Training in progress, step 100

config.json

@@ -1,35 +1,36 @@
 {
-  "_name_or_path": "klue/roberta-large",
+  "_name_or_path": "EleutherAI/polyglot-ko-1.3b",
   "architectures": [
-    "RobertaForSequenceClassification"
+    "GPTNeoXForSequenceClassification"
   ],
-  "attention_probs_dropout_prob": 0.1,
+  "attention_dropout": 0.0,
   "bos_token_id": 0,
-  "classifier_dropout": null,
+  "classifier_dropout": 0.1,
   "eos_token_id": 2,
-  "gradient_checkpointing": false,
   "hidden_act": "gelu",
-  "hidden_dropout_prob": 0.1,
-  "hidden_size": 1024,
+  "hidden_dropout": 0.0,
+  "hidden_size": 2048,
   "id2label": {
     "0": "LABEL_0"
   },
   "initializer_range": 0.02,
-  "intermediate_size": 4096,
+  "intermediate_size": 8192,
   "label2id": {
     "LABEL_0": 0
   },
   "layer_norm_eps": 1e-05,
-  "max_position_embeddings": 514,
-  "model_type": "roberta",
+  "max_position_embeddings": 2048,
+  "model_type": "gpt_neox",
   "num_attention_heads": 16,
   "num_hidden_layers": 24,
   "pad_token_id": 2,
-  "position_embedding_type": "absolute",
-  "tokenizer_class": "BertTokenizer",
+  "rope_scaling": null,
+  "rotary_emb_base": 10000,
+  "rotary_pct": 0.5,
+  "tie_word_embeddings": false,
   "torch_dtype": "float32",
   "transformers_version": "4.34.1",
-  "type_vocab_size": 1,
   "use_cache": true,
-  "vocab_size": 32000
+  "use_parallel_residual": true,
+  "vocab_size": 30080
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:62de9ff850ddfe83385cd791735b3bc98586feb411a922fe58bb711351b98032
-size 1346765678
+oid sha256:d5a038be930d7e4b9c06380cca5c12b961950b685498eb1df7b7188f4af6c5ba
+size 5080945162

special_tokens_map.json

@@ -1,9 +1,11 @@
 {
-  "bos_token": "[CLS]",
-  "cls_token": "[CLS]",
-  "eos_token": "[SEP]",
-  "mask_token": "[MASK]",
-  "pad_token": "[PAD]",
-  "sep_token": "[SEP]",
-  "unk_token": "[UNK]"
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "<|sep|>",
+    "<|acc|>",
+    "<|tel|>",
+    "<|rrn|>"
+  ],
+  "eos_token": "<|endoftext|>",
+  "pad_token": "<|endoftext|>"
 }

tokenizer_config.json

@@ -1,7 +1,7 @@
 {
   "added_tokens_decoder": {
     "0": {
-      "content": "[CLS]",
+      "content": "<|unused0|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -9,7 +9,7 @@
       "special": true
     },
     "1": {
-      "content": "[PAD]",
+      "content": "<|unused1|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -17,7 +17,7 @@
       "special": true
     },
     "2": {
-      "content": "[SEP]",
+      "content": "<|endoftext|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -25,15 +25,31 @@
       "special": true
     },
     "3": {
-      "content": "[UNK]",
+      "content": "<|sep|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
       "single_word": false,
       "special": true
     },
-    "4": {
-      "content": "[MASK]",
+    "30000": {
+      "content": "<|acc|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "30001": {
+      "content": "<|tel|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "30002": {
+      "content": "<|rrn|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -41,19 +57,16 @@
       "special": true
     }
   },
-  "bos_token": "[CLS]",
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "<|sep|>",
+    "<|acc|>",
+    "<|tel|>",
+    "<|rrn|>"
+  ],
   "clean_up_tokenization_spaces": true,
-  "cls_token": "[CLS]",
-  "do_basic_tokenize": true,
-  "do_lower_case": false,
-  "eos_token": "[SEP]",
-  "mask_token": "[MASK]",
-  "model_max_length": 512,
-  "never_split": null,
-  "pad_token": "[PAD]",
-  "sep_token": "[SEP]",
-  "strip_accents": null,
-  "tokenize_chinese_chars": true,
-  "tokenizer_class": "BertTokenizer",
-  "unk_token": "[UNK]"
+  "eos_token": "<|endoftext|>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e1b0be5f447301b577e44061c504ec9398fb05913a538169784e7cfc7134c7d3
+oid sha256:19f57ba66f27fb038e41f4b640f06321151aa6bdef95a3c3a615bf73330d3d10
 size 4600