End of training

README.md

@@ -1,14 +1,32 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/bert_base_km_20_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: bert_base_km_20_v1_qqp
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE QQP
+      type: glue
+      args: qqp
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.8163739797180312
+    - name: F1
+      type: f1
+      value: 0.7572110667800379
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -16,12 +34,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # bert_base_km_20_v1_qqp
 
-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 QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5794
-- Accuracy: 0.8356
-- F1: 0.7807
-- Combined Score: 0.8081
+- Loss: 0.3921
+- Accuracy: 0.8164
+- F1: 0.7572
+- Combined Score: 0.7868
 
 ## Model description
 

all_results.json

@@ -4,14 +4,14 @@
     "eval_combined_score": 0.7867925232490345,
     "eval_f1": 0.7572110667800379,
     "eval_loss": 0.3920540511608124,
-    "eval_runtime": 25.029,
+    "eval_runtime": 25.2211,
     "eval_samples": 40430,
-    "eval_samples_per_second": 1615.328,
-    "eval_steps_per_second": 6.313,
+    "eval_samples_per_second": 1603.021,
+    "eval_steps_per_second": 6.265,
     "total_flos": 3.3506166766980096e+17,
-    "train_loss": 0.25013284528018337,
-    "train_runtime": 4253.2472,
+    "train_loss": 0.2501540595037956,
+    "train_runtime": 4264.0449,
     "train_samples": 363846,
-    "train_samples_per_second": 4277.273,
-    "train_steps_per_second": 16.717
+    "train_samples_per_second": 4266.442,
+    "train_steps_per_second": 16.674
 }

eval_results.json

@@ -4,8 +4,8 @@
     "eval_combined_score": 0.7867925232490345,
     "eval_f1": 0.7572110667800379,
     "eval_loss": 0.3920540511608124,
-    "eval_runtime": 25.029,
+    "eval_runtime": 25.2211,
     "eval_samples": 40430,
-    "eval_samples_per_second": 1615.328,
-    "eval_steps_per_second": 6.313
+    "eval_samples_per_second": 1603.021,
+    "eval_steps_per_second": 6.265
 }

logs/events.out.tfevents.1745049845.s_005_m.2788007.7

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76da787a4d796c86043ca41a1a5fae72ad4e72d934a1a7ca74fae76c2e1e6f4c
+size 467

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 7.0,
     "total_flos": 3.3506166766980096e+17,
-    "train_loss": 0.25013284528018337,
-    "train_runtime": 4253.2472,
+    "train_loss": 0.2501540595037956,
+    "train_runtime": 4264.0449,
     "train_samples": 363846,
-    "train_samples_per_second": 4277.273,
-    "train_steps_per_second": 16.717
+    "train_samples_per_second": 4266.442,
+    "train_steps_per_second": 16.674
 }

