!ModelConfig
config_data: !DataConfig
  data_statistics: !DataStatistics
    average_len_target_per_bucket:
    - 5.996365000000221
    buckets:
    - !!python/tuple
      - 10
      - 10
    length_ratio_mean: 1.0
    length_ratio_stats_per_bucket:
    - !!python/tuple
      - 1.0
      - 0.0
    length_ratio_std: 0.0
    max_observed_len_source: 10
    max_observed_len_target: 10
    num_discarded: 0
    num_sents: 1000000
    num_sents_per_bucket:
    - 1000000
    num_tokens_source: 5996365
    num_tokens_target: 5996365
    num_unks_source: 0
    num_unks_target: 0
    size_vocab_source: 16
    size_vocab_target: 16
  max_seq_len_source: 10
  max_seq_len_target: 10
  num_source_factors: 2
  num_target_factors: 2
config_decoder: !TransformerConfig
  act_type: relu
  attention_heads: 4
  decoder_type: ssru_transformer
  depth_key_value: 16
  dropout_act: 0.1
  dropout_attention: 0.1
  dropout_prepost: 0.1
  feed_forward_num_hidden: 16
  max_seq_len_source: 10
  max_seq_len_target: 10
  model_size: 16
  num_layers: 1
  positional_embedding_type: fixed
  postprocess_sequence: dr
  preprocess_sequence: n
  use_glu: false
  use_lhuc: false
config_embed_source: !EmbeddingConfig
  allow_sparse_grad: false
  dropout: 0.0
  factor_configs:
  - !FactorConfig
    combine: sum
    num_embed: 16
    share_embedding: false
    vocab_size: 8
  num_embed: 16
  num_factors: 2
  vocab_size: 16
config_embed_target: !EmbeddingConfig
  allow_sparse_grad: false
  dropout: 0.0
  factor_configs:
  - !FactorConfig
    combine: sum
    num_embed: 16
    share_embedding: false
    vocab_size: 8
  num_embed: 16
  num_factors: 2
  vocab_size: 16
config_encoder: !TransformerConfig
  act_type: relu
  attention_heads: 4
  decoder_type: ssru_transformer
  depth_key_value: 16
  dropout_act: 0.1
  dropout_attention: 0.1
  dropout_prepost: 0.1
  feed_forward_num_hidden: 16
  max_seq_len_source: 10
  max_seq_len_target: 10
  model_size: 16
  num_layers: 1
  positional_embedding_type: fixed
  postprocess_sequence: dr
  preprocess_sequence: n
  use_glu: false
  use_lhuc: false
config_length_task: null
dtype: float32
lhuc: false
neural_vocab_selection: null
neural_vocab_selection_block_loss: false
vocab_source_size: 16
vocab_target_size: 16
weight_tying_type: src_trg_softmax
