End of training

all_results.json

@@ -1,4 +1,4 @@
 {
-    "eval_exact_match": 77.8996344300432,
-    "eval_f1": 77.8996344300432
+    "eval_exact_match": 78.92987703555998,
+    "eval_f1": 78.92987703555998
 }

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "cyclone/simcse-chinese-roberta-wwm-ext",
+  "_name_or_path": "hfl/chinese-roberta-wwm-ext-large",
   "architectures": [
     "BertForQuestionAnswering"
   ],
@@ -8,19 +8,18 @@
   "classifier_dropout": null,
   "directionality": "bidi",
   "eos_token_id": 2,
-  "gradient_checkpointing": false,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
-  "hidden_size": 768,
+  "hidden_size": 1024,
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
+  "intermediate_size": 4096,
   "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
   "model_type": "bert",
-  "num_attention_heads": 12,
-  "num_hidden_layers": 12,
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
   "output_past": true,
-  "pad_token_id": 1,
+  "pad_token_id": 0,
   "pooler_fc_size": 768,
   "pooler_num_attention_heads": 12,
   "pooler_num_fc_layers": 3,

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6c16a6b1731bbb1b3a9c4b65e27f5895044521eb61db41691bcbb46e07054e26
-size 14332554
+oid sha256:17f5d4c7aedd8688d51fea050aa3d206769f956c4fc79fc586ce63151da64112
+size 14265158

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c2e1ffe6f83bb62cc52589411bfeab349b5c23fc57b5fa900fdce8d146ea06a2
-size 406782121
+oid sha256:83376d941c0d8824fa13dec8e4f8efb571d462452919396ac2069d1bc0a4c0d6
+size 1298033321

tokenizer_config.json

@@ -46,18 +46,11 @@
   "cls_token": "[CLS]",
   "do_lower_case": true,
   "mask_token": "[MASK]",
-  "max_length": 64,
   "model_max_length": 1000000000000000019884624838656,
-  "pad_to_multiple_of": null,
   "pad_token": "[PAD]",
-  "pad_token_type_id": 0,
-  "padding_side": "right",
   "sep_token": "[SEP]",
-  "stride": 0,
   "strip_accents": null,
   "tokenize_chinese_chars": true,
   "tokenizer_class": "BertTokenizer",
-  "truncation_side": "right",
-  "truncation_strategy": "longest_first",
   "unk_token": "[UNK]"
 }