Vanilla and BLD KDA hybrid language models from Competing Learning in Hybrid Architectures (250M–1B parameters).
Xingyu Shen
xingyushenai
AI & ML interests
None yet
Recent Activity
upvoted a collection 3 days ago
Competing Learning Models liked a model 11 days ago
xingyushenai/CL-BLD-1B liked a model 11 days ago
xingyushenai/CL-Vanilla-750MOrganizations
None yet
models 10
xingyushenai/CL-Vanilla-1B
Text Generation • 1B • Updated • 56 • 1
xingyushenai/CL-BLD-1B
Text Generation • 1B • Updated • 66 • 2
xingyushenai/CL-Vanilla-750M
Text Generation • 0.8B • Updated • 72 • 2
xingyushenai/CL-BLD-750M
Text Generation • 0.8B • Updated • 54 • 1
xingyushenai/CL-Vanilla-390M
Text Generation • 0.4B • Updated • 44
xingyushenai/CL-BLD-510M
Text Generation • 0.5B • Updated • 52
xingyushenai/CL-Vanilla-510M
Text Generation • 0.5B • Updated • 49
xingyushenai/CL-Vanilla-250M
Text Generation • 0.3B • Updated • 49
xingyushenai/CL-BLD-250M
Text Generation • 0.3B • Updated • 43
xingyushenai/CL-BLD-390M
Text Generation • 0.4B • Updated • 47
datasets 0
None public yet