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 COLA
       type: glue
       config: cola
       split: validation
@@ -28,9 +30,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv2_new_pretrain_cola
 
-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 COLA dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6200
+- Loss: 0.6173
 - Matthews Correlation: 0.0
 
 ## Model description

all_results.json

@@ -1,14 +1,14 @@
 {
-    "epoch": 6.0,
-    "eval_loss": 9.847195625305176,
+    "epoch": 8.0,
+    "eval_loss": 0.6172745823860168,
     "eval_matthews_correlation": 0.0,
-    "eval_runtime": 1.8224,
+    "eval_runtime": 1.7868,
     "eval_samples": 1043,
-    "eval_samples_per_second": 572.309,
-    "eval_steps_per_second": 4.938,
-    "train_loss": 157.5859132074005,
-    "train_runtime": 312.4982,
+    "eval_samples_per_second": 583.712,
+    "eval_steps_per_second": 5.037,
+    "train_loss": 0.6167977461174353,
+    "train_runtime": 414.5842,
     "train_samples": 8551,
-    "train_samples_per_second": 1368.168,
-    "train_steps_per_second": 10.72
+    "train_samples_per_second": 1031.274,
+    "train_steps_per_second": 8.08
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 6.0,
-    "eval_loss": 9.847195625305176,
+    "epoch": 8.0,
+    "eval_loss": 0.6172745823860168,
     "eval_matthews_correlation": 0.0,
-    "eval_runtime": 1.8224,
+    "eval_runtime": 1.7868,
     "eval_samples": 1043,
-    "eval_samples_per_second": 572.309,
-    "eval_steps_per_second": 4.938
+    "eval_samples_per_second": 583.712,
+    "eval_steps_per_second": 5.037
 }

logs/events.out.tfevents.1686032849.serv-3317.3377083.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e40be66665dbca722083c087466a672dcaefe79a4fc26718b0f58b064922fc21
+size 375

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 6.0,
-    "train_loss": 157.5859132074005,
-    "train_runtime": 312.4982,
+    "epoch": 8.0,
+    "train_loss": 0.6167977461174353,
+    "train_runtime": 414.5842,
     "train_samples": 8551,
-    "train_samples_per_second": 1368.168,
-    "train_steps_per_second": 10.72
+    "train_samples_per_second": 1031.274,
+    "train_steps_per_second": 8.08
 }

