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 RTE
       type: glue
       config: rte
       split: validation
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.5415162454873647
+      value: 0.5595667870036101
 ---
 
 <!-- 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_rte
 
-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 RTE dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.1810
-- Accuracy: 0.5415
+- Loss: 0.6853
+- Accuracy: 0.5596
 
 ## Model description
 

all_results.json

@@ -1,14 +1,14 @@
 {
-    "epoch": 6.0,
-    "eval_accuracy": 0.4729241877256318,
-    "eval_loss": 0.7371178269386292,
-    "eval_runtime": 0.4898,
+    "epoch": 8.0,
+    "eval_accuracy": 0.5595667870036101,
+    "eval_loss": 0.6853141784667969,
+    "eval_runtime": 0.4833,
     "eval_samples": 277,
-    "eval_samples_per_second": 565.501,
-    "eval_steps_per_second": 6.125,
-    "train_loss": 9.316140111287435,
-    "train_runtime": 100.0781,
+    "eval_samples_per_second": 573.129,
+    "eval_steps_per_second": 6.207,
+    "train_loss": 0.5432160198688507,
+    "train_runtime": 143.285,
     "train_samples": 2490,
-    "train_samples_per_second": 1244.029,
-    "train_steps_per_second": 9.992
+    "train_samples_per_second": 868.898,
+    "train_steps_per_second": 6.979
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 6.0,
-    "eval_accuracy": 0.4729241877256318,
-    "eval_loss": 0.7371178269386292,
-    "eval_runtime": 0.4898,
+    "epoch": 8.0,
+    "eval_accuracy": 0.5595667870036101,
+    "eval_loss": 0.6853141784667969,
+    "eval_runtime": 0.4833,
     "eval_samples": 277,
-    "eval_samples_per_second": 565.501,
-    "eval_steps_per_second": 6.125
+    "eval_samples_per_second": 573.129,
+    "eval_steps_per_second": 6.207
 }

logs/events.out.tfevents.1686053254.serv-3317.3377083.17

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fae15e350757078b396b5fc1b25c4bd9d3cfd1bca4e4afa3b96bed956f1728af
+size 363

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 6.0,
-    "train_loss": 9.316140111287435,
-    "train_runtime": 100.0781,
+    "epoch": 8.0,
+    "train_loss": 0.5432160198688507,
+    "train_runtime": 143.285,
     "train_samples": 2490,
-    "train_samples_per_second": 1244.029,
-    "train_steps_per_second": 9.992
+    "train_samples_per_second": 868.898,
+    "train_steps_per_second": 6.979
 }

