hard time tho

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 10.0,
+    "eval_samples": 474,
+    "exact_match": 68.75,
+    "f1": 76.5876322751323,
+    "init_mem_cpu_alloc_delta": 17783119,
+    "init_mem_cpu_peaked_delta": 18230,
+    "init_mem_gpu_alloc_delta": 1414861824,
+    "init_mem_gpu_peaked_delta": 67165696,
+    "train_mem_cpu_alloc_delta": 2991504,
+    "train_mem_cpu_peaked_delta": 264499229,
+    "train_mem_gpu_alloc_delta": 4256099328,
+    "train_mem_gpu_peaked_delta": 11330810880,
+    "train_runtime": 8323.4424,
+    "train_samples": 7978,
+    "train_samples_per_second": 1.199
+}

optimizer.pt → checkpoint-9500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:db17e7ecb5d8209144c77f48530f20623e5012d732c516b793a3d7d34df7949b
+oid sha256:2939c7c78324e191101544dc3cb7ab8bdc8c8bf6ac6956b10a9ebe29a2c032b5
 size 2819470077

pytorch_model.bin → checkpoint-9500/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:46083e65954b4d081bd00995edde7772af0098346f2d0c19df14480d01797636
+oid sha256:2d1edd8b3db3558e5ad79f866e47ff5a4a502a2e99b60d82805de41bcbf3ad0f
 size 1413985490

scheduler.pt → checkpoint-9500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:04876499127dd83ce05f49419379d3ce78d67b20405c3a81867244d553097304
+oid sha256:58b1a986b1dea83d69bef3937079a177d0187daf845e863be53b4af16af445a6
 size 623

tokenizer_config.json → checkpoint-9500/tokenizer_config.json

@@ -1 +1 @@
-{"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "do_basic_tokenize": true, "never_split": null, "bos_token": "[CLS]", "eos_token": "[SEP]", "model_max_length": 512, "special_tokens_map_file": "/opt/ml/.cache/huggingface/transformers/1a24ab4628028ed80dea35ce3334a636dc656fd9a17a09bad377f88f0cbecdac.70c17d6e4d492c8f24f5bb97ab56c7f272e947112c6faf9dd846da42ba13eb23", "name_or_path": "klue/roberta-large", "tokenizer_class": "BertTokenizer"}
+{"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "do_basic_tokenize": true, "never_split": null, "bos_token": "[CLS]", "eos_token": "[SEP]", "model_max_length": 512, "tokenizer_class": "BertTokenizer", "special_tokens_map_file": "/opt/ml/.cache/huggingface/transformers/1a24ab4628028ed80dea35ce3334a636dc656fd9a17a09bad377f88f0cbecdac.70c17d6e4d492c8f24f5bb97ab56c7f272e947112c6faf9dd846da42ba13eb23", "name_or_path": "klue/roberta-large"}

