Training in progress, step 100

config.json

@@ -1,54 +1,46 @@
 {
-  "_name_or_path": "kykim/funnel-kor-base",
-  "activation_dropout": 0.0,
+  "_name_or_path": "kykim/bert-kor-base",
   "architectures": [
-    "FunnelForSequenceClassification"
+    "BertForSequenceClassification"
   ],
-  "attention_dropout": 0.1,
-  "attention_type": "relative_shift",
-  "block_repeats": [
-    1,
-    1,
-    1
-  ],
-  "block_sizes": [
-    6,
-    6,
-    6
-  ],
-  "d_head": 64,
-  "d_inner": 3072,
-  "d_model": 768,
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "directionality": "bidi",
   "embedding_size": 768,
-  "hidden_act": "gelu_new",
-  "hidden_dropout": 0.1,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
   "id2label": {
     "0": "item",
     "1": "non_item",
     "2": "garbage",
     "3": "review"
   },
-  "initializer_range": 0.1,
-  "initializer_std": 0.02,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
   "label2id": {
     "garbage": 2,
     "item": 0,
     "non_item": 1,
     "review": 3
   },
-  "layer_norm_eps": 1e-09,
+  "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
-  "model_type": "funnel",
-  "n_head": 12,
-  "num_decoder_layers": 2,
-  "pool_q_only": true,
-  "pooling_type": "mean",
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "pooler_fc_size": 768,
+  "pooler_num_attention_heads": 12,
+  "pooler_num_fc_layers": 3,
+  "pooler_size_per_head": 128,
+  "pooler_type": "first_token_transform",
+  "position_embedding_type": "absolute",
   "problem_type": "single_label_classification",
-  "rel_attn_type": "factorized",
-  "separate_cls": true,
   "torch_dtype": "float32",
   "transformers_version": "4.32.0",
-  "truncate_seq": true,
-  "type_vocab_size": 3,
+  "type_vocab_size": 2,
+  "use_cache": true,
   "vocab_size": 42000
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5fbf6b79baa3c5a72e3959086ca02b9e0eb96f11a9c6906670bcf808fa3b33c0
-size 684546438
+oid sha256:cfcb16109e2b7036c61479eacd459d8afd92c0812f406927009271be31b7dc60
+size 473270510

special_tokens_map.json

@@ -1,9 +1,7 @@
 {
-  "bos_token": "<s>",
-  "cls_token": "<cls>",
-  "eos_token": "</s>",
-  "mask_token": "<mask>",
-  "pad_token": "<pad>",
-  "sep_token": "<sep>",
-  "unk_token": "<unk>"
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
 }

tokenizer.json

@@ -10,7 +10,7 @@
   "added_tokens": [
     {
       "id": 0,
-      "content": "<pad>",
+      "content": "[PAD]",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -19,7 +19,7 @@
     },
     {
       "id": 1,
-      "content": "<unk>",
+      "content": "[UNK]",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -28,7 +28,7 @@
     },
     {
       "id": 2,
-      "content": "<cls>",
+      "content": "[CLS]",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -37,7 +37,7 @@
     },
     {
       "id": 3,
-      "content": "<sep>",
+      "content": "[SEP]",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -46,25 +46,7 @@
     },
     {
       "id": 4,
-      "content": "<mask>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 5,
-      "content": "<s>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 6,
-      "content": "</s>",
+      "content": "[MASK]",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -87,8 +69,8 @@
     "single": [
       {
         "SpecialToken": {
-          "id": "<cls>",
-          "type_id": 2
+          "id": "[CLS]",
+          "type_id": 0
         }
       },
       {
@@ -99,7 +81,7 @@
       },
       {
         "SpecialToken": {
-          "id": "<sep>",
+          "id": "[SEP]",
           "type_id": 0
         }
       }
@@ -107,8 +89,8 @@
     "pair": [
       {
         "SpecialToken": {
-          "id": "<cls>",
-          "type_id": 2
+          "id": "[CLS]",
+          "type_id": 0
         }
       },
       {
@@ -119,7 +101,7 @@
       },
       {
         "SpecialToken": {
-          "id": "<sep>",
+          "id": "[SEP]",
           "type_id": 0
         }
       },
@@ -131,28 +113,28 @@
       },
       {
         "SpecialToken": {
-          "id": "<sep>",
+          "id": "[SEP]",
           "type_id": 1
         }
       }
     ],
     "special_tokens": {
-      "<cls>": {
-        "id": "<cls>",
+      "[CLS]": {
+        "id": "[CLS]",
         "ids": [
           2
         ],
         "tokens": [
-          "<cls>"
+          "[CLS]"
         ]
       },
-      "<sep>": {
-        "id": "<sep>",
+      "[SEP]": {
+        "id": "[SEP]",
         "ids": [
           3
         ],
         "tokens": [
-          "<sep>"
+          "[SEP]"
         ]
       }
     }
@@ -164,19 +146,19 @@
   },
   "model": {
     "type": "WordPiece",
-    "unk_token": "<unk>",
+    "unk_token": "[UNK]",
     "continuing_subword_prefix": "##",
     "max_input_chars_per_word": 100,
     "vocab": {
-      "<pad>": 0,
-      "<unk>": 1,
-      "<cls>": 2,
-      "<sep>": 3,
-      "<mask>": 4,
-      "<s>": 5,
-      "</s>": 6,
-      "<eod>": 7,
-      "<eop>": 8,
+      "[PAD]": 0,
+      "[UNK]": 1,
+      "[CLS]": 2,
+      "[SEP]": 3,
+      "[MASK]": 4,
+      "[unused0]": 5,
+      "[unused1]": 6,
+      "[unused2]": 7,
+      "[unused3]": 8,
       "[unused4]": 9,
       "[unused5]": 10,
       "[unused6]": 11,

tokenizer_config.json

@@ -1,19 +1,15 @@
 {
-  "bos_token": "<s>",
-  "clean_text": true,
   "clean_up_tokenization_spaces": true,
-  "cls_token": "<cls>",
+  "cls_token": "[CLS]",
   "do_basic_tokenize": true,
   "do_lower_case": true,
-  "eos_token": "</s>",
-  "mask_token": "<mask>",
+  "mask_token": "[MASK]",
   "model_max_length": 512,
   "never_split": null,
-  "pad_token": "<pad>",
-  "sep_token": "<sep>",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
   "strip_accents": false,
   "tokenize_chinese_chars": true,
-  "tokenizer_class": "FunnelTokenizer",
-  "unk_token": "<unk>",
-  "wordpieces_prefix": "##"
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d9a5a04723c7b140f278b33665efcfdaa8934f2abd07e0d4349157e35f33bb1
+oid sha256:ba5c275c0bb7447c906e5ea04738c5d9ea5a84e1e3000af05de39d43a8eaacef
 size 4472

vocab.txt

@@ -1,12 +1,12 @@
-<pad>
-<unk>
-<cls>
-<sep>
-<mask>
-<s>
-</s>
-<eod>
-<eop>
+[PAD]
+[UNK]
+[CLS]
+[SEP]
+[MASK]
+[unused0]
+[unused1]
+[unused2]
+[unused3]
 [unused4]
 [unused5]
 [unused6]