|
{ |
|
"_name_or_path": "xlm-roberta-large", |
|
"architectures": [ |
|
"XLMRobertaForSequenceClassification" |
|
], |
|
"attention_probs_dropout_prob": 0.1, |
|
"bos_token_id": 0, |
|
"classifier_dropout": null, |
|
"eos_token_id": 2, |
|
"hidden_act": "gelu", |
|
"hidden_dropout_prob": 0.1, |
|
"hidden_size": 1024, |
|
"id2label": { |
|
"0": "MT", |
|
"1": "LY", |
|
"2": "SP", |
|
"3": "ID", |
|
"4": "NA", |
|
"5": "HI", |
|
"6": "IN", |
|
"7": "OP", |
|
"8": "IP", |
|
"9": "it", |
|
"10": "ne", |
|
"11": "sr", |
|
"12": "nb", |
|
"13": "re", |
|
"14": "en", |
|
"15": "ra", |
|
"16": "dtp", |
|
"17": "fi", |
|
"18": "lt", |
|
"19": "rv", |
|
"20": "ob", |
|
"21": "rs", |
|
"22": "av", |
|
"23": "ds", |
|
"24": "ed" |
|
}, |
|
"initializer_range": 0.02, |
|
"intermediate_size": 4096, |
|
"label2id": { |
|
"HI": 5, |
|
"ID": 3, |
|
"IN": 6, |
|
"IP": 8, |
|
"LY": 1, |
|
"MT": 0, |
|
"NA": 4, |
|
"OP": 7, |
|
"SP": 2, |
|
"av": 22, |
|
"ds": 23, |
|
"dtp": 16, |
|
"ed": 24, |
|
"en": 14, |
|
"fi": 17, |
|
"it": 9, |
|
"lt": 18, |
|
"nb": 12, |
|
"ne": 10, |
|
"ob": 20, |
|
"ra": 15, |
|
"re": 13, |
|
"rs": 21, |
|
"rv": 19, |
|
"sr": 11 |
|
}, |
|
"layer_norm_eps": 1e-05, |
|
"max_position_embeddings": 514, |
|
"model_type": "xlm-roberta", |
|
"num_attention_heads": 16, |
|
"num_hidden_layers": 24, |
|
"output_past": true, |
|
"pad_token_id": 1, |
|
"position_embedding_type": "absolute", |
|
"torch_dtype": "bfloat16", |
|
"transformers_version": "4.39.3", |
|
"type_vocab_size": 1, |
|
"use_cache": true, |
|
"vocab_size": 250002 |
|
} |
|
|