checkpoint-9500/trainer_state.json

@@ -0,0 +1,586 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 9.519038076152304,
+  "global_step": 9500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.997522904167843e-07,
+      "loss": 26.4633,
+      "step": 100
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.990094071072878e-07,
+      "loss": 9.8101,
+      "step": 200
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 9.977720861487699e-07,
+      "loss": 5.8569,
+      "step": 300
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 9.96041553526267e-07,
+      "loss": 5.5213,
+      "step": 400
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 9.938195239178374e-07,
+      "loss": 5.1344,
+      "step": 500
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 9.911081989955939e-07,
+      "loss": 4.9245,
+      "step": 600
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 9.879102652442023e-07,
+      "loss": 4.5925,
+      "step": 700
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 9.842288912990095e-07,
+      "loss": 4.676,
+      "step": 800
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 9.80067724806438e-07,
+      "loss": 4.4576,
+      "step": 900
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 9.754308888097582e-07,
+      "loss": 4.2986,
+      "step": 1000
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 9.703229776638185e-07,
+      "loss": 3.8739,
+      "step": 1100
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 9.647490524827833e-07,
+      "loss": 3.5629,
+      "step": 1200
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 9.587146361253867e-07,
+      "loss": 2.9575,
+      "step": 1300
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 9.522257077226716e-07,
+      "loss": 2.8326,
+      "step": 1400
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 9.452886967536388e-07,
+      "loss": 2.7891,
+      "step": 1500
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.379104766746722e-07,
+      "loss": 2.4971,
+      "step": 1600
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.30098358109054e-07,
+      "loss": 2.4297,
+      "step": 1700
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 9.218600816033199e-07,
+      "loss": 2.183,
+      "step": 1800
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 9.132038099576289e-07,
+      "loss": 2.0732,
+      "step": 1900
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 9.041381201377467e-07,
+      "loss": 2.2732,
+      "step": 2000
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 8.94671994776661e-07,
+      "loss": 1.7384,
+      "step": 2100
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 8.84814813274243e-07,
+      "loss": 1.8955,
+      "step": 2200
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 8.745763425037795e-07,
+      "loss": 2.0546,
+      "step": 2300
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 8.639667271345798e-07,
+      "loss": 1.6855,
+      "step": 2400
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 8.529964795802484e-07,
+      "loss": 1.7512,
+      "step": 2500
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 8.416764695825834e-07,
+      "loss": 1.7856,
+      "step": 2600
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 8.300179134414187e-07,
+      "loss": 1.7027,
+      "step": 2700
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 8.180323629010848e-07,
+      "loss": 1.5552,
+      "step": 2800
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 8.057316937044976e-07,
+      "loss": 1.6427,
+      "step": 2900
+    },
+    {
+      "epoch": 3.01,
+      "learning_rate": 7.931280938262168e-07,
+      "loss": 1.7996,
+      "step": 3000
+    },
+    {
+      "epoch": 3.11,
+      "learning_rate": 7.802340513961341e-07,
+      "loss": 1.4622,
+      "step": 3100
+    },
+    {
+      "epoch": 3.21,
+      "learning_rate": 7.670623423257547e-07,
+      "loss": 1.4137,
+      "step": 3200
+    },
+    {
+      "epoch": 3.31,
+      "learning_rate": 7.536260176493347e-07,
+      "loss": 1.4894,
+      "step": 3300
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 7.399383905924165e-07,
+      "loss": 1.4263,
+      "step": 3400
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 7.26013023380574e-07,
+      "loss": 1.491,
+      "step": 3500
+    },
+    {
+      "epoch": 3.61,
+      "learning_rate": 7.118637138014395e-07,
+      "loss": 1.2858,
+      "step": 3600
+    },
+    {
+      "epoch": 3.71,
+      "learning_rate": 6.975044815333281e-07,
+      "loss": 1.2851,
+      "step": 3700
+    },
+    {
+      "epoch": 3.81,
+      "learning_rate": 6.829495542540013e-07,
+      "loss": 1.3229,
+      "step": 3800
+    },
+    {
+      "epoch": 3.91,
+      "learning_rate": 6.682133535433393e-07,
+      "loss": 1.3675,
+      "step": 3900
+    },
+    {
+      "epoch": 4.01,
+      "learning_rate": 6.533104805938873e-07,
+      "loss": 1.3686,
+      "step": 4000
+    },
+    {
+      "epoch": 4.11,
+      "learning_rate": 6.382557017434331e-07,
+      "loss": 1.0781,
+      "step": 4100
+    },
+    {
+      "epoch": 4.21,
+      "learning_rate": 6.230639338439549e-07,
+      "loss": 1.1388,
+      "step": 4200
+    },
+    {
+      "epoch": 4.31,
+      "learning_rate": 6.077502294814311e-07,
+      "loss": 1.0966,
+      "step": 4300
+    },
+    {
+      "epoch": 4.41,
+      "learning_rate": 5.923297620611622e-07,
+      "loss": 1.1764,
+      "step": 4400
+    },
+    {
+      "epoch": 4.51,
+      "learning_rate": 5.76817810773379e-07,
+      "loss": 1.2625,
+      "step": 4500
+    },
+    {
+      "epoch": 4.61,
+      "learning_rate": 5.612297454540351e-07,
+      "loss": 1.1718,
+      "step": 4600
+    },
+    {
+      "epoch": 4.71,
+      "learning_rate": 5.455810113557839e-07,
+      "loss": 1.218,
+      "step": 4700
+    },
+    {
+      "epoch": 4.81,
+      "learning_rate": 5.298871138442307e-07,
+      "loss": 1.2203,
+      "step": 4800
+    },
+    {
+      "epoch": 4.91,
+      "learning_rate": 5.14163603034622e-07,
+      "loss": 1.0743,
+      "step": 4900
+    },
+    {
+      "epoch": 5.01,
+      "learning_rate": 4.984260583841952e-07,
+      "loss": 1.1825,
+      "step": 5000
+    },
+    {
+      "epoch": 5.11,
+      "learning_rate": 4.82690073255455e-07,
+      "loss": 0.8899,
+      "step": 5100
+    },
+    {
+      "epoch": 5.21,
+      "learning_rate": 4.6697123946567224e-07,
+      "loss": 0.9867,
+      "step": 5200
+    },
+    {
+      "epoch": 5.31,
+      "learning_rate": 4.512851318379138e-07,
+      "loss": 0.9842,
+      "step": 5300
+    },
+    {
+      "epoch": 5.41,
+      "learning_rate": 4.3564729276891087e-07,
+      "loss": 1.1441,
+      "step": 5400
+    },
+    {
+      "epoch": 5.51,
+      "learning_rate": 4.2007321682905594e-07,
+      "loss": 0.9965,
+      "step": 5500
+    },
+    {
+      "epoch": 5.61,
+      "learning_rate": 4.0457833540978923e-07,
+      "loss": 0.9929,
+      "step": 5600
+    },
+    {
+      "epoch": 5.71,
+      "learning_rate": 3.89178001433584e-07,
+      "loss": 1.152,
+      "step": 5700
+    },
+    {
+      "epoch": 5.81,
+      "learning_rate": 3.7388747414168295e-07,
+      "loss": 0.8896,
+      "step": 5800
+    },
+    {
+      "epoch": 5.91,
+      "learning_rate": 3.5872190397465635e-07,
+      "loss": 1.1022,
+      "step": 5900
+    },
+    {
+      "epoch": 6.01,
+      "learning_rate": 3.436963175607656e-07,
+      "loss": 0.8925,
+      "step": 6000
+    },
+    {
+      "epoch": 6.11,
+      "learning_rate": 3.288256028270033e-07,
+      "loss": 0.9647,
+      "step": 6100
+    },
+    {
+      "epoch": 6.21,
+      "learning_rate": 3.141244942475647e-07,
+      "loss": 0.7659,
+      "step": 6200
+    },
+    {
+      "epoch": 6.31,
+      "learning_rate": 2.996075582443658e-07,
+      "loss": 1.0098,
+      "step": 6300
+    },
+    {
+      "epoch": 6.41,
+      "learning_rate": 2.8528917875407433e-07,
+      "loss": 1.0704,
+      "step": 6400
+    },
+    {
+      "epoch": 6.51,
+      "learning_rate": 2.711835429759539e-07,
+      "loss": 1.0039,
+      "step": 6500
+    },
+    {
+      "epoch": 6.61,
+      "learning_rate": 2.573046273146427e-07,
+      "loss": 0.9493,
+      "step": 6600
+    },
+    {
+      "epoch": 6.71,
+      "learning_rate": 2.4366618353179644e-07,
+      "loss": 0.9582,
+      "step": 6700
+    },
+    {
+      "epoch": 6.81,
+      "learning_rate": 2.30281725120316e-07,
+      "loss": 0.8875,
+      "step": 6800
+    },
+    {
+      "epoch": 6.91,
+      "learning_rate": 2.1716451391466006e-07,
+      "loss": 0.8132,
+      "step": 6900
+    },
+    {
+      "epoch": 7.01,
+      "learning_rate": 2.0432754695051136e-07,
+      "loss": 1.0703,
+      "step": 7000
+    },
+    {
+      "epoch": 7.11,
+      "learning_rate": 1.9178354358681548e-07,
+      "loss": 1.0679,
+      "step": 7100
+    },
+    {
+      "epoch": 7.21,
+      "learning_rate": 1.7954493290295309e-07,
+      "loss": 0.8124,
+      "step": 7200
+    },
+    {
+      "epoch": 7.31,
+      "learning_rate": 1.6762384138353075e-07,
+      "loss": 0.9779,
+      "step": 7300
+    },
+    {
+      "epoch": 7.41,
+      "learning_rate": 1.5603208090299496e-07,
+      "loss": 0.7439,
+      "step": 7400
+    },
+    {
+      "epoch": 7.52,
+      "learning_rate": 1.4478113702197569e-07,
+      "loss": 0.858,
+      "step": 7500
+    },
+    {
+      "epoch": 7.62,
+      "learning_rate": 1.3388215760695098e-07,
+      "loss": 1.0146,
+      "step": 7600
+    },
+    {
+      "epoch": 7.72,
+      "learning_rate": 1.2334594178451424e-07,
+      "loss": 0.7714,
+      "step": 7700
+    },
+    {
+      "epoch": 7.82,
+      "learning_rate": 1.1318292924118584e-07,
+      "loss": 0.8889,
+      "step": 7800
+    },
+    {
+      "epoch": 7.92,
+      "learning_rate": 1.0340318987937097e-07,
+      "loss": 1.0001,
+      "step": 7900
+    },
+    {
+      "epoch": 8.02,
+      "learning_rate": 9.401641383971476e-08,
+      "loss": 0.9567,
+      "step": 8000
+    },
+    {
+      "epoch": 8.12,
+      "learning_rate": 8.503190189973914e-08,
+      "loss": 0.834,
+      "step": 8100
+    },
+    {
+      "epoch": 8.22,
+      "learning_rate": 7.645855625827657e-08,
+      "loss": 0.9169,
+      "step": 8200
+    },
+    {
+      "epoch": 8.32,
+      "learning_rate": 6.830487171482935e-08,
+      "loss": 0.841,
+      "step": 8300
+    },
+    {
+      "epoch": 8.42,
+      "learning_rate": 6.057892725259717e-08,
+      "loss": 1.0763,
+      "step": 8400
+    },
+    {
+      "epoch": 8.52,
+      "learning_rate": 5.328837803351083e-08,
+      "loss": 0.8885,
+      "step": 8500
+    },
+    {
+      "epoch": 8.62,
+      "learning_rate": 4.644044781320422e-08,
+      "loss": 0.7986,
+      "step": 8600
+    },
+    {
+      "epoch": 8.72,
+      "learning_rate": 4.004192178344029e-08,
+      "loss": 0.701,
+      "step": 8700
+    },
+    {
+      "epoch": 8.82,
+      "learning_rate": 3.4099139849083304e-08,
+      "loss": 0.9347,
+      "step": 8800
+    },
+    {
+      "epoch": 8.92,
+      "learning_rate": 2.8617990346277655e-08,
+      "loss": 1.0685,
+      "step": 8900
+    },
+    {
+      "epoch": 9.02,
+      "learning_rate": 2.3603904208058688e-08,
+      "loss": 0.8046,
+      "step": 9000
+    },
+    {
+      "epoch": 9.12,
+      "learning_rate": 1.9061849583176636e-08,
+      "loss": 0.8515,
+      "step": 9100
+    },
+    {
+      "epoch": 9.22,
+      "learning_rate": 1.499632691346375e-08,
+      "loss": 0.8998,
+      "step": 9200
+    },
+    {
+      "epoch": 9.32,
+      "learning_rate": 1.1411364474624264e-08,
+      "loss": 0.8923,
+      "step": 9300
+    },
+    {
+      "epoch": 9.42,
+      "learning_rate": 8.31051438486441e-09,
+      "loss": 0.7543,
+      "step": 9400
+    },
+    {
+      "epoch": 9.52,
+      "learning_rate": 5.696849085317645e-09,
+      "loss": 0.883,
+      "step": 9500
+    }
+  ],
+  "max_steps": 9980,
+  "num_train_epochs": 10,
+  "total_flos": 0.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin → checkpoint-9500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:afe3ac874b2d51fc44874710f9192cd0e67d40eba3403f25e0f6abd7d2dab16f
-size 2863
+oid sha256:574e5e2a956b3e59901c3ed9848a90b98bd67a72f27a5e69b189b003d4e05fd4
+size 2287

