config.json

{
  "_name_or_path": "google/mt5-base",
  "architectures": [
    "MT5ForConditionalGeneration"
  ],
  "d_ff": 2048,
  "d_kv": 64,
  "d_model": 768,
  "decoder_start_token_id": 0,
  "dense_act_fn": "gelu_new",
  "dropout_rate": 0.1,
  "eos_token_id": 1,
  "feed_forward_proj": "gated-gelu",
  "initializer_factor": 1.0,
  "is_encoder_decoder": true,
  "is_gated_act": true,
  "label2id": {
    "nc": {
      "autos": 9.0,
      "entertainment": 6.0,
      "finance": 3.0,
      "foodanddrink": 0.0,
      "health": 7.0,
      "lifestyle": 4.0,
      "news": 5.0,
      "sports": 1.0,
      "travel": 2.0,
      "video": 8.0
    },
    "paws-x": {
      "different": 0.0,
      "same": 1.0
    },
    "qadsm": {
      "Bad": 0.0,
      "Good": 1.0
    },
    "qam": {
      "False": 0.0,
      "True": 1.0
    },
    "wpr": {
      "Bad": 0.0,
      "Excellent": 3.0,
      "Fair": 1.0,
      "Good": 2.0,
      "Perfect": 4.0
    },
    "xnli": {
      "contradiction": 2.0,
      "entailment": 0.0,
      "neutral": 1.0
    }
  },
  "layer_norm_epsilon": 1e-06,
  "model_type": "mt5",
  "num_accum_batches": 4,
  "num_decoder_layers": 12,
  "num_heads": 12,
  "num_layers": 12,
  "output_past": true,
  "pad_token_id": 0,
  "r3f": true,
  "r4f": false,
  "relative_attention_max_distance": 128,
  "relative_attention_num_buckets": 32,
  "retain_gradients": false,
  "scale_loss": true,
  "tasks": [
    [
      "xnli",
      3
    ],
    [
      "paws-x",
      2
    ],
    [
      "nc",
      10
    ],
    [
      "qadsm",
      2
    ],
    [
      "wpr",
      5
    ],
    [
      "qam",
      2
    ]
  ],
  "tie_word_embeddings": false,
  "tokenizer_class": "T5Tokenizer",
  "torch_dtype": "float32",
  "transformers_version": "4.28.1",
  "use_cache": true,
  "vocab_size": 250112
}