model improved

README.md

@@ -15,7 +15,7 @@ widget:
 
 ## Model Description
 
-This is a DeBERTa(V2) model pre-trained on 青空文庫 texts. You can fine-tune `deberta-large-japanese-aozora` for downstream tasks, such as [POS-tagging](https://huggingface.co/KoichiYasuoka/deberta-large-japanese-luw-upos), [dependency-parsing](https://huggingface.co/KoichiYasuoka/deberta-large-japanese-aozora-ud-head), and so on.
+This is a DeBERTa(V2) model pre-trained on 青空文庫 texts. NVIDIA A100-SXM4-40GB took 127 hours 8 minutes for training. You can fine-tune `deberta-large-japanese-aozora` for downstream tasks, such as [POS-tagging](https://huggingface.co/KoichiYasuoka/deberta-large-japanese-luw-upos), [dependency-parsing](https://huggingface.co/KoichiYasuoka/deberta-large-japanese-aozora-ud-head), and so on.
 
 ## How to Use
 

config.json

@@ -20,12 +20,15 @@
   "pooler_dropout": 0,
   "pooler_hidden_act": "gelu",
   "pooler_hidden_size": 1024,
-  "pos_att_type": null,
-  "position_biased_input": true,
-  "relative_attention": false,
+  "pos_att_type": [
+    "p2c",
+    "c2p"
+  ],
+  "position_biased_input": false,
+  "relative_attention": true,
   "tokenizer_class": "DebertaV2TokenizerFast",
   "torch_dtype": "float32",
-  "transformers_version": "4.19.2",
+  "transformers_version": "4.22.1",
   "type_vocab_size": 0,
   "vocab_size": 32000
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ca0e9802d5d73c632b43918d0258abcf7f3ff958348ee445b49a20e26202071
-size 1346887533
+oid sha256:593d7a3dd4e1f65737c8a65ffcf54cffc04070e62bd39b15768975d9cadb886a
+size 1550541549

special_tokens_map.json

@@ -1 +1,9 @@
-{"bos_token": "[CLS]", "eos_token": "[SEP]", "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
+{
+  "bos_token": "[CLS]",
+  "cls_token": "[CLS]",
+  "eos_token": "[SEP]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

tokenizer_config.json

@@ -1 +1,14 @@
-{"do_lower_case": false, "bos_token": "[CLS]", "eos_token": "[SEP]", "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "split_by_punct": true, "keep_accents": true, "model_max_length": 512, "tokenizer_class": "DebertaV2TokenizerFast"}
+{
+  "bos_token": "[CLS]",
+  "cls_token": "[CLS]",
+  "do_lower_case": false,
+  "eos_token": "[SEP]",
+  "keep_accents": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "split_by_punct": true,
+  "tokenizer_class": "DebertaV2TokenizerFast",
+  "unk_token": "[UNK]"
+}