config copy.json

@@ -0,0 +1,27 @@
+{
+  "_name_or_path": "QAWithLSTMModel",
+  "architectures": [
+    "QAWithLSTMModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "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",
+  "transformers_version": "4.4.1",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 32000
+}

config.json

@@ -1,27 +0,0 @@
-{
-    "architectures": [
-      "klue/roberta-large"
-    ],
-    "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",
-    "transformers_version": "4.11.3",
-    "type_vocab_size": 1,
-    "use_cache": true,
-    "vocab_size": 32000
-  }

eval_results.json

@@ -0,0 +1,6 @@
+{
+    "epoch": 10.0,
+    "eval_samples": 474,
+    "exact_match": 68.75,
+    "f1": 76.5876322751323
+}

predictions.json

@@ -0,0 +1,242 @@
+{
+    "mrc-0-003264": "한보철강",
+    "mrc-0-004762": "1871년",
+    "mrc-1-001810": "나뭇잎",
+    "mrc-1-000219": "금대야",
+    "mrc-1-000285": "수평적 관계",
+    "mrc-0-005106": "쇼와 천황의 옥음방송",
+    "mrc-0-002076": "코칭 스티치",
+    "mrc-1-000414": "감염병",
+    "mrc-0-002875": "스페인",
+    "mrc-0-003828": "20세기 초",
+    "mrc-0-002778": "\"5월의 왕\"",
+    "mrc-0-003931": "'일급 비밀 프로젝트 2501'",
+    "mrc-0-002485": "테헤란",
+    "mrc-0-004483": "역사교육과정개발추진위원회",
+    "mrc-0-003032": "1967년 8월 16일",
+    "mrc-1-000724": "1963년",
+    "mrc-0-002138": "쿠빌라이",
+    "mrc-0-003727": "〈중앙일보〉",
+    "mrc-0-003115": "이이노야 성",
+    "mrc-0-003088": "전체 4순위",
+    "mrc-0-005296": "뇌물",
+    "mrc-1-001495": "준융합성 천연두",
+    "mrc-0-005289": "데스탱 장군",
+    "mrc-0-002240": "박트리아",
+    "mrc-0-000248": "'진전(陳田)'이라 새겨진 기와조각",
+    "mrc-0-001846": "《국가》",
+    "mrc-1-001653": "2011년 3월 19일",
+    "mrc-0-000785": "브라질 포르투갈어",
+    "mrc-0-001519": "릿지당",
+    "mrc-0-002457": "강도역왕",
+    "mrc-1-001479": "포드 극장",
+    "mrc-0-001707": "만주국 관리",
+    "mrc-0-002280": "『협동조합에 관하여』",
+    "mrc-1-001023": "도고쿠",
+    "mrc-0-001022": "루피의 할아버지인 몽키 D. 가프",
+    "mrc-0-001150": "조류",
+    "mrc-0-001807": "광주교도소",
+    "mrc-0-001161": "베게티우스",
+    "mrc-0-004654": "가오슝 시",
+    "mrc-0-002468": "망치",
+    "mrc-0-001870": "마리즈 교수",
+    "mrc-1-000753": "\"트로이아 노바\"",
+    "mrc-0-001073": "토요일과 일요일 이틀",
+    "mrc-1-001116": "의상대사",
+    "mrc-1-000995": "시모키타 반도",
+    "mrc-0-002175": "‘우유의 바다’",
+    "mrc-0-001894": "장어",
+    "mrc-0-000337": "소련",
+    "mrc-0-002013": "전치",
+    "mrc-1-001766": "흑색육(黑色肉)",
+    "mrc-0-002333": "욱",
+    "mrc-1-001814": "독일 인민당",
+    "mrc-0-001554": "하기노 역",
+    "mrc-0-002247": "회칠",
+    "mrc-0-000547": "황강다리",
+    "mrc-1-001177": "직업 교육",
+    "mrc-0-000939": "공산당",
+    "mrc-0-000223": "음력 정월",
+    "mrc-0-005386": "조계종",
+    "mrc-0-000118": "베이징",
+    "mrc-0-003529": "독일군",
+    "mrc-1-000459": "1932년",
+    "mrc-0-003801": "제서지전(齊西之戰)",
+    "mrc-0-004342": "잦은 이슬비",
+    "mrc-0-001311": "출생 천궁도",
+    "mrc-0-002931": "나말여초 불상",
+    "mrc-0-002767": "1939년",
+    "mrc-1-000961": "창 절제술",
+    "mrc-1-000796": "도버 밀",
+    "mrc-1-000449": "여정현(呂正鉉)",
+    "mrc-0-004133": "광배",
+    "mrc-0-003576": "울산",
+    "mrc-0-002692": "땅(은색)과 하늘(금색)을 맺고 푸는 권한",
+    "mrc-0-004899": "저수지",
+    "mrc-0-003677": "퀘이커 교도",
+    "mrc-1-000127": "피에르 오주로",
+    "mrc-0-005412": "엠넷미디어 지분의 일부",
+    "mrc-0-000561": "스탈린그라드 전투",
+    "mrc-0-004268": "제 3자",
+    "mrc-1-000835": "페이팔",
+    "mrc-0-001704": "가리타 히사노리에",
+    "mrc-1-001132": "점수",
+    "mrc-0-002512": "코스모케라톱스",
+    "mrc-0-003017": "사채회사",
+    "mrc-0-000215": "스위스",
+    "mrc-0-002873": "\"공유지의 비극\"",
+    "mrc-0-003118": "계급 모순",
+    "mrc-1-000384": "해리 트루먼 대통령",
+    "mrc-0-005270": "무위태수",
+    "mrc-0-002981": "크라운라이터",
+    "mrc-1-000158": "1990년",
+    "mrc-0-002189": "\"인간과 요괴의 완전한 평등\"",
+    "mrc-0-000905": "경애왕",
+    "mrc-0-001198": "도널드 트럼프 미국 대통령",
+    "mrc-0-003947": "바르톨로메오 브뤼기에르 신부",
+    "mrc-1-001328": "유교",
+    "mrc-0-000166": "경위대식 망원경",
+    "mrc-0-004090": "남큐슈 자동차 전용 도로",
+    "mrc-0-003522": "세실리아 페인",
+    "mrc-1-001398": "웹 2.0",
+    "mrc-0-000355": "수녀",
+    "mrc-0-002906": "맨해튼",
+    "mrc-0-001590": "상대성이론",
+    "mrc-1-001522": "대제 손권",
+    "mrc-0-004307": "마그넨티우스",
+    "mrc-0-002471": "제임스 뷰캐넌",
+    "mrc-1-001313": "트렁크",
+    "mrc-0-004083": "데코행진",
+    "mrc-0-005186": "약 600년",
+    "mrc-0-004197": "깃털 셔틀콕의 타구감을 선호하고, 또한 플라스틱보다 깃털 셔틀콕이 정교한 컨트롤을 하기에 보다 더 적합하기 때문",
