polish-qa-v2 / config.json
pkedzia's picture
Updated model, better eval loss, actually is 0,7166 on the same dataset
e82eb44
raw
history blame contribute delete
706 Bytes
{
"_name_or_path": "sdadas/polish-roberta-large-v2",
"architectures": [
"RobertaForQuestionAnswering"
],
"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,
"initializer_range": 0.02,
"intermediate_size": 4096,
"layer_norm_eps": 1e-05,
"max_position_embeddings": 514,
"model_type": "roberta",
"num_attention_heads": 16,
"num_hidden_layers": 24,
"pad_token_id": 1,
"position_embedding_type": "absolute",
"torch_dtype": "float32",
"transformers_version": "4.32.0.dev0",
"type_vocab_size": 1,
"use_cache": true,
"vocab_size": 128001
}