Update README.md

README.md

@@ -7,7 +7,7 @@ datasets:
 
 This model experiment was inspired by the work published in [Goat: Fine-tuned LLaMA Outperforms GPT-4 on Arithmetic Tasks](https://arxiv.org/pdf/2305.14201.pdf), which found good success in fine tuning Llama models on math.
 
-Fine tuning of [philschmid/Llama-2-7b-hf](https://huggingface.co/philschmid/Llama-2-7b-hf) was conducted with 5.8M math problems from the [AtlasUnified/atlas-math-sets](https://huggingface.co/datasets/AtlasUnified/atlas-math-sets) dataset.  
+Fine tuning of [philschmid/Llama-2-7b-hf](https://huggingface.co/philschmid/Llama-2-7b-hf) was conducted with 2.8M math problems from the [AtlasUnified/atlas-math-sets](https://huggingface.co/datasets/AtlasUnified/atlas-math-sets) dataset.  
 
 Training was conducted on a trn1.32xlarge instance. The model here was complied for 2 Neuron cores, which will run on AWS inf2.8xlarge and larger instances.