upd W: 3 ep, bs=256, lr=1e-3, soft_res

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/kaggle/input/train-seq2seq/output",
+  "_name_or_path": "/kaggle/input/bert2bert-weights/ep_2",
   "architectures": [
     "EncoderDecoderModel"
   ],

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e8f47d2e27f8a51afc117515a0b284174b3d96869e2dcc98ae9b85be6a7f9d4a
-size 827926053
+oid sha256:16e28cf4eed30e910c0bd31482cdf98def223a38483f999bd59f787eae45a384
+size 827914439

tokenizer_config.json

@@ -1 +1 @@
-{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "special_tokens_map_file": "/kaggle/input/deeppavlov-rubertbasecased/special_tokens_map.json", "name_or_path": "/kaggle/input/train-seq2seq/output", "do_basic_tokenize": true, "never_split": null}
+{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "special_tokens_map_file": "/kaggle/input/deeppavlov-rubertbasecased/special_tokens_map.json", "name_or_path": "/kaggle/input/bert2bert-weights/ep_2", "do_basic_tokenize": true, "never_split": null}