Update README.md

README.md

@@ -8,7 +8,7 @@ base_model:
 
 # Gladiator-Mini-exp-1211: A Compact and Powerful Reasoning Engine
 
-**Gladiator-Mini-exp-1211** is a 3-billion parameter language model designed for **complex reasoning tasks**. This experimental model, based on [meta-llama/Llama-3.2-3B-Instruct](link to the base model), offers surprisingly strong analytical capabilities for its size. It demonstrates the potential of smaller models to achieve impressive performance in analytical thinking. We chose to finetune on a Llama model due to finetuning difficulties with the Qwen 2.5 4B model.
+**Gladiator-Mini-exp-1211** is a 3-billion parameter language model designed for **complex reasoning tasks**. This experimental model, based on [meta-llama/Llama-3.2-3B-Instruct] [https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct], offers surprisingly strong analytical capabilities for its size. It demonstrates the potential of smaller models to achieve impressive performance in analytical thinking. We chose to finetune on a Llama model due to finetuning difficulties with the Qwen 2.5 4B model.
 
 **What Makes it Stand Out?**
 
@@ -38,10 +38,14 @@ We encourage you to experiment with Gladiator-Mini-exp-1211, test its limits, an
 
 **Limitations:**
 
-*   Requires system prompts for optimal performance. An incorrect system prompt can lead to an incorrect answer, while a correct prompt usually leads to a correct one.
+*   Requires system prompts for optimal performance. An incorrect system prompt can lead to an incorrect answer, while a correct prompt usually leads to a correct one. An incorrect system prompt example, is using a math reasoning system prompt for a task requiring chain of thought on a tricky text equation.
 *   Reasoning abilities are still being refined.
 *   May exhibit biases or inaccuracies.
 
 **Disclaimer:**
 
-Gladiator-Mini-exp-1211 is an experimental model and should be used with caution. Please critically evaluate its outputs and do not take them as absolute truth.
+Gladiator-Mini-exp-1211 is an experimental model and should be used with caution. Please critically evaluate its outputs and do not take them as absolute truth.
+
+Base model: https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct
+
+Thanks to Meta for the fantastic Llama-3.2-3B model!