End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 tags:
 - generated_from_trainer
 datasets:
@@ -12,7 +14,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE WNLI
       type: glue
       config: wnli
       split: validation
@@ -28,9 +30,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv2_new_pretrain_wnli
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new) on the GLUE WNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6963
+- Loss: 0.6857
 - Accuracy: 0.5634
 
 ## Model description

all_results.json

@@ -1,14 +1,14 @@
 {
-    "epoch": 7.0,
-    "eval_accuracy": 0.43661971830985913,
-    "eval_loss": 0.9150529503822327,
-    "eval_runtime": 0.1329,
+    "epoch": 15.0,
+    "eval_accuracy": 0.5633802816901409,
+    "eval_loss": 0.685739278793335,
+    "eval_runtime": 0.1358,
     "eval_samples": 71,
-    "eval_samples_per_second": 534.404,
-    "eval_steps_per_second": 7.527,
-    "train_loss": 9.753294808523995,
-    "train_runtime": 48.7913,
+    "eval_samples_per_second": 522.699,
+    "eval_steps_per_second": 7.362,
+    "train_loss": 0.7134783363342285,
+    "train_runtime": 107.954,
     "train_samples": 635,
-    "train_samples_per_second": 650.731,
-    "train_steps_per_second": 5.124
+    "train_samples_per_second": 294.107,
+    "train_steps_per_second": 2.316
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 7.0,
-    "eval_accuracy": 0.43661971830985913,
-    "eval_loss": 0.9150529503822327,
-    "eval_runtime": 0.1329,
+    "epoch": 15.0,
+    "eval_accuracy": 0.5633802816901409,
+    "eval_loss": 0.685739278793335,
+    "eval_runtime": 0.1358,
     "eval_samples": 71,
-    "eval_samples_per_second": 534.404,
-    "eval_steps_per_second": 7.527
+    "eval_samples_per_second": 522.699,
+    "eval_steps_per_second": 7.362
 }

logs/events.out.tfevents.1686054155.serv-3317.3377083.23

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:301777e06c9dd2333c24b36b6a68cd1aafbaa3520e7caec1696ef6ade457a08e
+size 357

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 7.0,
-    "train_loss": 9.753294808523995,
-    "train_runtime": 48.7913,
+    "epoch": 15.0,
+    "train_loss": 0.7134783363342285,
+    "train_runtime": 107.954,
     "train_samples": 635,
-    "train_samples_per_second": 650.731,
-    "train_steps_per_second": 5.124
+    "train_samples_per_second": 294.107,
+    "train_steps_per_second": 2.316
 }

