config.json

{
  "_name_or_path": "t5-large",
  "architectures": [
    "EffT5ForConditionalGeneration"
  ],
  "copy_skipped_hidden_states": true,
  "d_ff": 4096,
  "d_kv": 64,
  "d_model": 1024,
  "decoder_start_token_id": 0,
  "dense_act_fn": "relu",
  "distill_alpha": null,
  "distill_layer_alpha": null,
  "distill_layer_map": "dyna",
  "distill_temp": null,
  "do_layer_distill": false,
  "do_layer_transformation": false,
  "dropout_rate": 0.1,
  "eos_token_id": 1,
  "exit_conf_threshold": 1.0,
  "exit_conf_type": null,
  "exit_position_temp": null,
  "fallback_conf_threshold": 1.0,
  "fallback_conf_type": null,
  "feed_forward_proj": "relu",
  "initializer_factor": 1.0,
  "intermediate_loss_fn": "weighted_ce",
  "is_encoder_decoder": true,
  "is_gated_act": false,
  "layer_norm_epsilon": 1e-06,
  "max_answer_length": 30,
  "model_type": "t5",
  "n_positions": 512,
  "num_decoder_layers": 24,
  "num_heads": 16,
  "num_layers": 24,
  "output_hidden_states_decoder": true,
  "output_past": true,
  "pad_token_id": 0,
  "parallel_gen_token": false,
  "relative_attention_max_distance": 128,
  "relative_attention_num_buckets": 32,
  "rollback_conf_threshold": null,
  "small_exit_layer": null,
  "static_exit_layer": null,
  "task_specific_params": {
    "summarization": {
      "early_stopping": true,
      "length_penalty": 2.0,
      "max_length": 200,
      "min_length": 30,
      "no_repeat_ngram_size": 3,
      "num_beams": 4,
      "prefix": "summarize: "
    },
    "translation_en_to_de": {
      "early_stopping": true,
      "max_length": 300,
      "num_beams": 4,
      "prefix": "translate English to German: "
    },
    "translation_en_to_fr": {
      "early_stopping": true,
      "max_length": 300,
      "num_beams": 4,
      "prefix": "translate English to French: "
    },
    "translation_en_to_ro": {
      "early_stopping": true,
      "max_length": 300,
      "num_beams": 4,
      "prefix": "translate English to Romanian: "
    }
  },
  "torch_dtype": "float32",
  "train_meta_cm_head": false,
  "transformers_version": "4.27.4",
  "use_big_small": false,
  "use_cache": true,
  "vocab_size": 32128
}