End of training

README.md

@@ -1,14 +1,32 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/bert_base_km_10_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: bert_base_km_10_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: 0.8171407370764284
+    - name: F1
+      type: f1
+      value: 0.7492793434394818
 ---
 
 <!-- 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_10_v1_qqp
 
-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 QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6114
-- Accuracy: 0.8388
-- F1: 0.7771
-- Combined Score: 0.8079
+- Loss: 0.3983
+- Accuracy: 0.8171
+- F1: 0.7493
+- Combined Score: 0.7832
 
 ## Model description
 

all_results.json

@@ -1,17 +1,17 @@
 {
     "epoch": 7.0,
-    "eval_accuracy": 0.8168933959930744,
-    "eval_combined_score": 0.7826009835165454,
-    "eval_f1": 0.7483085710400164,
-    "eval_loss": 0.39908739924430847,
-    "eval_runtime": 25.4192,
+    "eval_accuracy": 0.8171407370764284,
+    "eval_combined_score": 0.7832100402579552,
+    "eval_f1": 0.7492793434394818,
+    "eval_loss": 0.3983299732208252,
+    "eval_runtime": 25.1253,
     "eval_samples": 40430,
-    "eval_samples_per_second": 1590.533,
-    "eval_steps_per_second": 6.216,
+    "eval_samples_per_second": 1609.133,
+    "eval_steps_per_second": 6.288,
     "total_flos": 3.3506166766980096e+17,
-    "train_loss": 0.24196342696668346,
-    "train_runtime": 4302.4345,
+    "train_loss": 0.2418764561612323,
+    "train_runtime": 4259.546,
     "train_samples": 363846,
-    "train_samples_per_second": 4228.373,
-    "train_steps_per_second": 16.526
+    "train_samples_per_second": 4270.948,
+    "train_steps_per_second": 16.692
 }

eval_results.json

@@ -1,11 +1,11 @@
 {
     "epoch": 7.0,
-    "eval_accuracy": 0.8168933959930744,
-    "eval_combined_score": 0.7826009835165454,
-    "eval_f1": 0.7483085710400164,
-    "eval_loss": 0.39908739924430847,
-    "eval_runtime": 25.4192,
+    "eval_accuracy": 0.8171407370764284,
+    "eval_combined_score": 0.7832100402579552,
+    "eval_f1": 0.7492793434394818,
+    "eval_loss": 0.3983299732208252,
+    "eval_runtime": 25.1253,
     "eval_samples": 40430,
-    "eval_samples_per_second": 1590.533,
-    "eval_steps_per_second": 6.216
+    "eval_samples_per_second": 1609.133,
+    "eval_steps_per_second": 6.288
 }

logs/events.out.tfevents.1745021469.s_005_m.2780655.7

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:999366d04672853c556fbe517e1b19570a6d11347938f1e252622972eec78a15
+size 467

