JunxiongWang/sftdatasetv3 download history
JunxiongWang/sftdatasetv3 is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 219 times (44 in the last 7 days), and 10,352 times in total. It ranks #62,790 among datasets by monthly downloads.
This is the dataset used in paper, The Mamba in the Llama: Distilling and Accelerating Hybrid Models. @article{junxiongdaniele2024mambainllama, title = {The Mamba in the Llama: Distilling and Accelerating Hybrid Models}, author = {Junxiong Wang and Daniele Paliotta and Avner May and Alexander
Models trained on sftdatasetv3
27 models list it as training data.
- amd/HyLo-Llama-8MLA8GDN-64K-SFT 178 downloads in 30 days
- amd/X-EcoMLA-1B1B-fixed-kv512-DPO 130 downloads in 30 days
- amd/Zebra-Llama-8B-8MLA-24Mamba-SFT 112 downloads in 30 days
- amd/Zebra-Llama-1B-4MLA-12Mamba-SFT 110 downloads in 30 days
- amd/Zebra-Llama-8B-16MLA-16Mamba-SFT 109 downloads in 30 days
- amd/HyLo-Llama-4MLA12GDN-64K-SFT 103 downloads in 30 days
- amd/X-EcoMLA-1B8B-fixed-kv64-DPO 101 downloads in 30 days
- amd/Zebra-Llama-1B-8MLA-8Mamba-SFT 98 downloads in 30 days
- amd/X-EcoMLA-3B3B-dynamic-0.95-DPO 98 downloads in 30 days
- amd/Zebra-Llama-3B-6MLA-22Mamba-SFT 97 downloads in 30 days
- amd/X-EcoMLA-3B3B-fixed-kv816-DPO 90 downloads in 30 days
- amd/X-EcoMLA-1B1B-dynamic-0.95-DPO 86 downloads in 30 days
Open JunxiongWang/sftdatasetv3 on Hugging Face
Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.