Parc Models Pythia 160M, Mamba 130M, and RWKV 169M models trained on OpenWebText for 4000 steps (context window: 1024; effective batch size: 512). 6 seeds each. jmichaelov/parc-pythia-seed0 Text Generation • 0.2B • Updated Sep 27, 2025 • 970 jmichaelov/parc-pythia-seed1 Text Generation • 0.2B • Updated Sep 27, 2025 • 885 jmichaelov/parc-pythia-seed2 Text Generation • 0.2B • Updated Sep 27, 2025 • 757 jmichaelov/parc-pythia-seed3 Text Generation • 0.2B • Updated Sep 27, 2025 • 652
Parc Models Pythia 160M, Mamba 130M, and RWKV 169M models trained on OpenWebText for 4000 steps (context window: 1024; effective batch size: 512). 6 seeds each. jmichaelov/parc-pythia-seed0 Text Generation • 0.2B • Updated Sep 27, 2025 • 970 jmichaelov/parc-pythia-seed1 Text Generation • 0.2B • Updated Sep 27, 2025 • 885 jmichaelov/parc-pythia-seed2 Text Generation • 0.2B • Updated Sep 27, 2025 • 757 jmichaelov/parc-pythia-seed3 Text Generation • 0.2B • Updated Sep 27, 2025 • 652