trainer_state.json

@@ -1,115 +1,145 @@
 {
-  "best_metric": 0.7371178269386292,
-  "best_model_checkpoint": "hBERTv2_new_pretrain_rte/checkpoint-20",
-  "epoch": 6.0,
-  "global_step": 120,
+  "best_metric": 0.6853141784667969,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_rte/checkpoint-60",
+  "epoch": 8.0,
+  "global_step": 160,
   "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": 9.1657,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 0.7512,
       "step": 20
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 0.7371178269386292,
-      "eval_runtime": 0.4926,
-      "eval_samples_per_second": 562.316,
-      "eval_steps_per_second": 6.09,
+      "eval_accuracy": 0.4981949458483754,
+      "eval_loss": 0.6928728818893433,
+      "eval_runtime": 0.4852,
+      "eval_samples_per_second": 570.956,
+      "eval_steps_per_second": 6.184,
       "step": 20
     },
     {
       "epoch": 2.0,
-      "learning_rate": 0.00048,
-      "loss": 9.3188,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.7204,
       "step": 40
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.5270758122743683,
-      "eval_loss": 0.7859036326408386,
-      "eval_runtime": 0.4896,
-      "eval_samples_per_second": 565.793,
-      "eval_steps_per_second": 6.128,
+      "eval_loss": 0.6908421516418457,
+      "eval_runtime": 0.4837,
+      "eval_samples_per_second": 572.644,
+      "eval_steps_per_second": 6.202,
       "step": 40
     },
     {
       "epoch": 3.0,
-      "learning_rate": 0.00047,
-      "loss": 9.3339,
+      "learning_rate": 3.76e-05,
+      "loss": 0.701,
       "step": 60
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.5270758122743683,
-      "eval_loss": 0.9236378073692322,
-      "eval_runtime": 0.4893,
-      "eval_samples_per_second": 566.073,
-      "eval_steps_per_second": 6.131,
+      "eval_accuracy": 0.5595667870036101,
+      "eval_loss": 0.6853141784667969,
+      "eval_runtime": 0.479,
+      "eval_samples_per_second": 578.295,
+      "eval_steps_per_second": 6.263,
       "step": 60
     },
     {
       "epoch": 4.0,
-      "learning_rate": 0.00046,
-      "loss": 9.7488,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.6315,
       "step": 80
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.5270758122743683,
-      "eval_loss": 0.914161205291748,
-      "eval_runtime": 0.4899,
-      "eval_samples_per_second": 565.365,
-      "eval_steps_per_second": 6.123,
+      "eval_accuracy": 0.5631768953068592,
+      "eval_loss": 0.7080695033073425,
+      "eval_runtime": 0.4845,
+      "eval_samples_per_second": 571.714,
+      "eval_steps_per_second": 6.192,
       "step": 80
     },
     {
       "epoch": 5.0,
-      "learning_rate": 0.00045000000000000004,
-      "loss": 9.1542,
+      "learning_rate": 3.6e-05,
+      "loss": 0.5807,
       "step": 100
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.5270758122743683,
-      "eval_loss": 0.8327292799949646,
-      "eval_runtime": 0.4923,
-      "eval_samples_per_second": 562.691,
-      "eval_steps_per_second": 6.094,
+      "eval_accuracy": 0.5342960288808665,
+      "eval_loss": 0.8746474981307983,
+      "eval_runtime": 0.4837,
+      "eval_samples_per_second": 572.719,
+      "eval_steps_per_second": 6.203,
       "step": 100
     },
     {
       "epoch": 6.0,
-      "learning_rate": 0.00044,
-      "loss": 9.1755,
+      "learning_rate": 3.52e-05,
+      "loss": 0.4079,
       "step": 120
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.5270758122743683,
-      "eval_loss": 0.8221527934074402,
-      "eval_runtime": 0.4828,
-      "eval_samples_per_second": 573.69,
-      "eval_steps_per_second": 6.213,
+      "eval_accuracy": 0.5631768953068592,
+      "eval_loss": 0.8831439018249512,
+      "eval_runtime": 0.4787,
+      "eval_samples_per_second": 578.592,
+      "eval_steps_per_second": 6.266,
       "step": 120
     },
     {
-      "epoch": 6.0,
-      "step": 120,
-      "total_flos": 2186307132456960.0,
-      "train_loss": 9.316140111287435,
-      "train_runtime": 100.0781,
-      "train_samples_per_second": 1244.029,
-      "train_steps_per_second": 9.992
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.3077,
+      "step": 140
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5487364620938628,
+      "eval_loss": 1.0778553485870361,
+      "eval_runtime": 0.479,
+      "eval_samples_per_second": 578.245,
+      "eval_steps_per_second": 6.263,
+      "step": 140
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.2453,
+      "step": 160
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5415162454873647,
+      "eval_loss": 1.1809593439102173,
+      "eval_runtime": 0.479,
+      "eval_samples_per_second": 578.29,
+      "eval_steps_per_second": 6.263,
+      "step": 160
+    },
+    {
+      "epoch": 8.0,
+      "step": 160,
+      "total_flos": 2915076176609280.0,
+      "train_loss": 0.5432160198688507,
+      "train_runtime": 143.285,
+      "train_samples_per_second": 868.898,
+      "train_steps_per_second": 6.979
     }
   ],
   "max_steps": 1000,
   "num_train_epochs": 50,
-  "total_flos": 2186307132456960.0,
+  "total_flos": 2915076176609280.0,
   "trial_name": null,
   "trial_params": null
 }