config.json

{
  "_name_or_path": "roberta-test",
  "adaptive": true,
  "architectures": [
    "LSGAlbertForMaskedLM"
  ],
  "attention_probs_dropout_prob": 0,
  "auto_map": {
    "AutoConfig": "modeling_lsg_albert.LSGAlbertConfig",
    "AutoModel": "modeling_lsg_albert.LSGAlbertModel",
    "AutoModelForMaskedLM": "modeling_lsg_albert.LSGAlbertForMaskedLM",
    "AutoModelForMultipleChoice": "modeling_lsg_albert.LSGAlbertForMultipleChoice",
    "AutoModelForPreTraining": "modeling_lsg_albert.LSGAlbertForPreTraining",
    "AutoModelForQuestionAnswering": "modeling_lsg_albert.LSGAlbertForQuestionAnswering",
    "AutoModelForSequenceClassification": "modeling_lsg_albert.LSGAlbertForSequenceClassification",
    "AutoModelForTokenClassification": "modeling_lsg_albert.LSGAlbertForTokenClassification"
  },
  "base_model_prefix": "lsg",
  "block_size": 128,
  "bos_token_id": 2,
  "classifier_dropout_prob": 0.1,
  "down_scale_factor": 1,
  "embedding_size": 128,
  "eos_token_id": 3,
  "gap_size": 0,
  "hidden_act": "gelu_new",
  "hidden_dropout_prob": 0,
  "hidden_size": 768,
  "initializer_range": 0.02,
  "inner_group_num": 1,
  "intermediate_size": 3072,
  "layer_norm_eps": 1e-12,
  "lsh_num_pre_rounds": 1,
  "mask_first_token": true,
  "max_position_embeddings": 4096,
  "model_type": "albert",
  "net_structure_type": 0,
  "num_attention_heads": 12,
  "num_global_tokens": 1,
  "num_hidden_groups": 1,
  "num_hidden_layers": 12,
  "num_memory_blocks": 0,
  "pad_token_id": 0,
  "pool_with_global": true,
  "position_embedding_type": "absolute",
  "sparse_block_size": 128,
  "sparsity_factor": 2,
  "sparsity_type": "norm",
  "torch_dtype": "float32",
  "transformers_version": "4.20.1",
  "type_vocab_size": 2,
  "vocab_size": 30000
}