config.json

{
  "latent_size":256,
  "seed": 42, 
  "ratio_increase": 0.5, 
  "ratio_zero": 0.05, 
  "dim_target_kl": 1.0, 
  "length_weighted_loss": true, 
  "beta_start": 0.0001, 
  "beta_m": 0.1, 
  "lambda_adv": 0.0, 
  "fb_mode": 1, 
  "mlm_probability": 0.15, 
  "drop_prob": 0.0, 
  "sub_prob": 0.0, 
  "shuffle_dist": 0, 
  "zeta": 0.0, 
  "learning_rate": 5e-05, 
  "weight_decay": 0.1, 
  "adam_epsilon": 1e-08, 
  "max_grad_norm": 1.0, 
  "warmup": 0.01, 
  "temperature": 1.0, 
  "top_k": 0, 
  "repetition_penalty": 2.0, 
  "top_p": 0.9, 
  "std_scale": 1.5, 
  "max_out_length": 100, 
  "logger": true, 
  "checkpoint_callback": null,
  "enable_checkpointing": true, 
  "process_position": 0,
  "encoder":{
  "architectures": [
    "BertForMaskedLM"
  ],
  "attention_probs_dropout_prob": 0.1,
  "directionality": "bidi",
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,
  "hidden_size": 768,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "layer_norm_eps": 1e-12,
  "max_position_embeddings": 512,
  "model_type": "bert",
  "num_attention_heads": 12,
  "num_hidden_layers": 12,
  "pad_token_id": 0,
  "pooler_fc_size": 768,
  "pooler_num_attention_heads": 12,
  "pooler_num_fc_layers": 3,
  "pooler_size_per_head": 128,
  "pooler_type": "first_token_transform",
  "type_vocab_size": 2,
  "vocab_size": 21128
},
  "decoder":{
    "num_layers":32,
    "vocab_size":50048,
    "hidden_size":1600,
    "num_attention_heads":25,
    "embedding_dropout_prob":0.1,
    "attention_dropout_prob":0.1,
    "output_dropout_prob":0.1,
    "max_sequence_length":512,
    "max_memory_length":512,
    "latent_size":256,
    "checkpoint_activations":false,
    "checkpoint_num_layers":1,
    "parallel_output":true,
    "relative_encoding":true
}
}