End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/bert_base_km_10_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: bert_base_km_10_v1_qnli
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE QNLI
+      type: glue
+      args: qnli
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.6498261028738789
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -15,10 +30,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # bert_base_km_10_v1_qnli
 
-This model is a fine-tuned version of [Hartunka/bert_base_km_10_v1](https://huggingface.co/Hartunka/bert_base_km_10_v1) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/bert_base_km_10_v1](https://huggingface.co/Hartunka/bert_base_km_10_v1) on the GLUE QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.2518
-- Accuracy: 0.6469
+- Loss: 0.6279
+- Accuracy: 0.6498
 
 ## Model description
 

all_results.json

@@ -1,15 +1,15 @@
 {
     "epoch": 7.0,
-    "eval_accuracy": 0.6490939044481054,
-    "eval_loss": 0.6261129379272461,
-    "eval_runtime": 3.4175,
+    "eval_accuracy": 0.6498261028738789,
+    "eval_loss": 0.6279211044311523,
+    "eval_runtime": 3.422,
     "eval_samples": 5463,
-    "eval_samples_per_second": 1598.526,
-    "eval_steps_per_second": 6.437,
+    "eval_samples_per_second": 1596.438,
+    "eval_steps_per_second": 6.429,
     "total_flos": 9.645664445050368e+16,
-    "train_loss": 0.4108444413241609,
-    "train_runtime": 1229.6341,
+    "train_loss": 0.4127295610381336,
+    "train_runtime": 1216.6029,
     "train_samples": 104743,
-    "train_samples_per_second": 4259.113,
-    "train_steps_per_second": 16.672
+    "train_samples_per_second": 4304.732,
+    "train_steps_per_second": 16.85
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 7.0,
-    "eval_accuracy": 0.6490939044481054,
-    "eval_loss": 0.6261129379272461,
-    "eval_runtime": 3.4175,
+    "eval_accuracy": 0.6498261028738789,
+    "eval_loss": 0.6279211044311523,
+    "eval_runtime": 3.422,
     "eval_samples": 5463,
-    "eval_samples_per_second": 1598.526,
-    "eval_steps_per_second": 6.437
+    "eval_samples_per_second": 1596.438,
+    "eval_steps_per_second": 6.429
 }

logs/events.out.tfevents.1745017143.s_005_m.2780655.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:460ab7f4974e8ea38b5c4514a119eef176a984a3ed4e42223ba88a17b97a8147
+size 363

