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: 0.5215083287570932
+      value: 0.528830313014827
 ---
 
 <!-- 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. -->
 
 # add_BERT_no_pretrain_qnli
 
-This model is a fine-tuned version of [](https://huggingface.co/) on the glue dataset.
+This model is a fine-tuned version of [](https://huggingface.co/) on the GLUE QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6922
-- Accuracy: 0.5215
+- Loss: 0.6899
+- Accuracy: 0.5288
 
 ## Model description
 

all_results.json

@@ -1,14 +1,14 @@
 {
-    "epoch": 6.0,
-    "eval_accuracy": 0.5053999633900788,
-    "eval_loss": 0.6930387616157532,
-    "eval_runtime": 8.4832,
+    "epoch": 9.0,
+    "eval_accuracy": 0.528830313014827,
+    "eval_loss": 0.6899052858352661,
+    "eval_runtime": 10.1354,
     "eval_samples": 5463,
-    "eval_samples_per_second": 643.979,
-    "eval_steps_per_second": 5.069,
-    "train_loss": 0.697838950555015,
-    "train_runtime": 2914.3249,
+    "eval_samples_per_second": 539.0,
+    "eval_steps_per_second": 4.243,
+    "train_loss": 0.6946712540991512,
+    "train_runtime": 5581.0967,
     "train_samples": 104743,
-    "train_samples_per_second": 1797.037,
-    "train_steps_per_second": 14.051
+    "train_samples_per_second": 938.373,
+    "train_steps_per_second": 7.337
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 6.0,
-    "eval_accuracy": 0.5053999633900788,
-    "eval_loss": 0.6930387616157532,
-    "eval_runtime": 8.4832,
+    "epoch": 9.0,
+    "eval_accuracy": 0.528830313014827,
+    "eval_loss": 0.6899052858352661,
+    "eval_runtime": 10.1354,
     "eval_samples": 5463,
-    "eval_samples_per_second": 643.979,
-    "eval_steps_per_second": 5.069
+    "eval_samples_per_second": 539.0,
+    "eval_steps_per_second": 4.243
 }

logs/events.out.tfevents.1686753111.serv-9219.1715196.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9514a24dcd093dc5d3fe4c58a99670587563fe74dafbbdd395809ab4c640db6b
+size 363

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 6.0,
-    "train_loss": 0.697838950555015,
-    "train_runtime": 2914.3249,
+    "epoch": 9.0,
+    "train_loss": 0.6946712540991512,
+    "train_runtime": 5581.0967,
     "train_samples": 104743,
-    "train_samples_per_second": 1797.037,
-    "train_steps_per_second": 14.051
+    "train_samples_per_second": 938.373,
+    "train_steps_per_second": 7.337
 }