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 7.0,
     "total_flos": 3.3506166766980096e+17,
-    "train_loss": 0.24196342696668346,
-    "train_runtime": 4302.4345,
+    "train_loss": 0.2418764561612323,
+    "train_runtime": 4259.546,
     "train_samples": 363846,
-    "train_samples_per_second": 4228.373,
-    "train_steps_per_second": 16.526
+    "train_samples_per_second": 4270.948,
+    "train_steps_per_second": 16.692
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 2844,
-  "best_metric": 0.39908739924430847,
+  "best_metric": 0.3983299732208252,
   "best_model_checkpoint": "bert_base_km_10_v1_qqp/checkpoint-2844",
   "epoch": 7.0,
   "eval_steps": 500,
@@ -11,138 +11,138 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 1.8532532453536987,
+      "grad_norm": 1.8717503547668457,
       "learning_rate": 4.9e-05,
-      "loss": 0.4755,
+      "loss": 0.4756,
       "step": 1422
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7899826861241652,
-      "eval_combined_score": 0.7309140483760923,
-      "eval_f1": 0.6718454106280193,
-      "eval_loss": 0.44356149435043335,
-      "eval_runtime": 25.799,
-      "eval_samples_per_second": 1567.112,
-      "eval_steps_per_second": 6.124,
+      "eval_accuracy": 0.7900321543408361,
+      "eval_combined_score": 0.7309520672751686,
+      "eval_f1": 0.671871980209501,
+      "eval_loss": 0.4438171982765198,
+      "eval_runtime": 25.8505,
+      "eval_samples_per_second": 1563.992,
+      "eval_steps_per_second": 6.112,
       "step": 1422
     },
     {
       "epoch": 2.0,
-      "grad_norm": 2.911086082458496,
+      "grad_norm": 2.944045305252075,
       "learning_rate": 4.8e-05,
-      "loss": 0.3635,
+      "loss": 0.3636,
       "step": 2844
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8168933959930744,
-      "eval_combined_score": 0.7826009835165454,
-      "eval_f1": 0.7483085710400164,
-      "eval_loss": 0.39908739924430847,
-      "eval_runtime": 25.503,
-      "eval_samples_per_second": 1585.302,
-      "eval_steps_per_second": 6.195,
+      "eval_accuracy": 0.8171407370764284,
+      "eval_combined_score": 0.7832100402579552,
+      "eval_f1": 0.7492793434394818,
+      "eval_loss": 0.3983299732208252,
+      "eval_runtime": 25.2779,
+      "eval_samples_per_second": 1599.421,
+      "eval_steps_per_second": 6.251,
       "step": 2844
     },
     {
       "epoch": 3.0,
-      "grad_norm": 3.2880516052246094,
+      "grad_norm": 3.461521863937378,
       "learning_rate": 4.7e-05,
-      "loss": 0.2764,
+      "loss": 0.2766,
       "step": 4266
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8274301261439525,
-      "eval_combined_score": 0.7966945001925022,
-      "eval_f1": 0.7659588742410519,
-      "eval_loss": 0.4213642179965973,
-      "eval_runtime": 25.414,
-      "eval_samples_per_second": 1590.855,
-      "eval_steps_per_second": 6.217,
+      "eval_accuracy": 0.8278258718773188,
+      "eval_combined_score": 0.7976598880746693,
+      "eval_f1": 0.7674939042720198,
+      "eval_loss": 0.419876366853714,
+      "eval_runtime": 25.2754,
+      "eval_samples_per_second": 1599.582,
+      "eval_steps_per_second": 6.251,
       "step": 4266
     },
     {
       "epoch": 4.0,
-      "grad_norm": 2.7092745304107666,
+      "grad_norm": 2.752514600753784,
       "learning_rate": 4.600000000000001e-05,
-      "loss": 0.2061,
+      "loss": 0.2059,
       "step": 5688
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.83447934701954,
-      "eval_combined_score": 0.7993729692685413,
-      "eval_f1": 0.7642665915175426,
-      "eval_loss": 0.47107040882110596,
-      "eval_runtime": 25.2725,
-      "eval_samples_per_second": 1599.766,
-      "eval_steps_per_second": 6.252,
+      "eval_accuracy": 0.8327232253277269,
+      "eval_combined_score": 0.7966158709591289,
+      "eval_f1": 0.7605085165905309,
+      "eval_loss": 0.46601402759552,
+      "eval_runtime": 25.4001,
+      "eval_samples_per_second": 1591.723,
+      "eval_steps_per_second": 6.22,
       "step": 5688
     },
     {
       "epoch": 5.0,
-      "grad_norm": 3.767702341079712,
+      "grad_norm": 3.1064395904541016,
       "learning_rate": 4.5e-05,
-      "loss": 0.1551,
+      "loss": 0.1549,
       "step": 7110
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.8337620578778135,
-      "eval_combined_score": 0.7942305672432244,
-      "eval_f1": 0.7546990766086353,
-      "eval_loss": 0.5268918871879578,
-      "eval_runtime": 25.4108,
-      "eval_samples_per_second": 1591.056,
-      "eval_steps_per_second": 6.218,
+      "eval_accuracy": 0.8351471679445956,
+      "eval_combined_score": 0.8002196381329486,
+      "eval_f1": 0.7652921083213016,
+      "eval_loss": 0.5142884254455566,
+      "eval_runtime": 25.3616,
+      "eval_samples_per_second": 1594.14,
+      "eval_steps_per_second": 6.23,
       "step": 7110
     },
     {
       "epoch": 6.0,
-      "grad_norm": 5.443331718444824,
+      "grad_norm": 5.583347320556641,
       "learning_rate": 4.4000000000000006e-05,
       "loss": 0.1209,
       "step": 8532
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.8357160524363096,
-      "eval_combined_score": 0.8038283599635194,
-      "eval_f1": 0.7719406674907293,
-      "eval_loss": 0.5602436065673828,
-      "eval_runtime": 25.2783,
-      "eval_samples_per_second": 1599.395,
-      "eval_steps_per_second": 6.25,
+      "eval_accuracy": 0.8362849369280237,
+      "eval_combined_score": 0.8034324064211018,
+      "eval_f1": 0.7705798759141798,
+      "eval_loss": 0.5888400673866272,
+      "eval_runtime": 25.2856,
+      "eval_samples_per_second": 1598.932,
+      "eval_steps_per_second": 6.249,
       "step": 8532
     },
     {
       "epoch": 7.0,
-      "grad_norm": 3.5738558769226074,
+      "grad_norm": 3.643422842025757,
       "learning_rate": 4.3e-05,
-      "loss": 0.0961,
+      "loss": 0.0957,
       "step": 9954
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.8356665842196389,
-      "eval_combined_score": 0.8052383897237023,
-      "eval_f1": 0.7748101952277657,
-      "eval_loss": 0.6076688170433044,
-      "eval_runtime": 25.4562,
-      "eval_samples_per_second": 1588.218,
-      "eval_steps_per_second": 6.207,
+      "eval_accuracy": 0.8387830818698986,
+      "eval_combined_score": 0.8079193833210486,
+      "eval_f1": 0.7770556847721987,
+      "eval_loss": 0.6113874316215515,
+      "eval_runtime": 25.3801,
+      "eval_samples_per_second": 1592.979,
+      "eval_steps_per_second": 6.225,
       "step": 9954
     },
     {
       "epoch": 7.0,
       "step": 9954,
       "total_flos": 3.3506166766980096e+17,
-      "train_loss": 0.24196342696668346,
-      "train_runtime": 4302.4345,
-      "train_samples_per_second": 4228.373,
-      "train_steps_per_second": 16.526
+      "train_loss": 0.2418764561612323,
+      "train_runtime": 4259.546,
+      "train_samples_per_second": 4270.948,
+      "train_steps_per_second": 16.692
     }
   ],
   "logging_steps": 1,