
+tokenizer = AutoTokenizer.from_pretrained("pAce576/llama3.2-1b-Instruct")
+
+prompt = "Explain gravity to a 5-year-old."
+inputs = tokenizer(prompt, return_tensors="pt")
+outputs = model.generate(**inputs, max_new_tokens=100)
+print(tokenizer.decode(outputs[0], skip_special_tokens=True))