trainer_state.json

@@ -1,115 +1,145 @@
 {
-  "best_metric": 9.847195625305176,
-  "best_model_checkpoint": "hBERTv2_new_pretrain_cola/checkpoint-67",
-  "epoch": 6.0,
-  "global_step": 402,
+  "best_metric": 0.6172745823860168,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_cola/checkpoint-201",
+  "epoch": 8.0,
+  "global_step": 536,
   "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": 225.85,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 0.6294,
       "step": 67
     },
     {
       "epoch": 1.0,
-      "eval_loss": 9.847195625305176,
+      "eval_loss": 0.6236423850059509,
       "eval_matthews_correlation": 0.0,
-      "eval_runtime": 1.8271,
-      "eval_samples_per_second": 570.863,
-      "eval_steps_per_second": 4.926,
+      "eval_runtime": 1.7907,
+      "eval_samples_per_second": 582.451,
+      "eval_steps_per_second": 5.026,
       "step": 67
     },
     {
       "epoch": 2.0,
-      "learning_rate": 0.00048,
-      "loss": 322.5451,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.6169,
       "step": 134
     },
     {
       "epoch": 2.0,
-      "eval_loss": 176.96644592285156,
+      "eval_loss": 0.6311982870101929,
       "eval_matthews_correlation": 0.0,
-      "eval_runtime": 1.8295,
-      "eval_samples_per_second": 570.11,
-      "eval_steps_per_second": 4.919,
+      "eval_runtime": 1.7917,
+      "eval_samples_per_second": 582.113,
+      "eval_steps_per_second": 5.023,
       "step": 134
     },
     {
       "epoch": 3.0,
-      "learning_rate": 0.00047,
-      "loss": 88.5896,
+      "learning_rate": 3.76e-05,
+      "loss": 0.6115,
       "step": 201
     },
     {
       "epoch": 3.0,
-      "eval_loss": 127.87056732177734,
+      "eval_loss": 0.6172745823860168,
       "eval_matthews_correlation": 0.0,
-      "eval_runtime": 1.8269,
-      "eval_samples_per_second": 570.919,
-      "eval_steps_per_second": 4.926,
+      "eval_runtime": 1.7888,
+      "eval_samples_per_second": 583.068,
+      "eval_steps_per_second": 5.031,
       "step": 201
     },
     {
       "epoch": 4.0,
-      "learning_rate": 0.00046,
-      "loss": 121.8729,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.6372,
       "step": 268
     },
     {
       "epoch": 4.0,
-      "eval_loss": 106.45637512207031,
+      "eval_loss": 0.620063841342926,
       "eval_matthews_correlation": 0.0,
-      "eval_runtime": 1.8301,
-      "eval_samples_per_second": 569.928,
-      "eval_steps_per_second": 4.918,
+      "eval_runtime": 1.7867,
+      "eval_samples_per_second": 583.759,
+      "eval_steps_per_second": 5.037,
       "step": 268
     },
     {
       "epoch": 5.0,
-      "learning_rate": 0.00045000000000000004,
-      "loss": 99.3696,
+      "learning_rate": 3.6e-05,
+      "loss": 0.6087,
       "step": 335
     },
     {
       "epoch": 5.0,
-      "eval_loss": 85.71812438964844,
+      "eval_loss": 0.6216873526573181,
       "eval_matthews_correlation": 0.0,
-      "eval_runtime": 1.827,
-      "eval_samples_per_second": 570.868,
-      "eval_steps_per_second": 4.926,
+      "eval_runtime": 1.7924,
+      "eval_samples_per_second": 581.886,
+      "eval_steps_per_second": 5.021,
       "step": 335
     },
     {
       "epoch": 6.0,
-      "learning_rate": 0.00044,
-      "loss": 87.2883,
+      "learning_rate": 3.52e-05,
+      "loss": 0.6086,
       "step": 402
     },
     {
       "epoch": 6.0,
-      "eval_loss": 79.47842407226562,
+      "eval_loss": 0.624801516532898,
       "eval_matthews_correlation": 0.0,
-      "eval_runtime": 1.8262,
-      "eval_samples_per_second": 571.122,
-      "eval_steps_per_second": 4.928,
+      "eval_runtime": 1.7896,
+      "eval_samples_per_second": 582.822,
+      "eval_steps_per_second": 5.029,
       "step": 402
     },
     {
-      "epoch": 6.0,
-      "step": 402,
-      "total_flos": 7508077427294208.0,
-      "train_loss": 157.5859132074005,
-      "train_runtime": 312.4982,
-      "train_samples_per_second": 1368.168,
-      "train_steps_per_second": 10.72
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.6113,
+      "step": 469
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 0.6282826662063599,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.7883,
+      "eval_samples_per_second": 583.234,
+      "eval_steps_per_second": 5.033,
+      "step": 469
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.6109,
+      "step": 536
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 0.620032012462616,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.8037,
+      "eval_samples_per_second": 578.242,
+      "eval_steps_per_second": 4.99,
+      "step": 536
+    },
+    {
+      "epoch": 8.0,
+      "step": 536,
+      "total_flos": 1.0010769903058944e+16,
+      "train_loss": 0.6167977461174353,
+      "train_runtime": 414.5842,
+      "train_samples_per_second": 1031.274,
+      "train_steps_per_second": 8.08
     }
   ],
   "max_steps": 3350,
   "num_train_epochs": 50,
-  "total_flos": 7508077427294208.0,
+  "total_flos": 1.0010769903058944e+16,
   "trial_name": null,
   "trial_params": null
 }