config.json

{
  "_name_or_path": "Unbabel/XLM-R-19L",
  "architectures": [
    "XLMRobertaModel"
  ],
  "attention_probs_dropout_prob": 0.1,
  "bos_token_id": 0,
  "classifier_dropout": null,
  "eos_token_id": 2,
  "gradient_checkpointing": false,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,
  "hidden_size": 1024,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "layer_norm_eps": 1e-05,
  "max_position_embeddings": 514,
  "model_type": "xlm-roberta",
  "num_attention_heads": 16,
  "num_hidden_layers": 19,
  "output_past": true,
  "pad_token_id": 1,
  "position_embedding_type": "absolute",
  "pruned_heads": {
    "0": [
      12,
      5,
      6,
      14
    ],
    "1": [
      13,
      12,
      5,
      15
    ],
    "2": [
      8,
      9,
      10,
      13
    ],
    "3": [
      9,
      10,
      4,
      12
    ],
    "4": [
      8,
      9,
      3,
      13
    ],
    "5": [
      2,
      3,
      4,
      12
    ],
    "6": [
      0,
      9,
      3,
      12
    ],
    "7": [
      0,
      9,
      2,
      15
    ],
    "8": [
      0,
      8,
      3,
      13
    ],
    "9": [
      0,
      15,
      14,
      7
    ],
    "10": [
      0,
      1,
      11,
      12
    ],
    "11": [
      11,
      15,
      6,
      7
    ],
    "12": [
      8,
      10,
      3,
      14
    ],
    "13": [
      8,
      2,
      4,
      5
    ],
    "14": [
      0,
      9,
      11,
      14
    ],
    "15": [
      8,
      3,
      4,
      6
    ],
    "16": [
      8,
      10,
      2,
      3
    ],
    "17": [
      0,
      8,
      12,
      6
    ],
    "18": [
      0,
      8,
      15,
      7
    ]
  },
  "torch_dtype": "float32",
  "transformers_version": "4.10.3",
  "type_vocab_size": 1,
  "use_cache": true,
  "vocab_size": 250002
}