Views
No views yet
1config = { 2 "d_model": 1180, 3 "n_rwkv_layers": 16, 4 "n_attn_layers": 4, 5 "n_heads": 10, 6 "seq_len": 1024, 7 "batch_size": 4, 8 "accum_steps": 8, 9 "lr": 4e-4, 10}