trainer_state.json

@@ -1,130 +1,250 @@
 {
-  "best_metric": 0.9150529503822327,
-  "best_model_checkpoint": "hBERTv2_new_pretrain_wnli/checkpoint-10",
-  "epoch": 7.0,
-  "global_step": 35,
+  "best_metric": 0.685739278793335,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_wnli/checkpoint-50",
+  "epoch": 15.0,
+  "global_step": 75,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "learning_rate": 0.00049,
-      "loss": 8.8479,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 0.8646,
       "step": 5
     },
     {
       "epoch": 1.0,
       "eval_accuracy": 0.43661971830985913,
-      "eval_loss": 1.9492391347885132,
-      "eval_runtime": 0.1327,
-      "eval_samples_per_second": 535.165,
-      "eval_steps_per_second": 7.538,
+      "eval_loss": 0.7421599626541138,
+      "eval_runtime": 0.1301,
+      "eval_samples_per_second": 545.622,
+      "eval_steps_per_second": 7.685,
       "step": 5
     },
     {
       "epoch": 2.0,
-      "learning_rate": 0.00048,
-      "loss": 9.5465,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.7094,
       "step": 10
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.43661971830985913,
-      "eval_loss": 0.9150529503822327,
-      "eval_runtime": 0.1358,
-      "eval_samples_per_second": 522.721,
-      "eval_steps_per_second": 7.362,
+      "eval_loss": 0.728983461856842,
+      "eval_runtime": 0.1305,
+      "eval_samples_per_second": 544.065,
+      "eval_steps_per_second": 7.663,
       "step": 10
     },
     {
       "epoch": 3.0,
-      "learning_rate": 0.00047,
-      "loss": 8.6319,
+      "learning_rate": 3.76e-05,
+      "loss": 0.7047,
       "step": 15
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.5633802816901409,
-      "eval_loss": 2.451029062271118,
-      "eval_runtime": 0.1296,
-      "eval_samples_per_second": 547.845,
-      "eval_steps_per_second": 7.716,
+      "eval_loss": 0.705270528793335,
+      "eval_runtime": 0.1314,
+      "eval_samples_per_second": 540.283,
+      "eval_steps_per_second": 7.61,
       "step": 15
     },
     {
       "epoch": 4.0,
-      "learning_rate": 0.00046,
-      "loss": 9.7722,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.7203,
       "step": 20
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.5633802816901409,
-      "eval_loss": 4.1092000007629395,
-      "eval_runtime": 0.1294,
-      "eval_samples_per_second": 548.875,
-      "eval_steps_per_second": 7.731,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7021897435188293,
+      "eval_runtime": 0.1332,
+      "eval_samples_per_second": 533.227,
+      "eval_steps_per_second": 7.51,
       "step": 20
     },
     {
       "epoch": 5.0,
-      "learning_rate": 0.00045000000000000004,
-      "loss": 10.479,
+      "learning_rate": 3.6e-05,
+      "loss": 0.7,
       "step": 25
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.5633802816901409,
-      "eval_loss": 4.987680912017822,
-      "eval_runtime": 0.1316,
-      "eval_samples_per_second": 539.47,
-      "eval_steps_per_second": 7.598,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.6976781487464905,
+      "eval_runtime": 0.1278,
+      "eval_samples_per_second": 555.458,
+      "eval_steps_per_second": 7.823,
       "step": 25
     },
     {
       "epoch": 6.0,
-      "learning_rate": 0.00044,
-      "loss": 10.4548,
+      "learning_rate": 3.52e-05,
+      "loss": 0.7098,
       "step": 30
     },
     {
       "epoch": 6.0,
       "eval_accuracy": 0.5633802816901409,
-      "eval_loss": 5.494718551635742,
-      "eval_runtime": 0.1299,
-      "eval_samples_per_second": 546.636,
-      "eval_steps_per_second": 7.699,
+      "eval_loss": 0.688545286655426,
+      "eval_runtime": 0.128,
+      "eval_samples_per_second": 554.484,
+      "eval_steps_per_second": 7.81,
       "step": 30
     },
     {
       "epoch": 7.0,
-      "learning_rate": 0.00043,
-      "loss": 10.5408,
+      "learning_rate": 3.44e-05,
+      "loss": 0.695,
       "step": 35
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.5633802816901409,
-      "eval_loss": 4.8345136642456055,
-      "eval_runtime": 0.136,
-      "eval_samples_per_second": 521.971,
-      "eval_steps_per_second": 7.352,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7045006155967712,
+      "eval_runtime": 0.1277,
+      "eval_samples_per_second": 555.952,
+      "eval_steps_per_second": 7.83,
       "step": 35
     },
     {
-      "epoch": 7.0,
-      "step": 35,
-      "total_flos": 650477628817408.0,
-      "train_loss": 9.753294808523995,
-      "train_runtime": 48.7913,
-      "train_samples_per_second": 650.731,
-      "train_steps_per_second": 5.124
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.7053,
+      "step": 40
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.685794472694397,
+      "eval_runtime": 0.1336,
+      "eval_samples_per_second": 531.6,
+      "eval_steps_per_second": 7.487,
+      "step": 40
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 3.28e-05,
+      "loss": 0.7095,
+      "step": 45
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7069764733314514,
+      "eval_runtime": 0.1289,
+      "eval_samples_per_second": 550.805,
+      "eval_steps_per_second": 7.758,
+      "step": 45
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.7012,
+      "step": 50
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.685739278793335,
+      "eval_runtime": 0.1298,
+      "eval_samples_per_second": 547.203,
+      "eval_steps_per_second": 7.707,
+      "step": 50
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.1200000000000006e-05,
+      "loss": 0.6995,
+      "step": 55
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.4507042253521127,
+      "eval_loss": 0.6968531012535095,
+      "eval_runtime": 0.1309,
+      "eval_samples_per_second": 542.604,
+      "eval_steps_per_second": 7.642,
+      "step": 55
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.0400000000000004e-05,
+      "loss": 0.6913,
+      "step": 60
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6875,
+      "eval_runtime": 0.1296,
+      "eval_samples_per_second": 548.033,
+      "eval_steps_per_second": 7.719,
+      "step": 60
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 2.96e-05,
+      "loss": 0.6963,
+      "step": 65
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.4788732394366197,
+      "eval_loss": 0.6958626508712769,
+      "eval_runtime": 0.1278,
+      "eval_samples_per_second": 555.431,
+      "eval_steps_per_second": 7.823,
+      "step": 65
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 2.8800000000000002e-05,
+      "loss": 0.6996,
+      "step": 70
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7190250158309937,
+      "eval_runtime": 0.1279,
+      "eval_samples_per_second": 555.225,
+      "eval_steps_per_second": 7.82,
+      "step": 70
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 2.8e-05,
+      "loss": 0.6957,
+      "step": 75
+    },
+    {
+      "epoch": 15.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6962751746177673,
+      "eval_runtime": 0.1333,
+      "eval_samples_per_second": 532.62,
+      "eval_steps_per_second": 7.502,
+      "step": 75
+    },
+    {
+      "epoch": 15.0,
+      "step": 75,
+      "total_flos": 1393880633180160.0,
+      "train_loss": 0.7134783363342285,
+      "train_runtime": 107.954,
+      "train_samples_per_second": 294.107,
+      "train_steps_per_second": 2.316
     }
   ],
   "max_steps": 250,
   "num_train_epochs": 50,
-  "total_flos": 650477628817408.0,
+  "total_flos": 1393880633180160.0,
   "trial_name": null,
   "trial_params": null
 }