Upload 11 files

config.json

@@ -1,11 +1,12 @@
 {
-  "_name_or_path": "monologg/koelectra-base-v3-discriminator",
+  "_name_or_path": "skt/kobert-base-v1",
   "architectures": [
-    "ElectraForSequenceClassification"
+    "BertForSequenceClassification"
   ],
   "attention_probs_dropout_prob": 0.1,
+  "author": "Heewon Jeon([email protected])",
   "classifier_dropout": null,
-  "embedding_size": 768,
+  "gradient_checkpointing": false,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
@@ -16,6 +17,7 @@
   },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
+  "kobert_version": 1.0,
   "label2id": {
     "LABEL_0": 0,
     "LABEL_1": 1,
@@ -23,19 +25,15 @@
   },
   "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
-  "model_type": "electra",
+  "model_type": "bert",
   "num_attention_heads": 12,
   "num_hidden_layers": 12,
-  "pad_token_id": 0,
+  "pad_token_id": 1,
   "position_embedding_type": "absolute",
   "problem_type": "single_label_classification",
-  "summary_activation": "gelu",
-  "summary_last_dropout": 0.1,
-  "summary_type": "first",
-  "summary_use_proj": true,
   "torch_dtype": "float32",
   "transformers_version": "4.45.1",
   "type_vocab_size": 2,
   "use_cache": true,
-  "vocab_size": 35000
+  "vocab_size": 8002
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b7ce6384be92f1c7f441e7d36892fd4af2d8942446e7ec62ab1300075d6eead1
-size 451718748
+oid sha256:f82d7a65656d1912f2914baa8290c48b9c55044f1f05bcafb8e1ee2590b9dd0a
+size 368780204

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5080a9a6edc7288b55de6518de49d2b65f4886fd9e8e3d9c400a0b0a9920cdda
-size 903551738
+oid sha256:af29b42c6fce25f4772e217cb873f5bcb9dff8b526dd14d37818bb174701e2cb
+size 737676026

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:38bdd138de6a56a2479d7804e54bf21693ad07775ca65e9e18d4fd68b3b67efc
+oid sha256:d7e2a646ca72077b2f4e495817913734817a4d0b59c43adf947fc829113f0db1
 size 13990

special_tokens_map.json

@@ -1,6 +1,14 @@
 {
+  "bos_token": "[CLS]",
   "cls_token": "[CLS]",
-  "mask_token": "[MASK]",
+  "eos_token": "[SEP]",
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
   "pad_token": "[PAD]",
   "sep_token": "[SEP]",
   "unk_token": "[UNK]"

spiece.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17dc471055592d3cc9e0a5831e769246a8a001a4d27551c9ed79668173c7b407
+size 371427

tokenizer_config.json

@@ -1,7 +1,7 @@
 {
   "added_tokens_decoder": {
     "0": {
-      "content": "[PAD]",
+      "content": "[UNK]",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -9,12 +9,12 @@
       "special": true
     },
     "1": {
-      "content": "[UNK]",
+      "content": "[PAD]",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
       "single_word": false,
-      "special": true
+      "special": false
     },
     "2": {
       "content": "[CLS]",
@@ -22,7 +22,7 @@
       "normalized": false,
       "rstrip": false,
       "single_word": false,
-      "special": true
+      "special": false
     },
     "3": {
       "content": "[SEP]",
@@ -30,29 +30,31 @@
       "normalized": false,
       "rstrip": false,
       "single_word": false,
-      "special": true
+      "special": false
     },
     "4": {
       "content": "[MASK]",
-      "lstrip": false,
-      "normalized": false,
+      "lstrip": true,
+      "normalized": true,
       "rstrip": false,
       "single_word": false,
       "special": true
     }
   },
-  "clean_up_tokenization_spaces": true,
+  "additional_special_tokens": [],
+  "bos_token": "[CLS]",
+  "clean_up_tokenization_spaces": false,
   "cls_token": "[CLS]",
-  "do_basic_tokenize": true,
   "do_lower_case": false,