+    "mrc-0-003208": "1998년",
+    "mrc-1-000297": "드라",
+    "mrc-1-000358": "가르미슈파르텐키르헨",
+    "mrc-1-001785": "정태적인 자본주의",
+    "mrc-0-005042": "데미안",
+    "mrc-1-000839": "목",
+    "mrc-0-004677": "34자",
+    "mrc-0-003564": "혼묘지",
+    "mrc-0-004202": "바르바로사 작전",
+    "mrc-0-001486": "동계건조(wintertrocken)",
+    "mrc-1-000291": "이질",
+    "mrc-0-003753": "중간 생���물",
+    "mrc-0-003033": "귀인 평향과 행동의 상관 관계",
+    "mrc-0-005478": "전류",
+    "mrc-0-005155": "기독교",
+    "mrc-1-000516": "1945년",
+    "mrc-0-002679": "브리튼인들",
+    "mrc-0-004527": "미국",
+    "mrc-0-001980": "8기의 수혈유구, 1기",
+    "mrc-0-004495": "전쟁",
+    "mrc-1-000037": "뇌졸중",
+    "mrc-0-004092": "김수환 추기경",
+    "mrc-0-004015": "4년 연상의 김옥성(金玉聲)과 결혼했다. 이후 그는 60년",
+    "mrc-0-000707": "펜실베이니아기",
+    "mrc-1-000658": "서남쪽",
+    "mrc-1-001446": "화이트 베이스",
+    "mrc-0-000412": "교수형",
+    "mrc-0-001288": "곽상",
+    "mrc-1-000197": "신도 케이",
+    "mrc-0-002253": "디오게네스",
+    "mrc-1-000367": "비밀 투표",
+    "mrc-0-003906": "111화",
+    "mrc-0-001254": "카누",
+    "mrc-0-003289": "알프스 전기 주식회사",
+    "mrc-0-004435": "크리스타",
+    "mrc-0-003844": "배상금",
+    "mrc-0-005355": "“어떤 경우에도 행할 수”",
+    "mrc-0-000521": "속도별 배치",
+    "mrc-1-000066": "1975년",
+    "mrc-0-001952": "제비초리형",
+    "mrc-0-002462": "석달",
+    "mrc-1-000899": "보은군 남쪽 상용리(현 영동군 용산면 상용리) 마을",
+    "mrc-0-002886": "태화관",
+    "mrc-0-004454": "거대 고래",
+    "mrc-0-001646": "하노이",
+    "mrc-1-001534": "토끼를 보고도 그냥 뛰어갔기에",
+    "mrc-1-000918": "좌우대칭",
+    "mrc-0-005046": "1913년",
+    "mrc-1-001611": "라자그리하",
+    "mrc-0-005222": "대령",
+    "mrc-0-004879": "걸프 카르텔",
+    "mrc-1-001481": "피터 케스카트 왓슨",
+    "mrc-0-003637": "충격파",
+    "mrc-0-003882": "대한민국 3보병사단",
+    "mrc-1-001285": "토머스 바클레이",
+    "mrc-0-005109": "감정 전염",
+    "mrc-1-001455": "스위치 이더넷",
+    "mrc-0-001274": "소비감소",
+    "mrc-1-000855": "마츠키 촌",
+    "mrc-0-003587": "환각, 망상, 사고장애",
+    "mrc-0-005031": "세포질",
+    "mrc-0-005154": "아케디아",
+    "mrc-0-000477": "과시하는 듯한 신앙심",
+    "mrc-0-002513": "소비자들의 반응",
+    "mrc-0-003022": "골룸바노",
+    "mrc-1-000946": "다케다",
+    "mrc-0-001240": "악당",
+    "mrc-0-001110": "유이엔",
+    "mrc-0-003149": "피네스 리먼",
+    "mrc-0-004132": "포스카르네트",
+    "mrc-0-005300": "면세증",
+    "mrc-0-004203": "숙의 정씨",
+    "mrc-0-003566": "남성연대 홈페이지",
+    "mrc-0-002926": "1895년",
+    "mrc-0-000536": "《천체의 회전에 관하여》",
+    "mrc-0-003848": "히로시마 종합 구장에서 열린 히로시마 도요 카프와의 경기에서 작은 논쟁으로 인해 현지의 히로시마",
+    "mrc-0-003760": "5·16 광장(여의도 광장)",
+    "mrc-0-001989": "6m",
+    "mrc-0-004863": "이등공송덕비건의소",
+    "mrc-1-000714": "헬리오시스",
+    "mrc-0-003211": "할머니",
+    "mrc-1-001173": "탄광 환경 문제",
+    "mrc-0-005360": "1796년",
+    "mrc-1-000096": "피자헛",
+    "mrc-0-003467": "가산",
+    "mrc-1-000593": "마거릿 대처",
+    "mrc-0-001001": "디지털 컴퓨터",
+    "mrc-1-001567": "35년간",
+    "mrc-1-000132": "문치미르",
+    "mrc-0-000787": "1947년",
+    "mrc-0-001494": "아이치 전기 전용선",
+    "mrc-0-003146": "'달빛 정원'(Moonlight Garden)",
+    "mrc-0-004411": "란다우 튜브",
+    "mrc-1-001622": "우천",
+    "mrc-0-002887": "＜젊은 근위대＞",
+    "mrc-0-003437": "10만필",
+    "mrc-1-001121": "기존 능력 성능 테스트를 기반으로 개발되며, 미래 능력 결정은 가능한 다양한 미래 시나리오의 도전을 해결할 수있는 군대",
+    "mrc-0-005286": "차고나라 전투",
+    "mrc-1-000295": "봅슬레이",
+    "mrc-0-000159": "일등병 토머스",
+    "mrc-0-004725": "1966년",
+    "mrc-0-004837": "클레몽트",
+    "mrc-1-000507": "17세기 조선 시대",
+    "mrc-0-001719": "창",
+    "mrc-0-001253": "수입 금지 조치",
+    "mrc-1-001270": "\"광휘에의 각성\"",
+    "mrc-0-005324": "카바레나 클럽",
+    "mrc-0-005105": "보수 성향",
+    "mrc-1-000983": "리퍼",
+    "mrc-0-002718": "국무회의",
+    "mrc-0-001552": "비단",
+    "mrc-0-003752": "사흘간",
+    "mrc-0-004530": "대과 급제자",
+    "mrc-0-003057": "오사카 시영 지하철",
+    "mrc-0-003850": "태조 왕건",
+    "mrc-0-003262": "맘루크를 몰아내기 위한 십자군 설교를 지시했다. 그 결과 1309년 7월 아비뇽 성문 앞에 일명 빈자의 십자군",
+    "mrc-0-001650": "멕시코",
+    "mrc-0-004662": "지질 과산화 생성물",
+    "mrc-0-003759": "독감",
+    "mrc-1-001279": "양당(楊黨)",
+    "mrc-0-001960": "슈바이츠 암 존탁",
+    "mrc-0-001162": "앨리스 페어살 스미스",
+    "mrc-0-004565": "다수결",
+    "mrc-0-000754": "코프먼과 저스티슨",
+    "mrc-1-000024": "물적 성과",
+    "mrc-0-000484": "제 양왕",
+    "mrc-0-002095": "'일곱 개의 신전 ���장'",
+    "mrc-0-003083": "윤치호",
+    "mrc-0-002978": "200,000명",
+    "mrc-1-000622": "오귀스트 블랑키(Auguste Blanqui)의 비밀결사주의"
+}

