m-a-p/340M-20B-GatedDeltaNet-hybrid-3-1
0.4B
•
Updated
•
5
Hybrid Linear Attention Research
All 1.3B & 340M hybrid linear-attention experiments.
-
-
m-a-p/340M-20B-GatedDeltaNet-hybrid-6-1
0.4B • Updated • 6 -
m-a-p/340M-20B-GLA-hybrid-6-1
0.3B • Updated • 6 -
m-a-p/340M-20B-GLA-hybrid-3-1
0.3B • Updated • 6 -
m-a-p/340M-20B-GLA-hybrid-12-1
0.3B • Updated • 6 -
m-a-p/340M-20B-GLA-hybrid-24-1
0.3B • Updated • 6 -
m-a-p/340M-20B-HGRN2-pure-baseline
0.3B • Updated • 7 -
m-a-p/340M-20B-HGRN2-hybrid-6-1
0.3B • Updated • 3 -
m-a-p/340M-20B-HGRN2-hybrid-3-1
0.3B • Updated • 3 -
m-a-p/340M-20B-HGRN2-hybrid-12-1
0.3B • Updated • 2 -
m-a-p/340M-20B-HGRN2-hybrid-24-1
0.3B • Updated • 3 -
m-a-p/340M-20B-DeltaNet-hybrid-3-1
0.3B • Updated • 3 -
m-a-p/340M-20B-DeltaNet-pure
0.3B • Updated • 11 -
m-a-p/340M-20B-DeltaNet-hybrid-6-1
0.3B • Updated • 7 -
m-a-p/340M-20B-DeltaNet-hybrid-12-1
0.3B • Updated • 6 -
m-a-p/340M-20B-DeltaNet-hybrid-24-1
0.3B • Updated • 4 -
m-a-p/340M-20B-RetNet-hybrid-24-1
0.3B • Updated • 4 -
m-a-p/340M-20B-RetNet-hybrid-12-1
0.3B • Updated • 4 -
m-a-p/340M-20B-RetNet-hybrid-6-1
0.3B • Updated • 3 -
m-a-p/340M-20B-RetNet-hybrid-3-1
0.3B • Updated • 6 -
m-a-p/340M-20B-RetNet-pure
0.3B • Updated • 7 -
m-a-p/1.3B-100B-GatedDeltaNet-hybrid-3-1
1B • Updated • 23 -
m-a-p/1.3B-100B-GatedDeltaNet-hybrid-6-1
1B • Updated • 5 -
m-a-p/1.3B-100B-GatedDeltaNet-hybrid-12-1
1B • Updated • 3 -
m-a-p/1.3B-100B-GatedDeltaNet-hybrid-24-1
1B • Updated • 4 -
m-a-p/1.3B-100B-GLA-hybrid-3-1
1B • Updated • 4 -
m-a-p/1.3B-100B-GLA-hybrid-6-1
1B • Updated • 3 -
m-a-p/1.3B-100B-GLA-hybrid-12-1
1B • Updated • 3 -
m-a-p/1.3B-100B-GLA-hybrid-24-1
1B • Updated • 3 -
m-a-p/1.3B-100B-HGRN2-hybrid-3-1
1B • Updated • 3 -
m-a-p/1.3B-100B-HGRN2-hybrid-6-1
1B • Updated • 5 -
m-a-p/1.3B-100B-HGRN2-hybrid-12-1
1B • Updated • 5 -
m-a-p/1.3B-100B-HGRN2-hybrid-24-1
1B • Updated • 4 -
m-a-p/1.3B-100B-HGRN-hybrid-3-1
1B • Updated • 25 -
m-a-p/1.3B-100B-HGRN-hybrid-6-1
1B • Updated • 6 -
m-a-p/1.3B-100B-HGRN-hybrid-12-1
1B • Updated • 3 -
m-a-p/1.3B-100B-HGRN-hybrid-24-1
1B • Updated • 3 -
m-a-p/1.3B-100B-HGRN-pure
1B • Updated • 25 -
m-a-p/340M-20B-HGRN-hybrid-3-1
0.3B • Updated • 6 -
m-a-p/340M-20B-HGRN-hybrid-6-1
0.3B • Updated • 3 -
m-a-p/340M-20B-HGRN-hybrid-12-1
0.3B • Updated • 3 -
m-a-p/340M-20B-HGRN-hybrid-24-1
0.3B • Updated • 3 -
m-a-p/340M-20B-HGRN-pure-baseline
0.3B • Updated • 3 -
m-a-p/1.3B-100B-RetNet-hybrid-3-1
1B • Updated • 3 -
m-a-p/1.3B-100B-RetNet-hybrid-6-1
1B • Updated • 7 -
m-a-p/1.3B-100B-RetNet-hybrid-12-1
1B • Updated • 3 -
m-a-p/1.3B-100B-RetNet-hybrid-24-1
1B • Updated • 3 -
m-a-p/1.3B-100B-RetNet-pure
1B • Updated • 3 -
m-a-p/1.3B-100B-GLA-pure
1B • Updated • 3 -
m-a-p/1.3B-100B-HGRN2-pure
1B • Updated • 3 -
m-a-p/1.3B-100B-GatedDeltaNet-pure
1B • Updated • 94 -
m-a-p/340M-20B-GLA-pure-baseline
0.3B • Updated • 5 -
m-a-p/340M-20B-GatedDeltaNet-hybrid-12-1
0.4B • Updated • 3 -
m-a-p/1.3B-100B-DeltaNet-hybrid-3-1
1B • Updated • 3 -
m-a-p/1.3B-100B-DeltaNet-hybrid-6-1
1B • Updated • 7 -
m-a-p/1.3B-100B-DeltaNet-hybrid-12-1
1B • Updated • 3 -
m-a-p/1.3B-100B-DeltaNet-hybrid-24-1
1B • Updated • 5 -
m-a-p/1.3B-100B-DeltaNet-pure
1B • Updated • 4 -
m-a-p/340M-20B-GatedDeltaNet-pure-baseline
0.4B • Updated • 7 -
m-a-p/340M-20B-GatedDeltaNet-New-hybrid-24-1
0.4B • Updated • 5