+  "eos_token": "[SEP]",
+  "keep_accents": false,
   "mask_token": "[MASK]",
-  "model_max_length": 512,
-  "never_split": null,
+  "model_max_length": 1000000000000000019884624838656,
   "pad_token": "[PAD]",
+  "remove_space": true,
   "sep_token": "[SEP]",
-  "strip_accents": null,
+  "sp_model_kwargs": {},
   "timeout": 60,
-  "tokenize_chinese_chars": true,
-  "tokenizer_class": "ElectraTokenizer",
+  "tokenizer_class": "XLNetTokenizer",
   "unk_token": "[UNK]"
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.9008317338451696,
-  "best_model_checkpoint": "./checkpoints/monologg_koelectra-base-v3-discriminator\\checkpoint-4689",
+  "best_metric": 0.6842610364683301,
+  "best_model_checkpoint": "./checkpoints/skt_kobert-base-v1\\checkpoint-3126",
   "epoch": 3.0,
   "eval_steps": 500,
   "global_step": 4689,
@@ -10,683 +10,683 @@
   "log_history": [
     {
       "epoch": 0.03198976327575176,
-      "grad_norm": 1.8438289165496826,
+      "grad_norm": 2.0306639671325684,
       "learning_rate": 1.9786734911494988e-05,
-      "loss": 0.9998,
+      "loss": 1.013,
       "step": 50
     },
     {
       "epoch": 0.06397952655150352,
-      "grad_norm": 4.708126544952393,
+      "grad_norm": 2.591745138168335,
       "learning_rate": 1.9573469822989978e-05,
-      "loss": 0.8181,
+      "loss": 0.9463,
       "step": 100
     },
     {
       "epoch": 0.09596928982725528,
-      "grad_norm": 7.408482551574707,
+      "grad_norm": 2.405771017074585,
       "learning_rate": 1.9360204734484968e-05,
-      "loss": 0.6265,
+      "loss": 0.912,
       "step": 150
     },
     {
       "epoch": 0.12795905310300704,
-      "grad_norm": 6.211757183074951,
+      "grad_norm": 3.699129104614258,
       "learning_rate": 1.9146939645979955e-05,
-      "loss": 0.5046,
+      "loss": 0.8349,
       "step": 200
     },
     {
       "epoch": 0.1599488163787588,
-      "grad_norm": 7.470839500427246,
+      "grad_norm": 3.0695576667785645,
       "learning_rate": 1.8933674557474945e-05,
-      "loss": 0.5358,
+      "loss": 0.8967,
       "step": 250
     },
     {
       "epoch": 0.19193857965451055,
-      "grad_norm": 2.0797882080078125,
+      "grad_norm": 2.762423038482666,
       "learning_rate": 1.872040946896993e-05,
-      "loss": 0.483,
+      "loss": 0.8609,
       "step": 300
     },
     {
       "epoch": 0.22392834293026231,
-      "grad_norm": 14.55879020690918,
+      "grad_norm": 5.626348495483398,
       "learning_rate": 1.8507144380464918e-05,
-      "loss": 0.4462,
+      "loss": 0.8619,
       "step": 350
     },
     {
       "epoch": 0.2559181062060141,
-      "grad_norm": 3.5867443084716797,
+      "grad_norm": 2.3783187866210938,
       "learning_rate": 1.8293879291959908e-05,
-      "loss": 0.4348,
+      "loss": 0.847,
       "step": 400
     },
     {
       "epoch": 0.28790786948176583,
-      "grad_norm": 11.995755195617676,
+      "grad_norm": 2.9742441177368164,
       "learning_rate": 1.8080614203454897e-05,
-      "loss": 0.4236,
+      "loss": 0.8161,
       "step": 450
     },
     {
       "epoch": 0.3198976327575176,
-      "grad_norm": 7.211252689361572,
+      "grad_norm": 1.8092529773712158,
       "learning_rate": 1.7867349114949884e-05,
-      "loss": 0.4099,
+      "loss": 0.7752,
       "step": 500
     },
     {
       "epoch": 0.35188739603326935,
-      "grad_norm": 5.884838581085205,
+      "grad_norm": 2.72413969039917,
       "learning_rate": 1.7654084026444874e-05,
-      "loss": 0.4265,
+      "loss": 0.8281,
       "step": 550
     },
     {
       "epoch": 0.3838771593090211,
-      "grad_norm": 4.443331241607666,