pytorch_model copy.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16a336762ed49849ec3f97bd08769e616293835b98b9041cb9e15b4bc1b38cd2
+size 1413985490

special_tokens_map copy.json

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

tokenizer_config copy.json

@@ -0,0 +1 @@
+{"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "do_basic_tokenize": true, "never_split": null, "bos_token": "[CLS]", "eos_token": "[SEP]", "model_max_length": 512, "tokenizer_class": "BertTokenizer", "special_tokens_map_file": "/opt/ml/.cache/huggingface/transformers/1a24ab4628028ed80dea35ce3334a636dc656fd9a17a09bad377f88f0cbecdac.70c17d6e4d492c8f24f5bb97ab56c7f272e947112c6faf9dd846da42ba13eb23", "name_or_path": "klue/roberta-large"}

train_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 10.0,
+    "init_mem_cpu_alloc_delta": 17783119,
+    "init_mem_cpu_peaked_delta": 18230,
+    "init_mem_gpu_alloc_delta": 1414861824,
+    "init_mem_gpu_peaked_delta": 67165696,
+    "train_mem_cpu_alloc_delta": 2991504,
+    "train_mem_cpu_peaked_delta": 264499229,
+    "train_mem_gpu_alloc_delta": 4256099328,
+    "train_mem_gpu_peaked_delta": 11330810880,
+    "train_runtime": 8323.4424,
+    "train_samples": 7978,
+    "train_samples_per_second": 1.199
+}

train_results.txt

@@ -0,0 +1,12 @@
+epoch = 10.0
+init_mem_cpu_alloc_delta = 17783119
+init_mem_cpu_peaked_delta = 18230
+init_mem_gpu_alloc_delta = 1414861824
+init_mem_gpu_peaked_delta = 67165696
+train_mem_cpu_alloc_delta = 2991504
+train_mem_cpu_peaked_delta = 264499229
+train_mem_gpu_alloc_delta = 4256099328
+train_mem_gpu_peaked_delta = 11330810880
+train_runtime = 8323.4424
+train_samples = 7978
+train_samples_per_second = 1.199

