End of training

README.md

@@ -1,14 +1,32 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/bert_base_km_5_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - matthews_correlation
 - accuracy
 model-index:
 - name: bert_base_km_5_v1_cola
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE COLA
+      type: glue
+      args: cola
+    metrics:
+    - name: Matthews Correlation
+      type: matthews_correlation
+      value: 0.04959298805408078
+    - name: Accuracy
+      type: accuracy
+      value: 0.6874400973320007
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -16,11 +34,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 # bert_base_km_5_v1_cola
 
-This model is a fine-tuned version of [Hartunka/bert_base_km_5_v1](https://huggingface.co/Hartunka/bert_base_km_5_v1) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/bert_base_km_5_v1](https://huggingface.co/Hartunka/bert_base_km_5_v1) on the GLUE COLA dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.8536
-- Matthews Correlation: 0.1002
-- Accuracy: 0.6462
+- Loss: 0.6169
+- Matthews Correlation: 0.0496
+- Accuracy: 0.6874
 
 ## Model description
 

all_results.json

@@ -1,16 +1,16 @@
 {
     "epoch": 8.0,
-    "eval_accuracy": 0.693192720413208,
-    "eval_loss": 0.6094943881034851,
-    "eval_matthews_correlation": 0.060190597377967524,
-    "eval_runtime": 0.6896,
+    "eval_accuracy": 0.6874400973320007,
+    "eval_loss": 0.6168506741523743,
+    "eval_matthews_correlation": 0.04959298805408078,
+    "eval_runtime": 0.7317,
     "eval_samples": 1043,
-    "eval_samples_per_second": 1512.498,
-    "eval_steps_per_second": 7.251,
+    "eval_samples_per_second": 1425.514,
+    "eval_steps_per_second": 6.834,
     "total_flos": 8999450537533440.0,
-    "train_loss": 0.4899906516075134,
-    "train_runtime": 134.3561,
+    "train_loss": 0.47968793616575356,
+    "train_runtime": 135.9284,
     "train_samples": 8551,
-    "train_samples_per_second": 3182.216,
-    "train_steps_per_second": 12.653
+    "train_samples_per_second": 3145.406,
+    "train_steps_per_second": 12.507
 }

eval_results.json

@@ -1,10 +1,10 @@
 {
     "epoch": 8.0,
-    "eval_accuracy": 0.693192720413208,
-    "eval_loss": 0.6094943881034851,
-    "eval_matthews_correlation": 0.060190597377967524,
-    "eval_runtime": 0.6896,
+    "eval_accuracy": 0.6874400973320007,
+    "eval_loss": 0.6168506741523743,
+    "eval_matthews_correlation": 0.04959298805408078,
+    "eval_runtime": 0.7317,
     "eval_samples": 1043,
-    "eval_samples_per_second": 1512.498,
-    "eval_steps_per_second": 7.251
+    "eval_samples_per_second": 1425.514,
+    "eval_steps_per_second": 6.834
 }

logs/events.out.tfevents.1744964173.s_005_m.2772371.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25cb565be9f59fb19e1c8ea6029211079cf1ecfad3a360cfae13a0b8d5a630a5
+size 427