trainer_state.json

@@ -1,115 +1,160 @@
 {
-  "best_metric": 0.6930387616157532,
-  "best_model_checkpoint": "add_BERT_no_pretrain_qnli/checkpoint-819",
-  "epoch": 6.0,
-  "global_step": 4914,
+  "best_metric": 0.6899052858352661,
+  "best_model_checkpoint": "add_BERT_no_pretrain_qnli/checkpoint-3276",
+  "epoch": 9.0,
+  "global_step": 7371,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "learning_rate": 0.0004900366300366301,
-      "loss": 0.7205,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 0.7079,
       "step": 819
     },
     {
       "epoch": 1.0,
       "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6930387616157532,
-      "eval_runtime": 8.5235,
-      "eval_samples_per_second": 640.931,
-      "eval_steps_per_second": 5.045,
+      "eval_loss": 0.7209677696228027,
+      "eval_runtime": 10.2227,
+      "eval_samples_per_second": 534.4,
+      "eval_steps_per_second": 4.206,
       "step": 819
     },
     {
       "epoch": 2.0,
-      "learning_rate": 0.00048003663003663005,
-      "loss": 0.6935,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.6952,
       "step": 1638
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6930967569351196,
-      "eval_runtime": 8.519,
-      "eval_samples_per_second": 641.274,
-      "eval_steps_per_second": 5.048,
+      "eval_accuracy": 0.4946000366099213,
+      "eval_loss": 0.6911802291870117,
+      "eval_runtime": 10.2371,
+      "eval_samples_per_second": 533.649,
+      "eval_steps_per_second": 4.2,
       "step": 1638
     },
     {
       "epoch": 3.0,
-      "learning_rate": 0.00047003663003663003,
-      "loss": 0.6932,
+      "learning_rate": 3.76e-05,
+      "loss": 0.6922,
       "step": 2457
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6930704116821289,
-      "eval_runtime": 8.5483,
-      "eval_samples_per_second": 639.078,
-      "eval_steps_per_second": 5.03,
+      "eval_accuracy": 0.5279150649826103,
+      "eval_loss": 0.6905035972595215,
+      "eval_runtime": 10.231,
+      "eval_samples_per_second": 533.964,
+      "eval_steps_per_second": 4.203,
       "step": 2457
     },
     {
       "epoch": 4.0,
-      "learning_rate": 0.00046003663003663006,
-      "loss": 0.6932,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.6918,
       "step": 3276
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.6931969523429871,
-      "eval_runtime": 8.5019,
-      "eval_samples_per_second": 642.559,
-      "eval_steps_per_second": 5.058,
+      "eval_accuracy": 0.528830313014827,
+      "eval_loss": 0.6899052858352661,
+      "eval_runtime": 10.2564,
+      "eval_samples_per_second": 532.644,
+      "eval_steps_per_second": 4.193,
       "step": 3276
     },
     {
       "epoch": 5.0,
-      "learning_rate": 0.00045003663003663003,
-      "loss": 0.6932,
+      "learning_rate": 3.6e-05,
+      "loss": 0.6922,
       "step": 4095
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6931020617485046,
-      "eval_runtime": 8.5439,
-      "eval_samples_per_second": 639.401,
-      "eval_steps_per_second": 5.033,
+      "eval_accuracy": 0.5152846421380194,
+      "eval_loss": 0.6922001838684082,
+      "eval_runtime": 10.2507,
+      "eval_samples_per_second": 532.941,
+      "eval_steps_per_second": 4.195,
       "step": 4095
     },
     {
       "epoch": 6.0,
-      "learning_rate": 0.00044003663003663,
-      "loss": 0.6935,
+      "learning_rate": 3.52e-05,
+      "loss": 0.6933,
       "step": 4914
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6933435797691345,
-      "eval_runtime": 8.5341,
-      "eval_samples_per_second": 640.137,
-      "eval_steps_per_second": 5.039,
+      "eval_accuracy": 0.5127219476478125,
+      "eval_loss": 0.6926350593566895,
+      "eval_runtime": 10.2253,
+      "eval_samples_per_second": 534.263,
+      "eval_steps_per_second": 4.205,
       "step": 4914
     },
     {
-      "epoch": 6.0,
-      "step": 4914,
-      "total_flos": 8.975886795669504e+16,
-      "train_loss": 0.697838950555015,
-      "train_runtime": 2914.3249,
-      "train_samples_per_second": 1797.037,
-      "train_steps_per_second": 14.051
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.6931,
+      "step": 5733
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.4946000366099213,
+      "eval_loss": 0.6952403783798218,
+      "eval_runtime": 10.2187,
+      "eval_samples_per_second": 534.609,
+      "eval_steps_per_second": 4.208,
+      "step": 5733
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.6933,
+      "step": 6552
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5112575507962658,
+      "eval_loss": 0.6927889585494995,
+      "eval_runtime": 10.2381,
+      "eval_samples_per_second": 533.597,
+      "eval_steps_per_second": 4.2,
+      "step": 6552
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 3.28e-05,
+      "loss": 0.693,
+      "step": 7371
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.5215083287570932,
+      "eval_loss": 0.6921678781509399,
+      "eval_runtime": 10.2872,
+      "eval_samples_per_second": 531.048,
+      "eval_steps_per_second": 4.18,
+      "step": 7371
+    },
+    {
+      "epoch": 9.0,
+      "step": 7371,
+      "total_flos": 1.3463830193504256e+17,
+      "train_loss": 0.6946712540991512,
+      "train_runtime": 5581.0967,
+      "train_samples_per_second": 938.373,
+      "train_steps_per_second": 7.337
     }
   ],
   "max_steps": 40950,
   "num_train_epochs": 50,
-  "total_flos": 8.975886795669504e+16,
+  "total_flos": 1.3463830193504256e+17,
   "trial_name": null,
   "trial_params": null
 }