trainer_state copy.json

@@ -0,0 +1,617 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 10.0,
+  "global_step": 9980,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.997522904167843e-07,
+      "loss": 26.4633,
+      "step": 100
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.990094071072878e-07,
+      "loss": 9.8101,
+      "step": 200
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 9.977720861487699e-07,
+      "loss": 5.8569,
+      "step": 300
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 9.96041553526267e-07,
+      "loss": 5.5213,
+      "step": 400
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 9.938195239178374e-07,
+      "loss": 5.1344,
+      "step": 500
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 9.911081989955939e-07,
+      "loss": 4.9245,
+      "step": 600
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 9.879102652442023e-07,
+      "loss": 4.5925,
+      "step": 700
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 9.842288912990095e-07,
+      "loss": 4.676,
+      "step": 800
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 9.80067724806438e-07,
+      "loss": 4.4576,
+      "step": 900
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 9.754308888097582e-07,
+      "loss": 4.2986,
+      "step": 1000
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 9.703229776638185e-07,
+      "loss": 3.8739,
+      "step": 1100
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 9.647490524827833e-07,
+      "loss": 3.5629,
+      "step": 1200
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 9.587146361253867e-07,
+      "loss": 2.9575,
+      "step": 1300
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 9.522257077226716e-07,
+      "loss": 2.8326,
+      "step": 1400
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 9.452886967536388e-07,
+      "loss": 2.7891,
+      "step": 1500
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.379104766746722e-07,
+      "loss": 2.4971,
+      "step": 1600
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.30098358109054e-07,
+      "loss": 2.4297,
+      "step": 1700
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 9.218600816033199e-07,
+      "loss": 2.183,
+      "step": 1800
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 9.132038099576289e-07,
+      "loss": 2.0732,
+      "step": 1900
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 9.041381201377467e-07,
+      "loss": 2.2732,
+      "step": 2000
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 8.94671994776661e-07,
+      "loss": 1.7384,
+      "step": 2100
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 8.84814813274243e-07,
+      "loss": 1.8955,
+      "step": 2200
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 8.745763425037795e-07,
+      "loss": 2.0546,
+      "step": 2300
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 8.639667271345798e-07,
+      "loss": 1.6855,
+      "step": 2400
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 8.529964795802484e-07,
+      "loss": 1.7512,
+      "step": 2500
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 8.416764695825834e-07,
+      "loss": 1.7856,
+      "step": 2600
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 8.300179134414187e-07,
+      "loss": 1.7027,
+      "step": 2700
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 8.180323629010848e-07,
+      "loss": 1.5552,
+      "step": 2800
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 8.057316937044976e-07,
+      "loss": 1.6427,
+      "step": 2900
+    },
+    {
+      "epoch": 3.01,
+      "learning_rate": 7.931280938262168e-07,
+      "loss": 1.7996,
+      "step": 3000
+    },
+    {
+      "epoch": 3.11,
+      "learning_rate": 7.802340513961341e-07,
+      "loss": 1.4622,
+      "step": 3100
+    },
+    {
+      "epoch": 3.21,
+      "learning_rate": 7.670623423257547e-07,
+      "loss": 1.4137,
+      "step": 3200
+    },
+    {
+      "epoch": 3.31,
+      "learning_rate": 7.536260176493347e-07,
+      "loss": 1.4894,
+      "step": 3300
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 7.399383905924165e-07,
+      "loss": 1.4263,
+      "step": 3400
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 7.26013023380574e-07,
+      "loss": 1.491,
+      "step": 3500
+    },
+    {
+      "epoch": 3.61,
+      "learning_rate": 7.118637138014395e-07,
+      "loss": 1.2858,
+      "step": 3600
+    },
+    {
+      "epoch": 3.71,
+      "learning_rate": 6.975044815333281e-07,
+      "loss": 1.2851,
+      "step": 3700
+    },
+    {
+      "epoch": 3.81,
+      "learning_rate": 6.829495542540013e-07,
+      "loss": 1.3229,
+      "step": 3800
+    },
+    {
+      "epoch": 3.91,
+      "learning_rate": 6.682133535433393e-07,
+      "loss": 1.3675,
+      "step": 3900
+    },
+    {
+      "epoch": 4.01,
+      "learning_rate": 6.533104805938873e-07,
+      "loss": 1.3686,
+      "step": 4000
+    },
+    {
+      "epoch": 4.11,
+      "learning_rate": 6.382557017434331e-07,
+      "loss": 1.0781,
+      "step": 4100
+    },
+    {
+      "epoch": 4.21,
+      "learning_rate": 6.230639338439549e-07,
+      "loss": 1.1388,
+      "step": 4200
+    },
+    {
+      "epoch": 4.31,
+      "learning_rate": 6.077502294814311e-07,
+      "loss": 1.0966,
+      "step": 4300
+    },
+    {
+      "epoch": 4.41,
+      "learning_rate": 5.923297620611622e-07,
+      "loss": 1.1764,
+      "step": 4400
+    },
+    {
+      "epoch": 4.51,
+      "learning_rate": 5.76817810773379e-07,
+      "loss": 1.2625,
+      "step": 4500
+    },
+    {
+      "epoch": 4.61,
+      "learning_rate": 5.612297454540351e-07,
+      "loss": 1.1718,
+      "step": 4600
+    },
+    {
+      "epoch": 4.71,
+      "learning_rate": 5.455810113557839e-07,
+      "loss": 1.218,
+      "step": 4700
+    },
+    {
+      "epoch": 4.81,
+      "learning_rate": 5.298871138442307e-07,
+      "loss": 1.2203,
+      "step": 4800
+    },
+    {
+      "epoch": 4.91,
+      "learning_rate": 5.14163603034622e-07,
+      "loss": 1.0743,
+      "step": 4900
+    },
+    {
+      "epoch": 5.01,
+      "learning_rate": 4.984260583841952e-07,
+      "loss": 1.1825,
+      "step": 5000
+    },
+    {
+      "epoch": 5.11,
+      "learning_rate": 4.82690073255455e-07,
+      "loss": 0.8899,
+      "step": 5100
+    },
+    {
+      "epoch": 5.21,
