config.json

{
  "_name_or_path": "allenai/longformer-base-4096",
  "activation_dropout": 0.0,
  "activation_function": "gelu",
  "architectures": [
    "LEDForConditionalGeneration"
  ],
  "attention_dropout": 0.1,
  "attention_mode": "longformer",
  "attention_window": [
    512,
    512,
    512,
    512,
    512,
    512,
    512,
    512,
    512,
    512,
    512,
    512
  ],
  "bos_token_id": 0,
  "classifier_dropout": 0.0,
  "d_model": 768,
  "decoder_attention_heads": 16,
  "decoder_ffn_dim": 4096,
  "decoder_layerdrop": 0.0,
  "decoder_layers": 12,
  "decoder_start_token_id": 2,
  "dropout": 0.1,
  "encoder_attention_heads": 12,
  "encoder_ffn_dim": 4096,
  "encoder_layerdrop": 0.0,
  "encoder_layers": 12,
  "eos_token_id": 2,
  "gradient_checkpointing": false,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,
  "ignore_attention_mask": false,
  "init_std": 0.02,
  "intermediate_size": 3072,
  "is_encoder_decoder": true,
  "layer_norm_eps": 1e-05,
  "max_decoder_position_embeddings": 1024,
  "max_encoder_position_embeddings": 16384,
  "max_position_embeddings": 4098,
  "model_type": "led",
  "num_hidden_layers": 12,
  "pad_token_id": 1,
  "sep_token_id": 2,
  "torch_dtype": "float32",
  "transformers_version": "4.29.2",
  "type_vocab_size": 1,
  "use_cache": true,
  "vocab_size": 50265
}