juancopi81's picture
Add t5x and mt3 models
b100e1c
# T5.1.0 11B model.
include 't5x/examples/t5/t5_1_0/base.gin' # imports vocab, optimizer and model.
# ------------------- Network specification overrides --------------------------
network.Transformer.config = @network.T5Config()
network.T5Config:
emb_dim = 1024
num_heads = 128
num_encoder_layers = 24
num_decoder_layers = 24
head_dim = 128
mlp_dim = 65536