|
{ |
|
"_name_or_path": "ltg/norbert3-small", |
|
"architectures": [ |
|
"NorbertForMaskedLM" |
|
], |
|
"attention_probs_dropout_prob": 0.05, |
|
"auto_map": { |
|
"AutoConfig": "ltg/norbert3-small--configuration_norbert.NorbertConfig", |
|
"AutoModel": "ltg/norbert3-small--modeling_norbert.NorbertModel", |
|
"AutoModelForMaskedLM": "ltg/norbert3-small--modeling_norbert.NorbertForMaskedLM", |
|
"AutoModelForMultipleChoice": "ltg/norbert3-small--modeling_norbert.NorbertForMultipleChoice", |
|
"AutoModelForQuestionAnswering": "ltg/norbert3-small--modeling_norbert.NorbertForQuestionAnswering", |
|
"AutoModelForSequenceClassification": "ltg/norbert3-small--modeling_norbert.NorbertForSequenceClassification", |
|
"AutoModelForTokenClassification": "ltg/norbert3-small--modeling_norbert.NorbertForTokenClassification" |
|
}, |
|
"hidden_dropout_prob": 0.05, |
|
"hidden_size": 384, |
|
"intermediate_size": 1024, |
|
"layer_norm_eps": 1e-07, |
|
"max_position_embeddings": 512, |
|
"num_attention_heads": 6, |
|
"num_hidden_layers": 12, |
|
"output_all_encoded_layers": true, |
|
"position_bucket_size": 32, |
|
"torch_dtype": "float32", |
|
"transformers_version": "4.37.2", |
|
"vocab_size": 50000 |
|
} |
|
|