Downloads · 30 days
9
28% of all-time downloads
MarioBarbeque/CyberSolve-DeepMind-LinAlg-1D-downsample-v2
CyberSolve-DeepMind-LinAlg-1D-downsample-v2 is a machine learning model from MarioBarbeque. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for transformers.
This model is a zeroth-generation, downsampled training of the CyberSolve LinAlg model. See the model card for the most updated full training of CyberSolve LinAlg here.
Downloads · 30 days
9
28% of all-time downloads
All-time downloads
32
Public
Parameters
783M
3.1 GB on disk
Likes
1
Public
Click a slice to open those files.
.safetensors3.1 GB · 100%
From the Hugging Face model README
This model is a zeroth-generation, downsampled training of the CyberSolve LinAlg model. See the model card for the most updated full training of CyberSolve LinAlg here.
Simulating the larger, full training and evaluation process, we trained and evaluated CyberSolve on a 10% split of the 2M total records available in the 1D Linear Algebra split of the Google DeepMind Mathematics dataset. The results found in this smaller training convinced us that the FLAN-T5 model would indeed learn to effectively solve linear equations. That is, this preliminary training green lighted the full model training for us.