+      "grad_norm": 3.9095618724823,
       "learning_rate": 1.744081893793986e-05,
-      "loss": 0.3863,
+      "loss": 0.7984,
       "step": 600
     },
     {
       "epoch": 0.41586692258477287,
-      "grad_norm": 3.68717622756958,
+      "grad_norm": 3.821829319000244,
       "learning_rate": 1.7227553849434847e-05,
-      "loss": 0.4244,
+      "loss": 0.845,
       "step": 650
     },
     {
       "epoch": 0.44785668586052463,
-      "grad_norm": 4.4258599281311035,
+      "grad_norm": 6.882596492767334,
       "learning_rate": 1.7014288760929837e-05,
-      "loss": 0.3701,
+      "loss": 0.8218,
       "step": 700
     },
     {
       "epoch": 0.4798464491362764,
-      "grad_norm": 8.484159469604492,
+      "grad_norm": 3.819617986679077,
       "learning_rate": 1.6801023672424827e-05,
-      "loss": 0.4079,
+      "loss": 0.8316,
       "step": 750
     },
     {
       "epoch": 0.5118362124120281,
-      "grad_norm": 8.976361274719238,
+      "grad_norm": 2.8060302734375,
       "learning_rate": 1.6587758583919813e-05,
-      "loss": 0.3413,
+      "loss": 0.8009,
       "step": 800
     },
     {
       "epoch": 0.5438259756877799,
-      "grad_norm": 7.113468170166016,
+      "grad_norm": 6.112952709197998,
       "learning_rate": 1.6374493495414803e-05,
-      "loss": 0.3928,
+      "loss": 0.7959,
       "step": 850
     },
     {
       "epoch": 0.5758157389635317,
-      "grad_norm": 7.242956161499023,
+      "grad_norm": 6.098555564880371,
       "learning_rate": 1.616122840690979e-05,
-      "loss": 0.4013,
+      "loss": 0.822,
       "step": 900
     },
     {
       "epoch": 0.6078055022392834,
-      "grad_norm": 6.204492092132568,
+      "grad_norm": 8.28978157043457,
       "learning_rate": 1.5947963318404776e-05,
-      "loss": 0.3838,
+      "loss": 0.7781,
       "step": 950
     },
     {
       "epoch": 0.6397952655150352,
-      "grad_norm": 5.312352180480957,
+      "grad_norm": 3.776757001876831,
       "learning_rate": 1.5734698229899766e-05,
-      "loss": 0.3838,
+      "loss": 0.8017,
       "step": 1000
     },
     {
       "epoch": 0.6717850287907869,
-      "grad_norm": 22.331918716430664,
+      "grad_norm": 2.7059381008148193,
       "learning_rate": 1.5521433141394756e-05,
-      "loss": 0.3618,
+      "loss": 0.7822,
       "step": 1050
     },
     {
       "epoch": 0.7037747920665387,
-      "grad_norm": 11.372283935546875,
+      "grad_norm": 6.732589244842529,
       "learning_rate": 1.5308168052889743e-05,
-      "loss": 0.3486,
+      "loss": 0.7965,
       "step": 1100
     },
     {
       "epoch": 0.7357645553422905,
-      "grad_norm": 31.921972274780273,
+      "grad_norm": 8.803948402404785,
       "learning_rate": 1.5094902964384733e-05,
-      "loss": 0.3807,
+      "loss": 0.772,
       "step": 1150
     },
     {
       "epoch": 0.7677543186180422,
-      "grad_norm": 6.360013484954834,
+      "grad_norm": 3.7579758167266846,
       "learning_rate": 1.488163787587972e-05,
-      "loss": 0.3903,
+      "loss": 0.877,
       "step": 1200
     },
     {
       "epoch": 0.799744081893794,
-      "grad_norm": 5.0191330909729,
+      "grad_norm": 3.5114402770996094,
       "learning_rate": 1.4668372787374708e-05,
-      "loss": 0.3477,
+      "loss": 0.7834,
       "step": 1250
     },
     {
       "epoch": 0.8317338451695457,
-      "grad_norm": 20.319231033325195,
+      "grad_norm": 3.5781891345977783,
       "learning_rate": 1.4455107698869698e-05,
-      "loss": 0.3193,
+      "loss": 0.7746,
       "step": 1300
     },
     {
       "epoch": 0.8637236084452975,
