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 QNLI
       type: glue
       config: qnli
       split: validation
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.5086948563060589
+      value: 0.5502471169686985
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -28,10 +30,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv2_new_pretrain_qnli
 
-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 QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6930
-- Accuracy: 0.5087
+- Loss: 0.6870
+- Accuracy: 0.5502
 
 ## Model description
 

all_results.json

@@ -1,14 +1,14 @@
 {
-    "epoch": 9.0,
-    "eval_accuracy": 0.4946000366099213,
-    "eval_loss": 0.7446396946907043,
-    "eval_runtime": 9.4896,
+    "epoch": 7.0,
+    "eval_accuracy": 0.5502471169686985,
+    "eval_loss": 0.6869744658470154,
+    "eval_runtime": 9.2856,
     "eval_samples": 5463,
-    "eval_samples_per_second": 575.68,
-    "eval_steps_per_second": 4.531,
-    "train_loss": 11.666525236674977,
-    "train_runtime": 4965.9529,
+    "eval_samples_per_second": 588.333,
+    "eval_steps_per_second": 4.631,
+    "train_loss": 0.69172189108015,
+    "train_runtime": 3811.1135,
     "train_samples": 104743,
-    "train_samples_per_second": 1054.611,
-    "train_steps_per_second": 8.246
+    "train_samples_per_second": 1374.178,
+    "train_steps_per_second": 10.745
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 9.0,
-    "eval_accuracy": 0.4946000366099213,
-    "eval_loss": 0.7446396946907043,
-    "eval_runtime": 9.4896,
+    "epoch": 7.0,
+    "eval_accuracy": 0.5502471169686985,
+    "eval_loss": 0.6869744658470154,
+    "eval_runtime": 9.2856,
     "eval_samples": 5463,
-    "eval_samples_per_second": 575.68,
-    "eval_steps_per_second": 4.531
+    "eval_samples_per_second": 588.333,
+    "eval_steps_per_second": 4.631
 }

logs/events.out.tfevents.1686037411.serv-3317.3377083.11

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c052fdd92038ad17825152b47c15cae4ba01c66d820eb071559f2170c8b2a5a
+size 363

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 9.0,
-    "train_loss": 11.666525236674977,
-    "train_runtime": 4965.9529,
+    "epoch": 7.0,
+    "train_loss": 0.69172189108015,
+    "train_runtime": 3811.1135,
     "train_samples": 104743,
-    "train_samples_per_second": 1054.611,
-    "train_steps_per_second": 8.246
+    "train_samples_per_second": 1374.178,
+    "train_steps_per_second": 10.745
 }