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 8.0,
     "total_flos": 8999450537533440.0,
-    "train_loss": 0.4899906516075134,
-    "train_runtime": 134.3561,
+    "train_loss": 0.47968793616575356,
+    "train_runtime": 135.9284,
     "train_samples": 8551,
-    "train_samples_per_second": 3182.216,
-    "train_steps_per_second": 12.653
+    "train_samples_per_second": 3145.406,
+    "train_steps_per_second": 12.507
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 102,
-  "best_metric": 0.6094943881034851,
+  "best_metric": 0.6168506741523743,
   "best_model_checkpoint": "bert_base_km_5_v1_cola/checkpoint-102",
   "epoch": 8.0,
   "eval_steps": 500,
@@ -11,148 +11,148 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 0.9332984089851379,
+      "grad_norm": 2.0184812545776367,
       "learning_rate": 4.9e-05,
-      "loss": 0.6192,
+      "loss": 0.6154,
       "step": 34
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6912751793861389,
-      "eval_loss": 0.6177230477333069,
-      "eval_matthews_correlation": 0.0,
-      "eval_runtime": 0.676,
-      "eval_samples_per_second": 1542.856,
-      "eval_steps_per_second": 7.396,
+      "eval_accuracy": 0.6893576383590698,
+      "eval_loss": 0.6183784604072571,
+      "eval_matthews_correlation": -0.007887379670285008,
+      "eval_runtime": 0.6744,
+      "eval_samples_per_second": 1546.637,
+      "eval_steps_per_second": 7.414,
       "step": 34
     },
     {
       "epoch": 2.0,
-      "grad_norm": 1.2563631534576416,
+      "grad_norm": 2.073906183242798,
       "learning_rate": 4.8e-05,
-      "loss": 0.6088,
+      "loss": 0.5959,
       "step": 68
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.6922339200973511,
-      "eval_loss": 0.6191443800926208,
-      "eval_matthews_correlation": 0.0463559874942472,
-      "eval_runtime": 0.6989,
-      "eval_samples_per_second": 1492.409,
-      "eval_steps_per_second": 7.154,
+      "eval_accuracy": 0.6740172505378723,
+      "eval_loss": 0.6171430349349976,
+      "eval_matthews_correlation": 0.005476740094736749,
+      "eval_runtime": 0.7236,
+      "eval_samples_per_second": 1441.409,
+      "eval_steps_per_second": 6.91,
       "step": 68
     },
     {
       "epoch": 3.0,
-      "grad_norm": 0.9797320365905762,
+      "grad_norm": 1.9242302179336548,
       "learning_rate": 4.7e-05,
-      "loss": 0.5837,
+      "loss": 0.5583,
       "step": 102
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.693192720413208,
-      "eval_loss": 0.6094943881034851,
-      "eval_matthews_correlation": 0.060190597377967524,
-      "eval_runtime": 0.6754,
-      "eval_samples_per_second": 1544.379,
-      "eval_steps_per_second": 7.404,
+      "eval_accuracy": 0.6874400973320007,
+      "eval_loss": 0.6168506741523743,
+      "eval_matthews_correlation": 0.04959298805408078,
+      "eval_runtime": 0.6807,
+      "eval_samples_per_second": 1532.323,
+      "eval_steps_per_second": 7.346,
       "step": 102
     },
     {
       "epoch": 4.0,
-      "grad_norm": 2.736848831176758,
+      "grad_norm": 3.0941638946533203,
       "learning_rate": 4.600000000000001e-05,
-      "loss": 0.5398,
+      "loss": 0.5145,
       "step": 136
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.6826462149620056,
-      "eval_loss": 0.6479612588882446,
-      "eval_matthews_correlation": 0.08323045697852056,
-      "eval_runtime": 0.6667,
-      "eval_samples_per_second": 1564.401,
-      "eval_steps_per_second": 7.5,
+      "eval_accuracy": 0.6634707450866699,
+      "eval_loss": 0.6423744559288025,
+      "eval_matthews_correlation": 0.09532212745017962,
+      "eval_runtime": 0.6687,
+      "eval_samples_per_second": 1559.856,
+      "eval_steps_per_second": 7.478,
       "step": 136
     },
     {
       "epoch": 5.0,
-      "grad_norm": 1.7911914587020874,
+      "grad_norm": 2.465773105621338,
       "learning_rate": 4.5e-05,
-      "loss": 0.4723,
+      "loss": 0.4591,
       "step": 170
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.6433365345001221,
-      "eval_loss": 0.6780942678451538,
-      "eval_matthews_correlation": 0.13224881537677105,
-      "eval_runtime": 0.6837,
-      "eval_samples_per_second": 1525.525,
-      "eval_steps_per_second": 7.313,
+      "eval_accuracy": 0.6490891575813293,
+      "eval_loss": 0.6875084638595581,
+      "eval_matthews_correlation": 0.09892672481640027,
+      "eval_runtime": 0.6908,
+      "eval_samples_per_second": 1509.867,
+      "eval_steps_per_second": 7.238,
       "step": 170
     },
     {
       "epoch": 6.0,
-      "grad_norm": 2.9178237915039062,
+      "grad_norm": 4.0577287673950195,
       "learning_rate": 4.4000000000000006e-05,
-      "loss": 0.4133,
+      "loss": 0.412,
       "step": 204
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.6155321002006531,
-      "eval_loss": 0.781400740146637,
-      "eval_matthews_correlation": 0.10001682090743959,
-      "eval_runtime": 0.6758,
-      "eval_samples_per_second": 1543.271,
-      "eval_steps_per_second": 7.398,
+      "eval_accuracy": 0.6308724880218506,
+      "eval_loss": 0.7450611591339111,
+      "eval_matthews_correlation": 0.06511003868382283,
+      "eval_runtime": 0.6768,
+      "eval_samples_per_second": 1541.14,
+      "eval_steps_per_second": 7.388,
       "step": 204
     },
     {
       "epoch": 7.0,
-      "grad_norm": 2.418649196624756,
+      "grad_norm": 2.787646532058716,
       "learning_rate": 4.3e-05,
-      "loss": 0.3619,
+      "loss": 0.3633,
       "step": 238
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.6289549469947815,
-      "eval_loss": 0.8527724146842957,
-      "eval_matthews_correlation": 0.11484818571489955,
-      "eval_runtime": 0.7193,
-      "eval_samples_per_second": 1450.052,
-      "eval_steps_per_second": 6.951,
+      "eval_accuracy": 0.6203259825706482,
+      "eval_loss": 0.7966165542602539,
+      "eval_matthews_correlation": 0.10894429086010696,
+      "eval_runtime": 0.6801,
+      "eval_samples_per_second": 1533.561,
+      "eval_steps_per_second": 7.352,
       "step": 238
     },
     {
       "epoch": 8.0,
-      "grad_norm": 4.517303943634033,
+      "grad_norm": 3.5457634925842285,
       "learning_rate": 4.2e-05,
-      "loss": 0.321,
+      "loss": 0.3189,
       "step": 272
     },
     {
       "epoch": 8.0,
       "eval_accuracy": 0.6462128758430481,
-      "eval_loss": 0.9528720378875732,
-      "eval_matthews_correlation": 0.12205331754627759,
-      "eval_runtime": 0.6661,
-      "eval_samples_per_second": 1565.727,
-      "eval_steps_per_second": 7.506,
+      "eval_loss": 0.8535791635513306,
+      "eval_matthews_correlation": 0.10018630520281055,
+      "eval_runtime": 0.7183,
+      "eval_samples_per_second": 1452.028,
+      "eval_steps_per_second": 6.961,
       "step": 272
     },
     {
       "epoch": 8.0,
       "step": 272,
       "total_flos": 8999450537533440.0,
-      "train_loss": 0.4899906516075134,
-      "train_runtime": 134.3561,
-      "train_samples_per_second": 3182.216,
-      "train_steps_per_second": 12.653
+      "train_loss": 0.47968793616575356,
+      "train_runtime": 135.9284,
+      "train_samples_per_second": 3145.406,
+      "train_steps_per_second": 12.507
     }
   ],
   "logging_steps": 1,