+      "learning_rate": 4.6697123946567224e-07,
+      "loss": 0.9867,
+      "step": 5200
+    },
+    {
+      "epoch": 5.31,
+      "learning_rate": 4.512851318379138e-07,
+      "loss": 0.9842,
+      "step": 5300
+    },
+    {
+      "epoch": 5.41,
+      "learning_rate": 4.3564729276891087e-07,
+      "loss": 1.1441,
+      "step": 5400
+    },
+    {
+      "epoch": 5.51,
+      "learning_rate": 4.2007321682905594e-07,
+      "loss": 0.9965,
+      "step": 5500
+    },
+    {
+      "epoch": 5.61,
+      "learning_rate": 4.0457833540978923e-07,
+      "loss": 0.9929,
+      "step": 5600
+    },
+    {
+      "epoch": 5.71,
+      "learning_rate": 3.89178001433584e-07,
+      "loss": 1.152,
+      "step": 5700
+    },
+    {
+      "epoch": 5.81,
+      "learning_rate": 3.7388747414168295e-07,
+      "loss": 0.8896,
+      "step": 5800
+    },
+    {
+      "epoch": 5.91,
+      "learning_rate": 3.5872190397465635e-07,
+      "loss": 1.1022,
+      "step": 5900
+    },
+    {
+      "epoch": 6.01,
+      "learning_rate": 3.436963175607656e-07,
+      "loss": 0.8925,
+      "step": 6000
+    },
+    {
+      "epoch": 6.11,
+      "learning_rate": 3.288256028270033e-07,
+      "loss": 0.9647,
+      "step": 6100
+    },
+    {
+      "epoch": 6.21,
+      "learning_rate": 3.141244942475647e-07,
+      "loss": 0.7659,
+      "step": 6200
+    },
+    {
+      "epoch": 6.31,
+      "learning_rate": 2.996075582443658e-07,
+      "loss": 1.0098,
+      "step": 6300
+    },
+    {
+      "epoch": 6.41,
+      "learning_rate": 2.8528917875407433e-07,
+      "loss": 1.0704,
+      "step": 6400
+    },
+    {
+      "epoch": 6.51,
+      "learning_rate": 2.711835429759539e-07,
+      "loss": 1.0039,
+      "step": 6500
+    },
+    {
+      "epoch": 6.61,
+      "learning_rate": 2.573046273146427e-07,
+      "loss": 0.9493,
+      "step": 6600
+    },
+    {
+      "epoch": 6.71,
+      "learning_rate": 2.4366618353179644e-07,
+      "loss": 0.9582,
+      "step": 6700
+    },
+    {
+      "epoch": 6.81,
+      "learning_rate": 2.30281725120316e-07,
+      "loss": 0.8875,
+      "step": 6800
+    },
+    {
+      "epoch": 6.91,
+      "learning_rate": 2.1716451391466006e-07,
+      "loss": 0.8132,
+      "step": 6900
+    },
+    {
+      "epoch": 7.01,
+      "learning_rate": 2.0432754695051136e-07,
+      "loss": 1.0703,
+      "step": 7000
+    },
+    {
+      "epoch": 7.11,
+      "learning_rate": 1.9178354358681548e-07,
+      "loss": 1.0679,
+      "step": 7100
+    },
+    {
+      "epoch": 7.21,
+      "learning_rate": 1.7954493290295309e-07,
+      "loss": 0.8124,
+      "step": 7200
+    },
+    {
+      "epoch": 7.31,
+      "learning_rate": 1.6762384138353075e-07,
+      "loss": 0.9779,
+      "step": 7300
+    },
+    {
+      "epoch": 7.41,
+      "learning_rate": 1.5603208090299496e-07,
+      "loss": 0.7439,
+      "step": 7400
+    },
+    {
+      "epoch": 7.52,
+      "learning_rate": 1.4478113702197569e-07,
+      "loss": 0.858,
+      "step": 7500
+    },
+    {
+      "epoch": 7.62,
+      "learning_rate": 1.3388215760695098e-07,
+      "loss": 1.0146,
+      "step": 7600
+    },
+    {
+      "epoch": 7.72,
+      "learning_rate": 1.2334594178451424e-07,
+      "loss": 0.7714,
+      "step": 7700
+    },
+    {
+      "epoch": 7.82,
+      "learning_rate": 1.1318292924118584e-07,
+      "loss": 0.8889,
+      "step": 7800
+    },
+    {
+      "epoch": 7.92,
+      "learning_rate": 1.0340318987937097e-07,
+      "loss": 1.0001,
+      "step": 7900
+    },
+    {
+      "epoch": 8.02,
+      "learning_rate": 9.401641383971476e-08,
+      "loss": 0.9567,
+      "step": 8000
+    },
+    {
+      "epoch": 8.12,
+      "learning_rate": 8.503190189973914e-08,
+      "loss": 0.834,
+      "step": 8100
+    },
+    {
+      "epoch": 8.22,
+      "learning_rate": 7.645855625827657e-08,
+      "loss": 0.9169,
+      "step": 8200
+    },
+    {
+      "epoch": 8.32,
+      "learning_rate": 6.830487171482935e-08,
+      "loss": 0.841,
+      "step": 8300
+    },
+    {
+      "epoch": 8.42,
+      "learning_rate": 6.057892725259717e-08,
+      "loss": 1.0763,
+      "step": 8400
+    },
+    {
+      "epoch": 8.52,
+      "learning_rate": 5.328837803351083e-08,
+      "loss": 0.8885,
+      "step": 8500
+    },
+    {
+      "epoch": 8.62,
+      "learning_rate": 4.644044781320422e-08,
+      "loss": 0.7986,
+      "step": 8600
+    },
+    {
+      "epoch": 8.72,
+      "learning_rate": 4.004192178344029e-08,
+      "loss": 0.701,
+      "step": 8700
+    },
+    {
+      "epoch": 8.82,
+      "learning_rate": 3.4099139849083304e-08,
+      "loss": 0.9347,
+      "step": 8800
+    },
+    {
+      "epoch": 8.92,
+      "learning_rate": 2.8617990346277655e-08,
+      "loss": 1.0685,
+      "step": 8900
+    },
+    {
+      "epoch": 9.02,
+      "learning_rate": 2.3603904208058688e-08,
+      "loss": 0.8046,
+      "step": 9000
+    },
+    {
+      "epoch": 9.12,
+      "learning_rate": 1.9061849583176636e-08,
+      "loss": 0.8515,
+      "step": 9100
+    },
+    {
+      "epoch": 9.22,
+      "learning_rate": 1.499632691346375e-08,
+      "loss": 0.8998,
+      "step": 9200
+    },
+    {
+      "epoch": 9.32,
+      "learning_rate": 1.1411364474624264e-08,
+      "loss": 0.8923,
+      "step": 9300
+    },
+    {
+      "epoch": 9.42,
+      "learning_rate": 8.31051438486441e-09,
+      "loss": 0.7543,
+      "step": 9400
+    },
+    {
+      "epoch": 9.52,
+      "learning_rate": 5.696849085317645e-09,
+      "loss": 0.883,
+      "step": 9500
+    },
+    {
+      "epoch": 9.62,
+      "learning_rate": 3.5729582957520486e-09,
+      "loss": 0.8342,
+      "step": 9600
+    },
+    {
+      "epoch": 9.72,
+      "learning_rate": 1.9409464485766746e-09,
+      "loss": 0.9553,
+      "step": 9700
+    },
+    {
+      "epoch": 9.82,
+      "learning_rate": 8.024306036893968e-10,
+      "loss": 0.9193,
+      "step": 9800
+    },
+    {
+      "epoch": 9.92,
+      "learning_rate": 1.5853884623195922e-10,
+      "loss": 0.8958,
+      "step": 9900
+    },
+    {
+      "epoch": 10.0,
+      "step": 9980,
+      "total_flos": 0.0,
+      "train_runtime": 8323.4424,
+      "train_samples_per_second": 1.199
+    }
+  ],
+  "max_steps": 9980,
+  "num_train_epochs": 10,
+  "total_flos": 0.0,
+  "trial_name": null,
+  "trial_params": null
+}

