Revise README for llama3-1-70b pretraining

Updated README to reflect specific workload details for llama3-1-70b.

Author: incredere

training/a4/llama3-1-70b/nemo-pretraining-gke/32node-bf16-seq8192-gbs2048/recipe/README.md

@@ -1,7 +1,7 @@
 <!-- mdformat global-off -->
-# Pretrain $USER-a4-llama3-1-70b workloads on a4 GKE Node pools with Nvidia NeMo Framework
+# Pretrain llama3-1-70b-seq8192-gbs2048-mbs1-gpus256 workloads on a4 GKE Node pools with Nvidia NeMo Framework
 
-This recipe outlines the steps for running a $USER-a4-llama3-1-70b pretraining
+This recipe outlines the steps for running a llama3-1-70b-seq8192-gbs2048-mbs1-gpus256 pretraining
 workload on [a4 GKE Node pools](https://cloud.google.com/kubernetes-engine) by using the
 [NVIDIA NeMo framework](https://github.com/NVIDIA/nemo).