414f6a1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
{ "architectures": [ "Lake1ForAnyToAny" ], "vocab_size": 20064, "hidden_size": 512, "num_hidden_layers": 112, "num_attention_heads": 64, "embd_pdrop": 0.0, "use_moe": true, "num_experts": 24, "expert_hidden_size": 124, "dropout": 0.0 }