-      "grad_norm": 3.2279179096221924,
+      "grad_norm": 4.3330278396606445,
       "learning_rate": 1.4241842610364684e-05,
-      "loss": 0.3583,
+      "loss": 0.8138,
       "step": 1350
     },
     {
       "epoch": 0.8957133717210493,
-      "grad_norm": 5.339118480682373,
+      "grad_norm": 3.0052335262298584,
       "learning_rate": 1.4028577521859672e-05,
-      "loss": 0.3343,
+      "loss": 0.7651,
       "step": 1400
     },
     {
       "epoch": 0.927703134996801,
-      "grad_norm": 3.4250569343566895,
+      "grad_norm": 3.8835694789886475,
       "learning_rate": 1.3815312433354662e-05,
-      "loss": 0.3756,
+      "loss": 0.7984,
       "step": 1450
     },
     {
       "epoch": 0.9596928982725528,
-      "grad_norm": 7.723750591278076,
+      "grad_norm": 5.093181133270264,
       "learning_rate": 1.3602047344849649e-05,
-      "loss": 0.3342,
+      "loss": 0.7943,
       "step": 1500
     },
     {
       "epoch": 0.9916826615483045,
-      "grad_norm": 9.071324348449707,
+      "grad_norm": 2.615415096282959,
       "learning_rate": 1.3388782256344637e-05,
-      "loss": 0.3282,
+      "loss": 0.7568,
       "step": 1550
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.8848368522072937,
-      "eval_f1": 0.8749968140479337,
-      "eval_loss": 0.3197398781776428,
-      "eval_runtime": 69.8877,
-      "eval_samples_per_second": 44.729,
-      "eval_steps_per_second": 2.804,
+      "eval_accuracy": 0.673064619321817,
+      "eval_f1": 0.5774602220992764,
+      "eval_loss": 0.755181610584259,
+      "eval_runtime": 146.1897,
+      "eval_samples_per_second": 21.383,
+      "eval_steps_per_second": 1.341,
       "step": 1563
     },
     {
       "epoch": 1.0236724248240563,
-      "grad_norm": 3.9245243072509766,
+      "grad_norm": 3.9289746284484863,
       "learning_rate": 1.3175517167839627e-05,
-      "loss": 0.2311,
+      "loss": 0.7968,
       "step": 1600
     },
     {
       "epoch": 1.055662188099808,
-      "grad_norm": 13.951614379882812,
+      "grad_norm": 3.1284472942352295,
       "learning_rate": 1.2962252079334613e-05,
-      "loss": 0.2844,
+      "loss": 0.7693,
       "step": 1650
     },
     {
       "epoch": 1.0876519513755598,
-      "grad_norm": 8.549053192138672,
+      "grad_norm": 6.5303544998168945,
       "learning_rate": 1.2748986990829602e-05,
-      "loss": 0.2573,
+      "loss": 0.7763,
       "step": 1700
     },
     {
       "epoch": 1.1196417146513116,
-      "grad_norm": 7.1960835456848145,
+      "grad_norm": 4.240009784698486,
       "learning_rate": 1.2535721902324592e-05,
-      "loss": 0.2515,
+      "loss": 0.7499,
       "step": 1750
     },
     {
       "epoch": 1.1516314779270633,
-      "grad_norm": 3.4280333518981934,
+      "grad_norm": 3.231553554534912,
       "learning_rate": 1.2322456813819578e-05,
-      "loss": 0.3032,
+      "loss": 0.8081,
       "step": 1800
     },
     {
       "epoch": 1.183621241202815,
-      "grad_norm": 7.896517753601074,
+      "grad_norm": 3.248556613922119,
       "learning_rate": 1.2109191725314566e-05,
-      "loss": 0.2503,
+      "loss": 0.7372,
       "step": 1850
     },
     {
       "epoch": 1.2156110044785668,
-      "grad_norm": 3.302367687225342,
+      "grad_norm": 4.725924968719482,
       "learning_rate": 1.1895926636809556e-05,
-      "loss": 0.2815,
+      "loss": 0.7564,
       "step": 1900
     },
     {
       "epoch": 1.2476007677543186,
-      "grad_norm": 7.449910640716553,
+      "grad_norm": 7.154216766357422,
       "learning_rate": 1.1682661548304543e-05,
-      "loss": 0.2483,
+      "loss": 0.7358,
       "step": 1950
