HackerTrans
TopNewTrendsCommentsPastAskShowJobs

opcode84

no profile record

Submissions

Nvidia trains 10T model in 4 bit precision (NVFP4)

developer.nvidia.com
9 points·by opcode84·11 เดือนที่ผ่านมา·3 comments

Nvidia announces 4-bit training with NVFP4

developer.nvidia.com
2 points·by opcode84·11 เดือนที่ผ่านมา·1 comments

Blackwell Brings Native Support for NVFP4: 4-Bit Inference with High Accuracy

developer.nvidia.com
1 points·by opcode84·ปีที่แล้ว·1 comments

New microscaling data formats supported in Blackwell and Triton

triton-lang.org
3 points·by opcode84·ปีที่แล้ว·1 comments

NVIDIA confirms Blackwell support for OCP microscaling formats MXFP8, MXFP4

developer.nvidia.com
4 points·by opcode84·ปีที่แล้ว·1 comments

Standardizing next-generation narrow precision data formats for AI

opencompute.org
103 points·by opcode84·3 ปีที่แล้ว·52 comments

comments

opcode84
·11 เดือนที่ผ่านมา·discuss
For narrow-precision formats to be practical in large-scale pretraining, they must ensure both model accuracy and stable convergence. To assess the viability of 4-bit precision in large-scale model training, experiments were conducted with FP8 and NVFP4 on a 12-billion parameter model based on a combined Mamba-Transformer architecture (12B Hybrid Mamba-Transformer model)—similar to NVIDIA Nemotron Nano 2. This model was trained on a massive dataset of 10 trillion tokens using a phased data-blending approach, switching to a different dataset mix in the second phase of training at 70%, and in the third phase of training at 90% during pretraining.

A version of the 12B Hybrid Mamba-Transformer model was initially trained with 8-bit precision—FP8, which has been shown in previous studies to closely match 16-bit precision, and hence served as our baseline for comparison. We then successfully trained this same 12B model from scratch using NVFP4, demonstrating that this new low-precision format can support full pretraining at trillion-token scale. The NVFP4 run exhibited stable convergence without the training instabilities or divergence issues that typically plague ultra-low precision training.

Figure 3 below shows that NVFP4’s validation loss curve closely matches the loss curves from the higher-precision baseline (i.e., FP8) throughout the entire duration of training. The quantization techniques outlined above ensure that even with aggressive bit-width reduction, the 4-bit pretraining dynamics closely resemble those of higher-precision runs.
opcode84
·11 เดือนที่ผ่านมา·discuss
A version of the 12B Hybrid Mamba-Transformer model was initially trained with 8-bit precision—FP8, which has been shown in previous studies to closely match 16-bit precision, and hence served as our baseline for comparison. We then successfully trained this same 12B model from scratch using NVFP4, demonstrating that this new low-precision format can support full pretraining at trillion-token scale. The NVFP4 run exhibited stable convergence without the training instabilities or divergence issues that typically plague ultra-low precision training.
opcode84
·ปีที่แล้ว·discuss
Introducing NVFP4 for Efficient and Accurate Low-Precision Inference
opcode84
·ปีที่แล้ว·discuss
The formats supported in the tutorial are the OCP microscaling formats, including mxfp4 and mxfp8, as well as NVIDIA’s nvfp4 format. These matrix multiplications are accelerated by fifth generation tensor core instructions on CUDA devices with compute capability 10.

Blog post: https://developer.nvidia.com/blog/openai-triton-on-nvidia-bl...
opcode84
·ปีที่แล้ว·discuss
From the article:

"NVIDIA Blackwell introduces revolutionary block-scaled floating point formats, including the Open Computing Project’s microscaling formats, which Triton now unlocks for NVIDIA Blackwell-powered hardware acceleration.

These formats provide higher average precision at higher performance than the non-native block-scaling techniques emulated frequently in LLM inference projects today.

For OCP format support, MXFP8 GEMMs on Triton showcase exceptional performance similar to the FP8 GEMMs performance accelerated and shown earlier in this post, while natively allowing for scaling in the Tensor Core.

Similarly, MXFP4 provides a new operating point in the precision-performance trade-off space but while offering double the hardware-accelerated performance of FP8 and MXFP8 GEMMs.

While Triton performance for MXFP8 is close to the NVIDIA Blackwell-accelerated FP8 shown earlier, we continue working with the community to accelerate and enable new use cases around block-scaling support."
opcode84
·2 ปีที่แล้ว·discuss
https://arxiv.org/pdf/2310.10537.pdf

Discussed in a previous post https://news.ycombinator.com/item?id=37930663
opcode84
·3 ปีที่แล้ว·discuss
The Microsoft chip has MX data types: https://news.ycombinator.com/item?id=37930663

https://arxiv.org/abs/2310.10537
opcode84
·3 ปีที่แล้ว·discuss
Earlier this year, AMD, Arm, Intel, Meta, Microsoft, NVIDIA, and Qualcomm Technologies, Inc. formed the Microscaling Formats (MX) Alliance with the goal of creating and standardizing next-generation 6- and 4-bit data types for AI training and inferencing. The key enabling technology that enables sub 8-bit formats to work, referred to as microscaling, builds on a foundation of years of design space exploration and research. MX enhances the robustness and ease-of-use of existing 8-bit formats such as FP8 and INT8, thus lowering the barrier for broader adoption of single digit bit training and inference.

Spec: https://www.opencompute.org/documents/ocp-microscaling-forma...

Whitepaper: https://arxiv.org/abs/2310.10537

Code: https://github.com/microsoft/microxcaling