From 9e515cea8f7adf12b0d39f875f9da30ad9d2e838 Mon Sep 17 00:00:00 2001 From: Anna Shors <71393111+ashors1@users.noreply.github.com> Date: Fri, 6 Dec 2024 09:02:17 -0800 Subject: [PATCH] docs: fix KD doc rendering (#434) Signed-off-by: ashors1 --- docs/user-guide/knowledge-distillation.rst | 1 + 1 file changed, 1 insertion(+) diff --git a/docs/user-guide/knowledge-distillation.rst b/docs/user-guide/knowledge-distillation.rst index 51387c162..ae0c9d70e 100644 --- a/docs/user-guide/knowledge-distillation.rst +++ b/docs/user-guide/knowledge-distillation.rst @@ -45,6 +45,7 @@ To start, we must first download both the pre-trained student and fine-tuned tea #. Download the `Llama3-8B LLM model and tokenizer `__ into the model's folder. You can use the Hugging Face CLI for this: .. code-block:: bash + huggingface-cli download nvidia/nemotron-3-8b-chat-4k-sft --local-dir teacher_checkpoint After these steps, you should have files ``2b_student.nemo`` and ``teacher_checkpoint/Nemotron-3-8B-Chat-4k-SFT.nemo`` to use in NeMo-Aligner.