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 7.0,
     "total_flos": 9.645664445050368e+16,
-    "train_loss": 0.4108444413241609,
-    "train_runtime": 1229.6341,
+    "train_loss": 0.4127295610381336,
+    "train_runtime": 1216.6029,
     "train_samples": 104743,
-    "train_samples_per_second": 4259.113,
-    "train_steps_per_second": 16.672
+    "train_samples_per_second": 4304.732,
+    "train_steps_per_second": 16.85
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 820,
-  "best_metric": 0.6261129379272461,
+  "best_metric": 0.6279211044311523,
   "best_model_checkpoint": "bert_base_km_10_v1_qnli/checkpoint-820",
   "epoch": 7.0,
   "eval_steps": 500,
@@ -11,124 +11,124 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 2.410963773727417,
+      "grad_norm": 2.2646937370300293,
       "learning_rate": 4.9e-05,
-      "loss": 0.6641,
+      "loss": 0.6642,
       "step": 410
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6377448288486179,
-      "eval_loss": 0.6366901397705078,
-      "eval_runtime": 3.557,
-      "eval_samples_per_second": 1535.854,
-      "eval_steps_per_second": 6.185,
+      "eval_accuracy": 0.6395753249130515,
+      "eval_loss": 0.6345698833465576,
+      "eval_runtime": 3.4904,
+      "eval_samples_per_second": 1565.133,
+      "eval_steps_per_second": 6.303,
       "step": 410
     },
     {
       "epoch": 2.0,
-      "grad_norm": 2.487971067428589,
+      "grad_norm": 2.06400728225708,
       "learning_rate": 4.8e-05,
-      "loss": 0.6198,
+      "loss": 0.6203,
       "step": 820
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.6490939044481054,
-      "eval_loss": 0.6261129379272461,
-      "eval_runtime": 3.453,
-      "eval_samples_per_second": 1582.081,
-      "eval_steps_per_second": 6.371,
+      "eval_accuracy": 0.6498261028738789,
+      "eval_loss": 0.6279211044311523,
+      "eval_runtime": 3.4233,
+      "eval_samples_per_second": 1595.845,
+      "eval_steps_per_second": 6.427,
       "step": 820
     },
     {
       "epoch": 3.0,
-      "grad_norm": 3.140673875808716,
+      "grad_norm": 3.729912281036377,
       "learning_rate": 4.7e-05,
-      "loss": 0.5408,
+      "loss": 0.5422,
       "step": 1230
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.6265788028555739,
-      "eval_loss": 0.6862198114395142,
-      "eval_runtime": 3.4467,
-      "eval_samples_per_second": 1585.001,
-      "eval_steps_per_second": 6.383,
+      "eval_accuracy": 0.6304228445908842,
+      "eval_loss": 0.6750292181968689,
+      "eval_runtime": 3.4493,
+      "eval_samples_per_second": 1583.801,
+      "eval_steps_per_second": 6.378,
       "step": 1230
     },
     {
       "epoch": 4.0,
-      "grad_norm": 7.148397922515869,
+      "grad_norm": 6.5136637687683105,
       "learning_rate": 4.600000000000001e-05,
-      "loss": 0.4205,
+      "loss": 0.422,
       "step": 1640
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.6441515650741351,
-      "eval_loss": 0.7419140338897705,
-      "eval_runtime": 3.4555,
-      "eval_samples_per_second": 1580.954,
-      "eval_steps_per_second": 6.367,
+      "eval_accuracy": 0.6467142595643419,
+      "eval_loss": 0.7332958579063416,
+      "eval_runtime": 3.4349,
+      "eval_samples_per_second": 1590.425,
+      "eval_steps_per_second": 6.405,
       "step": 1640
     },
     {
       "epoch": 5.0,
-      "grad_norm": 6.679465293884277,
+      "grad_norm": 6.036999702453613,
       "learning_rate": 4.5e-05,
-      "loss": 0.2925,
+      "loss": 0.2952,
       "step": 2050
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.6448837634999085,
-      "eval_loss": 0.8849136233329773,
-      "eval_runtime": 3.4598,
-      "eval_samples_per_second": 1578.979,
-      "eval_steps_per_second": 6.359,
+      "eval_accuracy": 0.6551345414607359,
+      "eval_loss": 0.8971741199493408,
+      "eval_runtime": 3.4347,
+      "eval_samples_per_second": 1590.509,
+      "eval_steps_per_second": 6.405,
       "step": 2050
     },
     {
       "epoch": 6.0,
-      "grad_norm": 8.848244667053223,
+      "grad_norm": 8.021710395812988,
       "learning_rate": 4.4000000000000006e-05,
-      "loss": 0.1969,
+      "loss": 0.1998,
       "step": 2460
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.638843126487278,
-      "eval_loss": 1.09120774269104,
-      "eval_runtime": 3.4458,
-      "eval_samples_per_second": 1585.388,
-      "eval_steps_per_second": 6.385,
+      "eval_accuracy": 0.6445176642870217,
+      "eval_loss": 1.0509867668151855,
+      "eval_runtime": 3.4393,
+      "eval_samples_per_second": 1588.41,
+      "eval_steps_per_second": 6.397,
       "step": 2460
     },
     {
       "epoch": 7.0,
-      "grad_norm": 9.282306671142578,
+      "grad_norm": 11.778338432312012,
       "learning_rate": 4.3e-05,
-      "loss": 0.1413,
+      "loss": 0.1454,
       "step": 2870
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.6390261760937214,
-      "eval_loss": 1.2178927659988403,
-      "eval_runtime": 3.4453,
-      "eval_samples_per_second": 1585.62,
-      "eval_steps_per_second": 6.385,
+      "eval_accuracy": 0.6468973091707853,
+      "eval_loss": 1.2517988681793213,
+      "eval_runtime": 3.434,
+      "eval_samples_per_second": 1590.876,
+      "eval_steps_per_second": 6.407,
       "step": 2870
     },
     {
       "epoch": 7.0,
       "step": 2870,
       "total_flos": 9.645664445050368e+16,
-      "train_loss": 0.4108444413241609,
-      "train_runtime": 1229.6341,
-      "train_samples_per_second": 4259.113,
-      "train_steps_per_second": 16.672
+      "train_loss": 0.4127295610381336,
+      "train_runtime": 1216.6029,
+      "train_samples_per_second": 4304.732,
+      "train_steps_per_second": 16.85
     }
   ],
   "logging_steps": 1,