From 7f427a469b40de3991a768bdd535f04206f055fb Mon Sep 17 00:00:00 2001 From: Pablo Gonzalez Date: Wed, 27 Nov 2024 12:23:10 -0500 Subject: [PATCH] Host dataset + add instructions to download it --- language/llama3-405b/README.md | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/language/llama3-405b/README.md b/language/llama3-405b/README.md index c3359fa365..9f2eff3cab 100644 --- a/language/llama3-405b/README.md +++ b/language/llama3-405b/README.md @@ -93,7 +93,15 @@ rclone config create mlc-inference s3 provider=Cloudflare access_key_id=f65ba5ee ``` You can then navigate in the terminal to your desired download directory and run the following command to download the dataset: -**TODO: Host dataset and grant access to submitters** +``` +rclone copy mlc-inference:mlcommons-inference-wg-public/llama3_405b/mlperf_llama3.1_405b_dataset_8313_processed_fp16_eval.pkl ./ -P +``` + +You can also download the calibration dataset from the Cloudflare R2 bucket by running the following command: + +``` +rclone copy mlc-inference:mlcommons-inference-wg-public/llama3_405b/mlperf_llama3.1_405b_calibration_dataset_512_processed_fp16_eval.pkl ./ -P +``` ## Run Performance Benchmarks