Update README.md

README.md

@@ -11,12 +11,12 @@ tags:
 
 # Model Card for Model ID
 
-A fine-tuned Llama-3.1-8B, designed. 
+A fine-tuned version of Llama-3.1-8B, designed to generate Angry Birds levels based on simple text descriptions. The model is trained using the Unsloth library and is optimized to produce game-level designs that can be directly imported into the official Angry Birds game.
 ## Model Details
 
 ### Model Description
 
-<!-- Provide a longer summary of what this model is. -->
+This model can be used to generate new levels for the Angry Birds game using simple text inputs. Users can describe elements like "a tall tower made of wood with a pig on top" or "multiple structures with TNT boxes and glass blocks," and the model will create a level design matching the description.
 
 
 
@@ -24,7 +24,7 @@ A fine-tuned Llama-3.1-8B, designed.
 - **Model type:** Text Generation
 - **Language(s) (NLP):** English
 - **License:** Apache license 2.0
-- **Finetuned from model Llama 3.1 8B:** trained with unsloth.
+- **Finetuned from:** Meta-Llama-3.1-8B using the Unsloth library
 
 ### Model Sources [optional]
 
@@ -78,9 +78,15 @@ Use the code below to get started with the model.
 
 ### Training Data
 
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+The model was fine-tuned using a diverse dataset that includes:
+
+- Existing Angry Birds levels, descriptions, and user-generated content to capture the typical structure, patterns, and elements of the game.
+- Additional levels created by us to introduce new variations and elements not found in the original dataset.
+- Data augmentation techniques were employed, such as creating variations of the same level with different bird types (e.g., red birds and then yellow birds), to enhance the diversity of the generated outputs.
+
+https://huggingface.co/datasets/raccoote/angry-birds-levels
+
 
-[More Information Needed]
 
 ### Training Procedure