NeMo Framework Megatron Backend

NVIDIA

Container

NVIDIA

NeMo Framework Megatron Backend

NVIDIA NeMo™ framework Megatron backend supports pre-training, post-training, and reinforcement learning of LLMs and multi-modal generative AI models with state-of-the-art data processing, model training techniques, and flexible deployment options.

NVIDIA AI Enterprise Supported

Layer	Label		Created
sha256:036716dc523fe81bfd7784cd11d65b80c0b9e9ddb675eacce1ed4bf0f043c1ff	RUN	`python -c "from transformers import AutoTokenizer; tok_llama3_8b=AutoTokenizer.from_pretrained('meta-llama/Meta-Llama-3.1-8B'); tok_llama3_70b=AutoTokenizer.from_pretrained('meta-llama/Meta-Llama-3.1-8B-instruct'); tok_llama3_8b_inst=AutoTokenizer.from_pretrained('meta-llama/Meta-Llama-3.1-70B'); tok_llama3_70b_inst=AutoTokenizer.from_pretrained('meta-llama/Meta-Llama-3.1-70B-instruct');"`	07/26/2024 5:12 PM UTC
sha256:d07d47f36a273b5f4bfc13f56b94719b014207864df358461566600063689059	RUN	`python -c "from transformers import AutoTokenizer; tok_llama3_8b=AutoTokenizer.from_pretrained('meta-llama/Meta-Llama-3-8B'); tok_llama3_70b=AutoTokenizer.from_pretrained('meta-llama/Meta-Llama-3-70B'); tok_llama3_8b_inst=AutoTokenizer.from_pretrained('meta-llama/Meta-Llama-3-8B-instruct'); tok_llama3_70b_inst=AutoTokenizer.from_pretrained('meta-llama/Meta-Llama-3-70B-instruct');"`	07/26/2024 5:12 PM UTC
sha256:59b0541e212b07d01f583693f7f684f6b8106bb72cf3a7962fa068a2c95f3de8	RUN	`RUN /bin/sh -c <<EOFINNER bash -eux rsync -ahP /mnt/tokenizer-dir/ ~/.cache/huggingface/hub/ EOFINNER # buildkit`	07/26/2024 5:12 PM UTC
sha256:dd30817144e824629e08a45be7d41fdb34ea216c8865a6da8a95822ec97f5449	RUN	`python -c "from transformers import AutoTokenizer; tok_llama3_8b=AutoTokenizer.from_pretrained('meta-llama/Meta-Llama-3-8B'); tok_llama3_70b=AutoTokenizer.from_pretrained('meta-llama/Meta-Llama-3-70B'); tok_llama3_8b_inst=AutoTokenizer.from_pretrained('meta-llama/Meta-Llama-3-8B-instruct'); tok_llama3_70b_inst=AutoTokenizer.from_pretrained('meta-llama/Meta-Llama-3-70B-instruct');"`	07/26/2024 12:33 AM UTC
sha256:fd1dd59f52fe6734e5c15d32495b6052a3cc36ff5f8883da7c5de75e4b314d0c	RUN	`RUN /bin/sh -c <<EOFINNER bash -eux cd /mnt/tokenizer-dir mkdir -p /tmp_assets tar -xzvf llama3_8b_tokenizer.tar.gz -C /tmp_assets tar -xzvf llama3_70b_tokenizer.tar.gz -C /tmp_assets cp -rf /tmp_assets/hub ~/.cache/huggingface rm -rf /tmp_assets EOFINNER # buildkit`	07/26/2024 12:33 AM UTC
sha256:d1e317d93f348137496390ccf994016b75a0bb42d2f237c6a81cc95edd9c66fe	RUN	`RUN /bin/sh -c <<EOFINNER bash -eux git config --global user.email "nemo@nvidia.com" git config --global user.name "NeMo Framework" pushd /opt/NeMo-Aligner git am --3way <./aligner-llama3.1.patch git am --3way <./aligner-calculator.patch rm -f ./aligner-llama3.1.patch ./aligner-calculator.patch popd pushd /opt/megatron-lm git am --3way <./mcore-calculator.patch rm -f ./mcore-calculator.patch popd git config --global --unset user.email git config --global --unset user.name EOFINNER # buildkit`	07/26/2024 12:33 AM UTC
sha256:52f81e7c8f4b1112b5a6838fa3d1aa3dd2d424e6472493153a2b0c690b84057a	COPY	`././mcore-calculator.patch /opt/megatron-lm`	07/26/2024 12:33 AM UTC
sha256:af1de615dfc47fb39151f64b74bb5fcf0a135be5fbc6ed08b48de2d1c1a96b42	COPY	`././aligner-calculator.patch /opt/NeMo-Aligner`	07/26/2024 12:09 AM UTC
sha256:57c59fcf03e7d0a469615c11570e1c35b87aa5416629871657978858c0a89632	COPY	`././aligner-llama3.1.patch /opt/NeMo-Aligner`	07/25/2024 10:40 PM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4	LABEL	`org.opencontainers.image.title=nvcr.io/nvidian/nemo:24.05.llama3.1_nc`	07/24/2024 5:19 PM UTC