     },
     {
       "epoch": 1.2795905310300704,
-      "grad_norm": 6.277085304260254,
+      "grad_norm": 5.529367923736572,
       "learning_rate": 1.1469396459799531e-05,
-      "loss": 0.2709,
+      "loss": 0.7921,
       "step": 2000
     },
     {
       "epoch": 1.3115802943058221,
-      "grad_norm": 4.672796726226807,
+      "grad_norm": 2.6566271781921387,
       "learning_rate": 1.1256131371294521e-05,
-      "loss": 0.2555,
+      "loss": 0.7684,
       "step": 2050
     },
     {
       "epoch": 1.3435700575815739,
-      "grad_norm": 11.666603088378906,
+      "grad_norm": 3.7145371437072754,
       "learning_rate": 1.1042866282789508e-05,
-      "loss": 0.2381,
+      "loss": 0.7612,
       "step": 2100
     },
     {
       "epoch": 1.3755598208573256,
-      "grad_norm": 9.201951026916504,
+      "grad_norm": 4.2554707527160645,
       "learning_rate": 1.0829601194284496e-05,
-      "loss": 0.2762,
+      "loss": 0.7622,
       "step": 2150
     },
     {
       "epoch": 1.4075495841330774,
-      "grad_norm": 2.2189481258392334,
+      "grad_norm": 3.207913398742676,
       "learning_rate": 1.0616336105779486e-05,
-      "loss": 0.2466,
+      "loss": 0.7331,
       "step": 2200
     },
     {
       "epoch": 1.4395393474088292,
-      "grad_norm": 8.02763557434082,
+      "grad_norm": 7.0538153648376465,
       "learning_rate": 1.0403071017274472e-05,
-      "loss": 0.2556,
+      "loss": 0.7578,
       "step": 2250
     },
     {
       "epoch": 1.471529110684581,
-      "grad_norm": 2.0652530193328857,
+      "grad_norm": 4.054543972015381,
       "learning_rate": 1.018980592876946e-05,
-      "loss": 0.2789,
+      "loss": 0.7711,
       "step": 2300
     },
     {
       "epoch": 1.5035188739603327,
-      "grad_norm": 13.113646507263184,
+      "grad_norm": 5.265899658203125,
       "learning_rate": 9.976540840264449e-06,
-      "loss": 0.2235,
+      "loss": 0.742,
       "step": 2350
     },
     {
       "epoch": 1.5355086372360844,
-      "grad_norm": 10.362279891967773,
+      "grad_norm": 5.515292167663574,
       "learning_rate": 9.763275751759437e-06,
-      "loss": 0.2482,
+      "loss": 0.763,
       "step": 2400
     },
     {
       "epoch": 1.5674984005118362,
-      "grad_norm": 1.568915843963623,
+      "grad_norm": 3.4068896770477295,
       "learning_rate": 9.550010663254427e-06,
-      "loss": 0.2622,
+      "loss": 0.7787,
       "step": 2450
     },
     {
       "epoch": 1.599488163787588,
-      "grad_norm": 10.095715522766113,
+      "grad_norm": 3.8786113262176514,
       "learning_rate": 9.336745574749414e-06,
-      "loss": 0.2668,
+      "loss": 0.7665,
       "step": 2500
     },
     {
       "epoch": 1.6314779270633397,
-      "grad_norm": 12.999874114990234,
+      "grad_norm": 7.389112949371338,
       "learning_rate": 9.123480486244403e-06,
-      "loss": 0.2361,
+      "loss": 0.7324,
       "step": 2550
     },
     {
       "epoch": 1.6634676903390915,
-      "grad_norm": 7.512565612792969,
+      "grad_norm": 3.790463924407959,
       "learning_rate": 8.910215397739392e-06,
-      "loss": 0.2646,
+      "loss": 0.7479,
       "step": 2600
     },
     {
       "epoch": 1.6954574536148432,
-      "grad_norm": 6.645097732543945,
+      "grad_norm": 3.296069622039795,
       "learning_rate": 8.696950309234378e-06,
-      "loss": 0.2478,
+      "loss": 0.6861,
       "step": 2650
     },
     {
       "epoch": 1.727447216890595,
-      "grad_norm": 4.22629976272583,
+      "grad_norm": 3.9884438514709473,
       "learning_rate": 8.483685220729368e-06,
-      "loss": 0.2686,
+      "loss": 0.7621,
