Update README.md

README.md

@@ -1,57 +1,26 @@
 ---
 base_model: LSX-UniWue/LLaMmlein_7B_chat
-library_name: peft
 license: mit
-language:
-- de
 ---
 
-# Model Card for Model ID
+# LLäMmlein QLoRA – Grounding Act Classification
 
-<!-- Provide a quick summary of what the model is/does. -->
+This model is a fine-tuned version of [LSX-UniWue/LLaMmlein_7B_chat](https://huggingface.co/LSX-UniWue/LLaMmlein_7B_chat), optimized using QLoRA  for efficient binary classification of German dialogue utterances into:
 
-
-
-## Model Details
-
-
-
-### Model Description
-
-<!-- Provide a longer summary of what this model is. -->
-
-# QLoRA Fine-Tuned Model: Advance vs. Non-Advance Classification
-
-This model is a fine-tuned version of [LSX-UniWue/LLaMmlein_7B_chat](https://huggingface.co/LSX-UniWue/LLaMmlein_7B_chat) using QLoRA and a classification head.
-
-It was trained to classify short utterances (e.g., student or teacher dialogue) into two categories:
-- **advance**: utterances that move the conversation forward (e.g., answering, explaining)
-- **non_advance**: utterances that do not move the conversation forward (e.g., hesitations, misunderstandings)
-
----
-
-## 🔧 Fine-tuning Details
-
-- **Base model**: LSX-UniWue/LLaMmlein_7B_chat
-- **Method**: QLoRA (4-bit quantization, LoRA adapters)
-- **Task**: Sequence Classification (2 classes)
-- **Training dataset size**: 1200 examples
-- **Validation dataset size**: 120 examples
-- **Training epochs**: 2
-- **Learning rate**: 2e-4
-- **Weight decay**: 0.01
+- **advance**: Contribution that moves the dialogue forward (e.g. confirmations, follow-ups, elaborations)
+- **non_advance**: Other utterances (e.g. vague responses, misunderstandings, irrelevant comments)
 
 ---
 
-## 📊 Performance
+## Use Cases
 
-- Final training loss: ~0.10
-- Final validation loss: ~0.35
-- No overfitting observed during training
+- Dialogue system analysis
+- Teacher-student interaction classification
+- Grounding in institutional advising or classroom discourse
 
 ---
 
-## 📚 How to Use
+## How to Use
 
 ```python
 from transformers import AutoModelForSequenceClassification, AutoTokenizer