|
{ |
|
"_name_or_path": "./", |
|
"architectures": [ |
|
"T5ForConditionalGeneration" |
|
], |
|
"d_ff": 2560, |
|
"d_kv": 64, |
|
"d_model": 768, |
|
"decoder_start_token_id": 0, |
|
"dropout_rate": 0.0, |
|
"eos_token_id": 1, |
|
"feed_forward_proj": "gated-gelu", |
|
"initializer_factor": 0.57, |
|
"is_encoder_decoder": true, |
|
"layer_norm_epsilon": 1e-06, |
|
"model_type": "t5", |
|
"num_decoder_layers": 36, |
|
"num_heads": 12, |
|
"num_layers": 36, |
|
"output_past": true, |
|
"pad_token_id": 0, |
|
"relative_attention_max_distance": 128, |
|
"relative_attention_num_buckets": 32, |
|
"tie_word_embeddings": false, |
|
"torch_dtype": "float32", |
|
"transformers_version": "4.18.0", |
|
"use_cache": true, |
|
"vocab_size": 32103, |
|
"early_stopping": true, |
|
"max_length": 128, |
|
"num_beams": 4, |
|
"prefix": "translate English to Dutch: ", |
|
"task_specific_params": { |
|
"translation_en_to_nl": { |
|
"early_stopping": true, |
|
"max_length": 128, |
|
"num_beams": 4, |
|
"prefix": "translate English to Dutch: " |
|
}, |
|
"translation_nl_to_en": { |
|
"early_stopping": true, |
|
"max_length": 128, |
|
"num_beams": 4, |
|
"prefix": "translate Dutch to English: " |
|
} |
|
} |
|
} |