       "step": 2700
     },
     {
       "epoch": 1.7594369801663468,
-      "grad_norm": 8.55104923248291,
+      "grad_norm": 4.0090532302856445,
       "learning_rate": 8.270420132224356e-06,
-      "loss": 0.2615,
+      "loss": 0.7455,
       "step": 2750
     },
     {
       "epoch": 1.7914267434420985,
-      "grad_norm": 4.150308132171631,
+      "grad_norm": 5.72037935256958,
       "learning_rate": 8.057155043719343e-06,
-      "loss": 0.2669,
+      "loss": 0.7615,
       "step": 2800
     },
     {
       "epoch": 1.8234165067178503,
-      "grad_norm": 15.736576080322266,
+      "grad_norm": 5.367597579956055,
       "learning_rate": 7.843889955214333e-06,
-      "loss": 0.252,
+      "loss": 0.7478,
       "step": 2850
     },
     {
       "epoch": 1.855406269993602,
-      "grad_norm": 1.219308853149414,
+      "grad_norm": 4.492830753326416,
       "learning_rate": 7.630624866709321e-06,
-      "loss": 0.2619,
+      "loss": 0.7644,
       "step": 2900
     },
     {
       "epoch": 1.8873960332693538,
-      "grad_norm": 4.5792999267578125,
+      "grad_norm": 4.395884990692139,
       "learning_rate": 7.4173597782043085e-06,
-      "loss": 0.2516,
+      "loss": 0.726,
       "step": 2950
     },
     {
       "epoch": 1.9193857965451055,
-      "grad_norm": 7.346240043640137,
+      "grad_norm": 5.388432025909424,
       "learning_rate": 7.204094689699297e-06,
-      "loss": 0.2374,
+      "loss": 0.7305,
       "step": 3000
     },
     {
       "epoch": 1.9513755598208573,
-      "grad_norm": 15.958673477172852,
+      "grad_norm": 3.6554412841796875,
       "learning_rate": 6.990829601194286e-06,
-      "loss": 0.2568,
+      "loss": 0.7312,
       "step": 3050
     },
     {
       "epoch": 1.983365323096609,
-      "grad_norm": 8.143491744995117,
+      "grad_norm": 2.9544596672058105,
       "learning_rate": 6.777564512689273e-06,
-      "loss": 0.2196,
+      "loss": 0.7535,
       "step": 3100
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8902751119641714,
-      "eval_f1": 0.8821468217693207,
-      "eval_loss": 0.35885584354400635,
-      "eval_runtime": 70.4611,
-      "eval_samples_per_second": 44.365,
-      "eval_steps_per_second": 2.782,
+      "eval_accuracy": 0.6842610364683301,
+      "eval_f1": 0.5914024960556081,
+      "eval_loss": 0.71613609790802,
+      "eval_runtime": 145.4665,
+      "eval_samples_per_second": 21.489,
+      "eval_steps_per_second": 1.347,
       "step": 3126
     },
     {
       "epoch": 2.015355086372361,
-      "grad_norm": 2.4161245822906494,
+      "grad_norm": 2.3898167610168457,
       "learning_rate": 6.5642994241842614e-06,
-      "loss": 0.211,
+      "loss": 0.7209,
       "step": 3150
     },
     {
       "epoch": 2.0473448496481126,
-      "grad_norm": 6.2580156326293945,
+      "grad_norm": 3.216728448867798,
       "learning_rate": 6.3510343356792505e-06,
-      "loss": 0.1819,
+      "loss": 0.7328,
       "step": 3200
     },
     {
       "epoch": 2.0793346129238643,
-      "grad_norm": 17.149974822998047,
+      "grad_norm": 5.2315673828125,
       "learning_rate": 6.137769247174238e-06,
-      "loss": 0.188,
+      "loss": 0.6979,
       "step": 3250
     },
     {
       "epoch": 2.111324376199616,
-      "grad_norm": 4.2034382820129395,
+      "grad_norm": 5.086841583251953,
       "learning_rate": 5.924504158669226e-06,
-      "loss": 0.1653,
+      "loss": 0.7401,
       "step": 3300
     },
     {
       "epoch": 2.143314139475368,
-      "grad_norm": 7.357328414916992,
+      "grad_norm": 3.975651502609253,
       "learning_rate": 5.711239070164215e-06,
