Delta-Vector/Pascal-Base-12B
118
Hparams:
- LR: 1e-5
- Weight Decay: 0.02
- grad-clip: 0.2
- R64, A32, Rslora, 0.1 drop out, 16bit lora (qlora is cope.)
pretrained on a buncha LNs, books, etc. Used for a base for Pascal.
Hparams:
pretrained on a buncha LNs, books, etc. Used for a base for Pascal.