trainer_state.json

@@ -1,130 +0,0 @@
-{
-  "best_metric": null,
-  "best_model_checkpoint": null,
-  "epoch": 9.519038076152304,
-  "global_step": 9500,
-  "is_hyper_param_search": false,
-  "is_local_process_zero": true,
-  "is_world_process_zero": true,
-  "log_history": [
-    {
-      "epoch": 0.5,
-      "learning_rate": 9.498997995991983e-07,
-      "loss": 10.6313,
-      "step": 500
-    },
-    {
-      "epoch": 1.0,
-      "learning_rate": 8.997995991983968e-07,
-      "loss": 5.4942,
-      "step": 1000
-    },
-    {
-      "epoch": 1.5,
-      "learning_rate": 8.496993987975952e-07,
-      "loss": 3.0541,
-      "step": 1500
-    },
-    {
-      "epoch": 2.0,
-      "learning_rate": 7.995991983967935e-07,
-      "loss": 2.0122,
-      "step": 2000
-    },
-    {
-      "epoch": 2.51,
-      "learning_rate": 7.494989979959919e-07,
-      "loss": 1.6182,
-      "step": 2500
-    },
-    {
-      "epoch": 3.01,
-      "learning_rate": 6.993987975951904e-07,
-      "loss": 1.6316,
-      "step": 3000
-    },
-    {
-      "epoch": 3.51,
-      "learning_rate": 6.492985971943887e-07,
-      "loss": 1.4137,
-      "step": 3500
-    },
-    {
-      "epoch": 4.01,
-      "learning_rate": 5.991983967935872e-07,
-      "loss": 1.2858,
-      "step": 4000
-    },
-    {
-      "epoch": 4.51,
-      "learning_rate": 5.490981963927856e-07,
-      "loss": 1.1587,
-      "step": 4500
-    },
-    {
-      "epoch": 5.01,
-      "learning_rate": 4.989979959919839e-07,
-      "loss": 1.1424,
-      "step": 5000
-    },
-    {
-      "epoch": 5.51,
-      "learning_rate": 4.4889779559118236e-07,
-      "loss": 0.9812,
-      "step": 5500
-    },
-    {
-      "epoch": 6.01,
-      "learning_rate": 3.987975951903808e-07,
-      "loss": 1.0498,
-      "step": 6000
-    },
-    {
-      "epoch": 6.51,
-      "learning_rate": 3.4869739478957914e-07,
-      "loss": 0.9131,
-      "step": 6500
-    },
-    {
-      "epoch": 7.01,
-      "learning_rate": 2.9859719438877756e-07,
-      "loss": 0.9897,
-      "step": 7000
-    },
-    {
-      "epoch": 7.52,
-      "learning_rate": 2.4849699398797593e-07,
-      "loss": 0.8315,
-      "step": 7500
-    },
-    {
-      "epoch": 8.02,
-      "learning_rate": 1.9839679358717435e-07,
-      "loss": 0.9413,
-      "step": 8000
-    },
-    {
-      "epoch": 8.52,
-      "learning_rate": 1.4829659318637274e-07,
-      "loss": 0.9108,
-      "step": 8500
-    },
-    {
-      "epoch": 9.02,
-      "learning_rate": 9.819639278557115e-08,
-      "loss": 0.7817,
-      "step": 9000
-    },
-    {
-      "epoch": 9.52,
-      "learning_rate": 4.809619238476954e-08,
-      "loss": 0.7999,
-      "step": 9500
-    }
-  ],
-  "max_steps": 9980,
-  "num_train_epochs": 10,
-  "total_flos": 0.0,
-  "trial_name": null,
-  "trial_params": null
-}

rng_state.pth → training_args copy.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7a2d729fdcde9734ee5b1b7d1c85b821418dfebfff8b57f99daf6a35b9e78ee3
-size 14657
+oid sha256:574e5e2a956b3e59901c3ed9848a90b98bd67a72f27a5e69b189b003d4e05fd4
+size 2287