End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/bert_base_km_20_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: bert_base_km_20_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: 0.6353651839648545
 ---
 
 <!-- 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_20_v1_qnli
 
-This model is a fine-tuned version of [Hartunka/bert_base_km_20_v1](https://huggingface.co/Hartunka/bert_base_km_20_v1) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/bert_base_km_20_v1](https://huggingface.co/Hartunka/bert_base_km_20_v1) on the GLUE QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.1769
-- Accuracy: 0.6458
+- Loss: 0.6347
+- Accuracy: 0.6354
 
 ## Model description
 

all_results.json

@@ -1,15 +1,15 @@
 {
     "epoch": 7.0,
-    "eval_accuracy": 0.629873695771554,
-    "eval_loss": 0.6378833651542664,
-    "eval_runtime": 3.4372,
+    "eval_accuracy": 0.6353651839648545,
+    "eval_loss": 0.6347135305404663,
+    "eval_runtime": 3.4354,
     "eval_samples": 5463,
-    "eval_samples_per_second": 1589.388,
-    "eval_steps_per_second": 6.401,
+    "eval_samples_per_second": 1590.194,
+    "eval_steps_per_second": 6.404,
     "total_flos": 9.645664445050368e+16,
-    "train_loss": 0.43505029711573795,
-    "train_runtime": 1217.4791,
+    "train_loss": 0.4254190969965599,
+    "train_runtime": 1216.1304,
     "train_samples": 104743,
-    "train_samples_per_second": 4301.634,
-    "train_steps_per_second": 16.838
+    "train_samples_per_second": 4306.405,
+    "train_steps_per_second": 16.857
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 7.0,
-    "eval_accuracy": 0.629873695771554,
-    "eval_loss": 0.6378833651542664,
-    "eval_runtime": 3.4372,
+    "eval_accuracy": 0.6353651839648545,
+    "eval_loss": 0.6347135305404663,
+    "eval_runtime": 3.4354,
     "eval_samples": 5463,
-    "eval_samples_per_second": 1589.388,
-    "eval_steps_per_second": 6.401
+    "eval_samples_per_second": 1590.194,
+    "eval_steps_per_second": 6.404
 }

logs/events.out.tfevents.1745045504.s_005_m.2788007.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85812bf91b4bccc5fbdd76c7e84c6062a6c2fdb7d05dceb0bc344774ed8ee752
+size 363