trainer_state.json

@@ -22,9 +22,9 @@
       "eval_combined_score": 0.7114269828123259,
       "eval_f1": 0.6421218360179242,
       "eval_loss": 0.4579184949398041,
-      "eval_runtime": 25.8671,
-      "eval_samples_per_second": 1562.987,
-      "eval_steps_per_second": 6.108,
+      "eval_runtime": 26.0123,
+      "eval_samples_per_second": 1554.263,
+      "eval_steps_per_second": 6.074,
       "step": 1422
     },
     {
@@ -40,9 +40,9 @@
       "eval_combined_score": 0.7867925232490345,
       "eval_f1": 0.7572110667800379,
       "eval_loss": 0.3920540511608124,
-      "eval_runtime": 25.2449,
-      "eval_samples_per_second": 1601.511,
-      "eval_steps_per_second": 6.259,
+      "eval_runtime": 25.2701,
+      "eval_samples_per_second": 1599.916,
+      "eval_steps_per_second": 6.252,
       "step": 2844
     },
     {
@@ -58,91 +58,91 @@
       "eval_combined_score": 0.7945958553596527,
       "eval_f1": 0.7648286140089419,
       "eval_loss": 0.3958491384983063,
-      "eval_runtime": 25.2475,
-      "eval_samples_per_second": 1601.344,
-      "eval_steps_per_second": 6.258,
+      "eval_runtime": 25.3413,
+      "eval_samples_per_second": 1595.42,
+      "eval_steps_per_second": 6.235,
       "step": 4266
     },
     {
       "epoch": 4.0,
-      "grad_norm": 2.9538912773132324,
+      "grad_norm": 3.1108007431030273,
       "learning_rate": 4.600000000000001e-05,
-      "loss": 0.2175,
+      "loss": 0.2176,
       "step": 5688
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8311897106109325,
-      "eval_combined_score": 0.7985560653064951,
-      "eval_f1": 0.7659224200020578,
-      "eval_loss": 0.4462401866912842,
-      "eval_runtime": 25.2658,
-      "eval_samples_per_second": 1600.186,
-      "eval_steps_per_second": 6.254,
+      "eval_accuracy": 0.8303240168191937,
+      "eval_combined_score": 0.7973735468711353,
+      "eval_f1": 0.7644230769230769,
+      "eval_loss": 0.44668564200401306,
+      "eval_runtime": 25.356,
+      "eval_samples_per_second": 1594.496,
+      "eval_steps_per_second": 6.231,
       "step": 5688
     },
     {
       "epoch": 5.0,
-      "grad_norm": 2.7615535259246826,
+      "grad_norm": 2.534144639968872,
       "learning_rate": 4.5e-05,
       "loss": 0.1641,
       "step": 7110
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.8319317338609943,
-      "eval_combined_score": 0.7997921620236754,
-      "eval_f1": 0.7676525901863567,
-      "eval_loss": 0.47893673181533813,
-      "eval_runtime": 25.3639,
-      "eval_samples_per_second": 1593.995,
-      "eval_steps_per_second": 6.229,
+      "eval_accuracy": 0.8321543408360128,
+      "eval_combined_score": 0.798826496563559,
+      "eval_f1": 0.7654986522911051,
+      "eval_loss": 0.4786772131919861,
+      "eval_runtime": 25.3806,
+      "eval_samples_per_second": 1592.949,
+      "eval_steps_per_second": 6.225,
       "step": 7110
     },
     {
       "epoch": 6.0,
-      "grad_norm": 3.9577207565307617,
+      "grad_norm": 4.56919002532959,
       "learning_rate": 4.4000000000000006e-05,
-      "loss": 0.1258,
+      "loss": 0.1263,
       "step": 8532
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.835171902052931,
-      "eval_combined_score": 0.8034685457247326,
-      "eval_f1": 0.771765189396534,
-      "eval_loss": 0.5442411303520203,
-      "eval_runtime": 25.1356,
-      "eval_samples_per_second": 1608.473,
-      "eval_steps_per_second": 6.286,
+      "eval_accuracy": 0.8334899826861242,
+      "eval_combined_score": 0.8009475508586806,
+      "eval_f1": 0.7684051190312371,
+      "eval_loss": 0.5397569537162781,
+      "eval_runtime": 25.3607,
+      "eval_samples_per_second": 1594.201,
+      "eval_steps_per_second": 6.23,
       "step": 8532
     },
     {
       "epoch": 7.0,
-      "grad_norm": 5.18740701675415,
+      "grad_norm": 6.266264915466309,
       "learning_rate": 4.3e-05,
-      "loss": 0.0998,
+      "loss": 0.0993,
       "step": 9954
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.8336878555528073,
-      "eval_combined_score": 0.8060038908517037,
-      "eval_f1": 0.7783199261506,
-      "eval_loss": 0.5709294676780701,
-      "eval_runtime": 25.1251,
-      "eval_samples_per_second": 1609.149,
-      "eval_steps_per_second": 6.289,
+      "eval_accuracy": 0.8356418501113035,
+      "eval_combined_score": 0.8081493620914663,
+      "eval_f1": 0.780656874071629,
+      "eval_loss": 0.5794485807418823,
+      "eval_runtime": 25.3072,
+      "eval_samples_per_second": 1597.566,
+      "eval_steps_per_second": 6.243,
       "step": 9954
     },
     {
       "epoch": 7.0,
       "step": 9954,
       "total_flos": 3.3506166766980096e+17,
-      "train_loss": 0.25013284528018337,
-      "train_runtime": 4253.2472,
-      "train_samples_per_second": 4277.273,
-      "train_steps_per_second": 16.717
+      "train_loss": 0.2501540595037956,
+      "train_runtime": 4264.0449,
+      "train_samples_per_second": 4266.442,
+      "train_steps_per_second": 16.674
     }
   ],
   "logging_steps": 1,