trainer_state.json

@@ -1,160 +1,130 @@
 {
-  "best_metric": 0.7446396946907043,
-  "best_model_checkpoint": "hBERTv2_new_pretrain_qnli/checkpoint-3276",
-  "epoch": 9.0,
-  "global_step": 7371,
+  "best_metric": 0.6869744658470154,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_qnli/checkpoint-1638",
+  "epoch": 7.0,
+  "global_step": 5733,
   "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": 9.3816,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 0.6921,
       "step": 819
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.874665379524231,
-      "eval_runtime": 9.5189,
-      "eval_samples_per_second": 573.91,
-      "eval_steps_per_second": 4.517,
+      "eval_accuracy": 0.505033864177192,
+      "eval_loss": 0.6903212070465088,
+      "eval_runtime": 9.3321,
+      "eval_samples_per_second": 585.401,
+      "eval_steps_per_second": 4.608,
       "step": 819
     },
     {
       "epoch": 2.0,
-      "learning_rate": 0.00048,
-      "loss": 9.2855,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.6906,
       "step": 1638
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.8931218981742859,
-      "eval_runtime": 9.5253,
-      "eval_samples_per_second": 573.527,
-      "eval_steps_per_second": 4.514,
+      "eval_accuracy": 0.5502471169686985,
+      "eval_loss": 0.6869744658470154,
+      "eval_runtime": 9.3282,
+      "eval_samples_per_second": 585.645,
+      "eval_steps_per_second": 4.61,
       "step": 1638
     },
     {
       "epoch": 3.0,
-      "learning_rate": 0.00047,
-      "loss": 9.3355,
+      "learning_rate": 3.76e-05,
+      "loss": 0.6889,
       "step": 2457
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.7446758151054382,
-      "eval_runtime": 9.5312,
-      "eval_samples_per_second": 573.172,
-      "eval_steps_per_second": 4.512,
+      "eval_accuracy": 0.5231557752150833,
+      "eval_loss": 0.692358672618866,
+      "eval_runtime": 9.3343,
+      "eval_samples_per_second": 585.259,
+      "eval_steps_per_second": 4.607,
       "step": 2457
     },
     {
       "epoch": 4.0,
-      "learning_rate": 0.00046,
-      "loss": 9.3249,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.6932,
       "step": 3276
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.7446396946907043,
-      "eval_runtime": 9.5208,
-      "eval_samples_per_second": 573.799,
-      "eval_steps_per_second": 4.516,
+      "eval_accuracy": 0.5260845689181768,
+      "eval_loss": 0.6916021704673767,
+      "eval_runtime": 9.3343,
+      "eval_samples_per_second": 585.262,
+      "eval_steps_per_second": 4.607,
       "step": 3276
     },
     {
       "epoch": 5.0,
-      "learning_rate": 0.00045000000000000004,
-      "loss": 9.2866,
+      "learning_rate": 3.6e-05,
+      "loss": 0.6927,
       "step": 4095
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.9298637509346008,
-      "eval_runtime": 9.5464,
-      "eval_samples_per_second": 572.255,
-      "eval_steps_per_second": 4.504,
+      "eval_accuracy": 0.5313930075050338,
+      "eval_loss": 0.692005455493927,
+      "eval_runtime": 9.3373,
+      "eval_samples_per_second": 585.071,
+      "eval_steps_per_second": 4.605,
       "step": 4095
     },
     {
       "epoch": 6.0,
-      "learning_rate": 0.00044,
-      "loss": 9.5308,
+      "learning_rate": 3.52e-05,
+      "loss": 0.6918,
       "step": 4914
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 4.201156139373779,
-      "eval_runtime": 9.523,
-      "eval_samples_per_second": 573.663,
-      "eval_steps_per_second": 4.515,
+      "eval_accuracy": 0.5297455610470437,
+      "eval_loss": 0.689666211605072,
+      "eval_runtime": 9.3306,
+      "eval_samples_per_second": 585.493,
+      "eval_steps_per_second": 4.608,
       "step": 4914
     },
     {
       "epoch": 7.0,
-      "learning_rate": 0.00043,
-      "loss": 9.1933,
+      "learning_rate": 3.44e-05,
+      "loss": 0.6927,
       "step": 5733
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 1.3372163772583008,
-      "eval_runtime": 9.5484,
-      "eval_samples_per_second": 572.138,
-      "eval_steps_per_second": 4.503,
+      "eval_accuracy": 0.5086948563060589,
+      "eval_loss": 0.692981481552124,
+      "eval_runtime": 9.3271,
+      "eval_samples_per_second": 585.713,
+      "eval_steps_per_second": 4.61,
       "step": 5733
     },
     {
-      "epoch": 8.0,
-      "learning_rate": 0.00042,
-      "loss": 32.7671,
-      "step": 6552
-    },
-    {
-      "epoch": 8.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 18.067407608032227,
-      "eval_runtime": 9.5244,
-      "eval_samples_per_second": 573.582,
-      "eval_steps_per_second": 4.515,
-      "step": 6552
-    },
-    {
-      "epoch": 9.0,
-      "learning_rate": 0.00041,
-      "loss": 6.8935,
-      "step": 7371
-    },
-    {
-      "epoch": 9.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 1.65470552444458,
-      "eval_runtime": 9.5357,
-      "eval_samples_per_second": 572.901,
-      "eval_steps_per_second": 4.509,
-      "step": 7371
-    },
-    {
-      "epoch": 9.0,
-      "step": 7371,
-      "total_flos": 1.3795203456014746e+17,
-      "train_loss": 11.666525236674977,
-      "train_runtime": 4965.9529,
-      "train_samples_per_second": 1054.611,
-      "train_steps_per_second": 8.246
+      "epoch": 7.0,
+      "step": 5733,
+      "total_flos": 1.0729602688011469e+17,
+      "train_loss": 0.69172189108015,
+      "train_runtime": 3811.1135,
+      "train_samples_per_second": 1374.178,
+      "train_steps_per_second": 10.745
     }
   ],
   "max_steps": 40950,
   "num_train_epochs": 50,
-  "total_flos": 1.3795203456014746e+17,
+  "total_flos": 1.0729602688011469e+17,
   "trial_name": null,
   "trial_params": null
 }