File size: 363 Bytes
b100e1c
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
# mT5 XL model.

include 't5x/examples/t5/mt5/base.gin'  # imports vocab, optimizer and model.

# ------------------- Network specification overrides --------------------------
network.Transformer.config = @network.T5Config()
network.T5Config:
  emb_dim = 2048
  num_heads = 32
  num_encoder_layers = 24
  num_decoder_layers = 24
  head_dim = 64
  mlp_dim = 5120