-      "loss": 0.1748,
+      "loss": 0.7695,
       "step": 3350
     },
     {
       "epoch": 2.1753039027511196,
-      "grad_norm": 13.677566528320312,
+      "grad_norm": 5.14952278137207,
       "learning_rate": 5.497973981659203e-06,
-      "loss": 0.2177,
+      "loss": 0.7259,
       "step": 3400
     },
     {
       "epoch": 2.2072936660268714,
-      "grad_norm": 8.67300033569336,
+      "grad_norm": 10.398428916931152,
       "learning_rate": 5.284708893154191e-06,
-      "loss": 0.1571,
+      "loss": 0.7099,
       "step": 3450
     },
     {
       "epoch": 2.239283429302623,
-      "grad_norm": 22.893770217895508,
+      "grad_norm": 5.051026821136475,
       "learning_rate": 5.07144380464918e-06,
-      "loss": 0.2009,
+      "loss": 0.7256,
       "step": 3500
     },
     {
       "epoch": 2.271273192578375,
-      "grad_norm": 3.602328300476074,
+      "grad_norm": 5.900022506713867,
       "learning_rate": 4.858178716144167e-06,
-      "loss": 0.1329,
+      "loss": 0.6803,
       "step": 3550
     },
     {
       "epoch": 2.3032629558541267,
-      "grad_norm": 18.8084659576416,
+      "grad_norm": 3.650153636932373,
       "learning_rate": 4.644913627639156e-06,
-      "loss": 0.1814,
+      "loss": 0.6936,
       "step": 3600
     },
     {
       "epoch": 2.3352527191298784,
-      "grad_norm": 0.4535435736179352,
+      "grad_norm": 3.192567825317383,
       "learning_rate": 4.431648539134144e-06,
-      "loss": 0.2069,
+      "loss": 0.7388,
       "step": 3650
     },
     {
       "epoch": 2.36724248240563,
-      "grad_norm": 11.971237182617188,
+      "grad_norm": 2.792283058166504,
       "learning_rate": 4.218383450629132e-06,
-      "loss": 0.188,
+      "loss": 0.7199,
       "step": 3700
     },
     {
       "epoch": 2.399232245681382,
-      "grad_norm": 5.598160266876221,
+      "grad_norm": 4.723822593688965,
       "learning_rate": 4.005118362124121e-06,
-      "loss": 0.1943,
+      "loss": 0.7536,
       "step": 3750
     },
     {
       "epoch": 2.4312220089571337,
-      "grad_norm": 11.97499942779541,
+      "grad_norm": 6.793713092803955,
       "learning_rate": 3.791853273619109e-06,
-      "loss": 0.2155,
+      "loss": 0.7043,
       "step": 3800
     },
     {
       "epoch": 2.4632117722328855,
-      "grad_norm": 9.149175643920898,
+      "grad_norm": 4.855713367462158,
       "learning_rate": 3.5785881851140968e-06,
-      "loss": 0.2025,
+      "loss": 0.7587,
       "step": 3850
     },
     {
       "epoch": 2.495201535508637,
-      "grad_norm": 7.965933799743652,
+      "grad_norm": 3.8809781074523926,
       "learning_rate": 3.3653230966090854e-06,
-      "loss": 0.2121,
+      "loss": 0.7007,
       "step": 3900
     },
     {
       "epoch": 2.527191298784389,
-      "grad_norm": 9.397235870361328,
+      "grad_norm": 3.591956377029419,
       "learning_rate": 3.1520580081040737e-06,
-      "loss": 0.1724,
+      "loss": 0.6805,
       "step": 3950
     },
     {
       "epoch": 2.5591810620601407,
-      "grad_norm": 14.580031394958496,
+      "grad_norm": 4.135153293609619,
       "learning_rate": 2.9387929195990615e-06,
-      "loss": 0.1636,
+      "loss": 0.7628,
       "step": 4000
     },
     {
       "epoch": 2.5911708253358925,
-      "grad_norm": 2.2896785736083984,
+      "grad_norm": 4.887539386749268,
       "learning_rate": 2.72552783109405e-06,
-      "loss": 0.1603,
+      "loss": 0.7203,
       "step": 4050
     },
     {
       "epoch": 2.6231605886116443,
-      "grad_norm": 22.936344146728516,
+      "grad_norm": 5.000283241271973,
       "learning_rate": 2.5122627425890384e-06,
