first commit

README.md

@@ -0,0 +1,11 @@
+---
+language: ko
+tags:
+- bart
+license: mit
+---
+
+## KoBART-base-v2
+
+A model trained to handle the semantics of sequences longer than KoBART.
+

added_tokens.json

@@ -0,0 +1,2 @@
+{}
+

config.json

@@ -0,0 +1,54 @@
+{
+  "activation_dropout": 0.0,
+  "activation_function": "gelu",
+  "add_bias_logits": false,
+  "add_final_layer_norm": false,
+  "architectures": [
+    "BartModel"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "classif_dropout": 0.1,
+  "classifier_dropout": 0.1,
+  "d_model": 768,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 3072,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 6,
+  "decoder_start_token_id": 2,
+  "do_blenderbot_90_layernorm": false,
+  "dropout": 0.1,
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 3072,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 6,
+  "eos_token_id": 1,
+  "extra_pos_embeddings": 2,
+  "force_bos_token_to_be_generated": false,
+  "forced_eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "NEGATIVE",
+    "1": "POSITIVE"
+  },
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "NEGATIVE": 0,
+    "POSITIVE": 1
+  },
+  "max_position_embeddings": 1026,
+  "model_type": "bart",
+  "normalize_before": false,
+  "normalize_embedding": true,
+  "num_hidden_layers": 6,
+  "pad_token_id": 3,
+  "scale_embedding": false,
+  "static_position_embeddings": false,
+  "transformers_version": "4.5.1",
+  "use_cache": true,
+  "vocab_size": 30000,
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "author": "Heewon Jeon(madjakarta@gmail.com)",
+  "kobart_version": 1.0
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7082c7f1ae4efcc22c4b0081d71911059f4c5f65f278436c948d14b2d3d934d
+size 495536138

special_tokens_map.json

@@ -0,0 +1,2 @@
+{"bos_token": "</s>", "eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": "<mask>"}
+