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 7.0,
     "total_flos": 9.645664445050368e+16,
-    "train_loss": 0.43505029711573795,
-    "train_runtime": 1217.4791,
+    "train_loss": 0.4254190969965599,
+    "train_runtime": 1216.1304,
     "train_samples": 104743,
-    "train_samples_per_second": 4301.634,
-    "train_steps_per_second": 16.838
+    "train_samples_per_second": 4306.405,
+    "train_steps_per_second": 16.857
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 820,
-  "best_metric": 0.6378833651542664,
+  "best_metric": 0.6347135305404663,
   "best_model_checkpoint": "bert_base_km_20_v1_qnli/checkpoint-820",
   "epoch": 7.0,
   "eval_steps": 500,
@@ -11,124 +11,124 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 1.7936151027679443,
+      "grad_norm": 1.621480107307434,
       "learning_rate": 4.9e-05,
-      "loss": 0.6684,
+      "loss": 0.668,
       "step": 410
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6203551162365001,
-      "eval_loss": 0.6478282809257507,
-      "eval_runtime": 3.6324,
-      "eval_samples_per_second": 1503.965,
-      "eval_steps_per_second": 6.057,
+      "eval_accuracy": 0.6232839099395936,
+      "eval_loss": 0.6446160674095154,
+      "eval_runtime": 3.5028,
+      "eval_samples_per_second": 1559.629,
+      "eval_steps_per_second": 6.281,
       "step": 410
     },
     {
       "epoch": 2.0,
-      "grad_norm": 1.8211458921432495,
+      "grad_norm": 1.6884652376174927,
       "learning_rate": 4.8e-05,
-      "loss": 0.632,
+      "loss": 0.6309,
       "step": 820
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.629873695771554,
-      "eval_loss": 0.6378833651542664,
-      "eval_runtime": 3.4314,
-      "eval_samples_per_second": 1592.059,
-      "eval_steps_per_second": 6.411,
+      "eval_accuracy": 0.6353651839648545,
+      "eval_loss": 0.6347135305404663,
+      "eval_runtime": 3.4361,
+      "eval_samples_per_second": 1589.87,
+      "eval_steps_per_second": 6.403,
       "step": 820
     },
     {
       "epoch": 3.0,
-      "grad_norm": 2.655383348464966,
+      "grad_norm": 2.6409614086151123,
       "learning_rate": 4.7e-05,
-      "loss": 0.572,
+      "loss": 0.5661,
       "step": 1230
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.6313380926231008,
-      "eval_loss": 0.6602140069007874,
-      "eval_runtime": 3.4526,
-      "eval_samples_per_second": 1582.266,
-      "eval_steps_per_second": 6.372,
+      "eval_accuracy": 0.6317041918359876,
+      "eval_loss": 0.6613866686820984,
+      "eval_runtime": 3.4605,
+      "eval_samples_per_second": 1578.666,
+      "eval_steps_per_second": 6.357,
       "step": 1230
     },
     {
       "epoch": 4.0,
-      "grad_norm": 4.3831000328063965,
+      "grad_norm": 5.787648677825928,
       "learning_rate": 4.600000000000001e-05,
-      "loss": 0.4607,
+      "loss": 0.4463,
       "step": 1640
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.6492769540545488,
-      "eval_loss": 0.6907069683074951,
-      "eval_runtime": 3.4461,
-      "eval_samples_per_second": 1585.267,
-      "eval_steps_per_second": 6.384,
+      "eval_accuracy": 0.6414058209774849,
+      "eval_loss": 0.6944608688354492,
+      "eval_runtime": 3.4319,
+      "eval_samples_per_second": 1591.851,
+      "eval_steps_per_second": 6.411,
       "step": 1640
     },
     {
       "epoch": 5.0,
-      "grad_norm": 4.579970359802246,
+      "grad_norm": 5.31504487991333,
       "learning_rate": 4.5e-05,
-      "loss": 0.3293,
+      "loss": 0.3097,
       "step": 2050
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.6485447556287754,
-      "eval_loss": 0.8158578276634216,
-      "eval_runtime": 3.4247,
-      "eval_samples_per_second": 1595.176,
-      "eval_steps_per_second": 6.424,
+      "eval_accuracy": 0.6447007138934652,
+      "eval_loss": 0.841460108757019,
+      "eval_runtime": 3.4478,
+      "eval_samples_per_second": 1584.501,
+      "eval_steps_per_second": 6.381,
       "step": 2050
     },
     {
       "epoch": 6.0,
-      "grad_norm": 9.527710914611816,
+      "grad_norm": 8.552193641662598,
       "learning_rate": 4.4000000000000006e-05,
-      "loss": 0.223,
+      "loss": 0.2074,
       "step": 2460
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.6404905729452681,
-      "eval_loss": 1.0376250743865967,
-      "eval_runtime": 3.4217,
-      "eval_samples_per_second": 1596.598,
-      "eval_steps_per_second": 6.43,
+      "eval_accuracy": 0.6355482335712979,
+      "eval_loss": 1.076311707496643,
+      "eval_runtime": 3.4374,
+      "eval_samples_per_second": 1589.288,
+      "eval_steps_per_second": 6.4,
       "step": 2460
     },
     {
       "epoch": 7.0,
-      "grad_norm": 9.083931922912598,
+      "grad_norm": 8.511574745178223,
       "learning_rate": 4.3e-05,
-      "loss": 0.1601,
+      "loss": 0.1497,
       "step": 2870
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.642504118616145,
-      "eval_loss": 1.2133229970932007,
-      "eval_runtime": 3.4431,
-      "eval_samples_per_second": 1586.657,
-      "eval_steps_per_second": 6.39,
+      "eval_accuracy": 0.6457990115321252,
+      "eval_loss": 1.1768972873687744,
+      "eval_runtime": 3.4374,
+      "eval_samples_per_second": 1589.297,
+      "eval_steps_per_second": 6.4,
       "step": 2870
     },
     {
       "epoch": 7.0,
       "step": 2870,
       "total_flos": 9.645664445050368e+16,
-      "train_loss": 0.43505029711573795,
-      "train_runtime": 1217.4791,
-      "train_samples_per_second": 4301.634,
-      "train_steps_per_second": 16.838
+      "train_loss": 0.4254190969965599,
+      "train_runtime": 1216.1304,
+      "train_samples_per_second": 4306.405,
+      "train_steps_per_second": 16.857
     }
   ],
   "logging_steps": 1,