-      "loss": 0.1944,
+      "loss": 0.709,
       "step": 4100
     },
     {
       "epoch": 2.655150351887396,
-      "grad_norm": 37.436187744140625,
+      "grad_norm": 5.802310943603516,
       "learning_rate": 2.2989976540840266e-06,
-      "loss": 0.2059,
+      "loss": 0.6938,
       "step": 4150
     },
     {
       "epoch": 2.6871401151631478,
-      "grad_norm": 26.6770076751709,
+      "grad_norm": 2.909243583679199,
       "learning_rate": 2.085732565579015e-06,
-      "loss": 0.1448,
+      "loss": 0.7236,
       "step": 4200
     },
     {
       "epoch": 2.7191298784388995,
-      "grad_norm": 20.561311721801758,
+      "grad_norm": 3.879182815551758,
       "learning_rate": 1.872467477074003e-06,
-      "loss": 0.2094,
+      "loss": 0.7327,
       "step": 4250
     },
     {
       "epoch": 2.7511196417146513,
-      "grad_norm": 4.114218711853027,
+      "grad_norm": 5.3820295333862305,
       "learning_rate": 1.6592023885689915e-06,
-      "loss": 0.1682,
+      "loss": 0.6762,
       "step": 4300
     },
     {
       "epoch": 2.783109404990403,
-      "grad_norm": 27.9166202545166,
+      "grad_norm": 4.583397388458252,
       "learning_rate": 1.4459373000639796e-06,
-      "loss": 0.1931,
+      "loss": 0.7111,
       "step": 4350
     },
     {
       "epoch": 2.815099168266155,
-      "grad_norm": 13.576592445373535,
+      "grad_norm": 3.990649938583374,
       "learning_rate": 1.2326722115589678e-06,
-      "loss": 0.1989,
+      "loss": 0.7177,
       "step": 4400
     },
     {
       "epoch": 2.8470889315419066,
-      "grad_norm": 4.428896427154541,
+      "grad_norm": 5.5702433586120605,
       "learning_rate": 1.019407123053956e-06,
-      "loss": 0.18,
+      "loss": 0.7068,
       "step": 4450
     },
     {
       "epoch": 2.8790786948176583,
-      "grad_norm": 10.087977409362793,
+      "grad_norm": 4.175040245056152,
       "learning_rate": 8.061420345489445e-07,
-      "loss": 0.2102,
+      "loss": 0.7092,
       "step": 4500
     },
     {
       "epoch": 2.91106845809341,
-      "grad_norm": 16.517681121826172,
+      "grad_norm": 4.912069797515869,
       "learning_rate": 5.928769460439326e-07,
-      "loss": 0.1662,
+      "loss": 0.7022,
       "step": 4550
     },
     {
       "epoch": 2.943058221369162,
-      "grad_norm": 15.329133987426758,
+      "grad_norm": 4.595305442810059,
       "learning_rate": 3.796118575389209e-07,
-      "loss": 0.1997,
+      "loss": 0.7228,
       "step": 4600
     },
     {
       "epoch": 2.9750479846449136,
-      "grad_norm": 18.751319885253906,
+      "grad_norm": 3.835669755935669,
       "learning_rate": 1.6634676903390917e-07,
-      "loss": 0.1924,
+      "loss": 0.6946,
       "step": 4650
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.9008317338451696,
-      "eval_f1": 0.892446816336566,
-      "eval_loss": 0.3720133900642395,
-      "eval_runtime": 71.8708,
-      "eval_samples_per_second": 43.495,
-      "eval_steps_per_second": 2.727,
+      "eval_accuracy": 0.6829814459373,
+      "eval_f1": 0.5863851194804445,
+      "eval_loss": 0.7020198106765747,
+      "eval_runtime": 144.2948,
+      "eval_samples_per_second": 21.664,
+      "eval_steps_per_second": 1.358,
       "step": 4689
     }
   ],
@@ -707,7 +707,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 837402132857850.0,
+  "total_flos": 1757912647073562.0,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:80719dde8ee12368c548b9d0fd820c908d79ccc5612eb348afb50e76be1caccc
-size 5240
+oid sha256:7b1a2694e34f5fe6ee1f92624820bc26f69ac4d9b0f9439ad2d5b4381e756508
+size 5176