Upload 11 files

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "train_loss": 0.33745264788858254,
+    "train_runtime": 18170.6622,
+    "train_samples": 61200,
+    "train_samples_per_second": 10.104,
+    "train_steps_per_second": 0.632
+}

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "CurtisJeon/klue-roberta-large-korquad_v1_qa",
+  "architectures": [
+    "CNN_RobertaForQuestionAnswering"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "tokenizer_class": "BertTokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.25.1",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 32000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47981b361deee511cebade5bf3b74f9082bd39a5615593388e8442dd94b60926
+size 1426552802

special_tokens_map.json

@@ -0,0 +1,9 @@
+{
+  "bos_token": "[CLS]",
+  "cls_token": "[CLS]",
+  "eos_token": "[SEP]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

tokenizer_config.json

@@ -0,0 +1,18 @@
+{
+  "bos_token": "[CLS]",
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": false,
+  "eos_token": "[SEP]",
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "name_or_path": "CurtisJeon/klue-roberta-large-korquad_v1_qa",
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "special_tokens_map_file": "/data/ephemeral/huggingface/models--klue--roberta-large/snapshots/28d911204e9022eda172571ca8cc61eaffd942f7/special_tokens_map.json",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "train_loss": 0.33745264788858254,
+    "train_runtime": 18170.6622,
+    "train_samples": 61200,
+    "train_samples_per_second": 10.104,
+    "train_steps_per_second": 0.632
+}

train_results.txt

@@ -0,0 +1,6 @@
+epoch = 3.0
+train_loss = 0.33745264788858254
+train_runtime = 18170.6622
+train_samples = 61200
+train_samples_per_second = 10.104
+train_steps_per_second = 0.632

trainer_state.json

@@ -0,0 +1,157 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "global_step": 11475,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9128540305010893e-06,
+      "loss": 0.7285,
+      "step": 500
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.8257080610021786e-06,
+      "loss": 0.5986,
+      "step": 1000
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.738562091503268e-06,
+      "loss": 0.5403,
+      "step": 1500
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.6514161220043572e-06,
+      "loss": 0.5121,
+      "step": 2000
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5642701525054465e-06,
+      "loss": 0.4629,
+      "step": 2500
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.4771241830065359e-06,
+      "loss": 0.4306,
+      "step": 3000
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.3899782135076252e-06,
+      "loss": 0.3937,
+      "step": 3500
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.3028322440087144e-06,
+      "loss": 0.3667,
+      "step": 4000
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.2156862745098038e-06,
+      "loss": 0.3126,
+      "step": 4500
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.1285403050108933e-06,
+      "loss": 0.3054,
+      "step": 5000
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.0413943355119825e-06,
+      "loss": 0.2873,
+      "step": 5500
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.542483660130718e-07,
+      "loss": 0.2772,
+      "step": 6000
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.671023965141612e-07,
+      "loss": 0.2762,
+      "step": 6500
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.799564270152506e-07,
+      "loss": 0.2742,
+      "step": 7000
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 6.928104575163398e-07,
+      "loss": 0.2607,
+      "step": 7500
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 6.056644880174292e-07,
+      "loss": 0.2272,
+      "step": 8000
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 5.185185185185185e-07,
+      "loss": 0.237,
+      "step": 8500
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 4.313725490196078e-07,
+      "loss": 0.2185,
+      "step": 9000
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 3.442265795206972e-07,
+      "loss": 0.2088,
+      "step": 9500
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 2.570806100217865e-07,
+      "loss": 0.2267,
+      "step": 10000
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 1.699346405228758e-07,
+      "loss": 0.1996,
+      "step": 10500
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 8.278867102396514e-08,
+      "loss": 0.2056,
+      "step": 11000
+    },
+    {
+      "epoch": 3.0,
+      "step": 11475,
+      "total_flos": 1.8232138178977805e+17,
+      "train_loss": 0.33745264788858254,
+      "train_runtime": 18170.6622,
+      "train_samples_per_second": 10.104,
+      "train_steps_per_second": 0.632
+    }
+  ],
+  "max_steps": 11475,
+  "num_train_epochs": 3,
+  "total_flos": 1.8232138178977805e+17,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc79a2fdf1b5f2da0fdfc187173835b6538df3249ced788721a6dd4d3855d152
+size 3960