End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: gokulsrinivasagan/bert_base_lda_100_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: bert_base_lda_100_v1_mnli
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE MNLI
+      type: glue
+      args: mnli
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.7162327095199349
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -15,10 +30,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # bert_base_lda_100_v1_mnli
 
-This model is a fine-tuned version of [gokulsrinivasagan/bert_base_lda_100_v1](https://huggingface.co/gokulsrinivasagan/bert_base_lda_100_v1) on an unknown dataset.
+This model is a fine-tuned version of [gokulsrinivasagan/bert_base_lda_100_v1](https://huggingface.co/gokulsrinivasagan/bert_base_lda_100_v1) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0094
-- Accuracy: 0.7111
+- Loss: 0.6799
+- Accuracy: 0.7162
 
 ## Model description
 

all_results.json

@@ -1,22 +1,22 @@
 {
-    "epoch": 14.0,
-    "epoch_mm": 14.0,
-    "eval_accuracy": 0.3273560876209883,
-    "eval_accuracy_mm": 0.3295362082994304,
-    "eval_loss": 1.0962358713150024,
-    "eval_loss_mm": 1.0962363481521606,
-    "eval_runtime": 6.2458,
-    "eval_runtime_mm": 6.2414,
+    "epoch": 9.0,
+    "epoch_mm": 9.0,
+    "eval_accuracy": 0.7112582781456953,
+    "eval_accuracy_mm": 0.7162327095199349,
+    "eval_loss": 0.7055869698524475,
+    "eval_loss_mm": 0.6798676252365112,
+    "eval_runtime": 6.4534,
+    "eval_runtime_mm": 6.4286,
     "eval_samples": 9815,
     "eval_samples_mm": 9832,
-    "eval_samples_per_second": 1571.45,
-    "eval_samples_per_second_mm": 1575.297,
-    "eval_steps_per_second": 6.244,
-    "eval_steps_per_second_mm": 6.249,
-    "total_flos": 7.232761575803474e+17,
-    "train_loss": 1.098970256541823,
-    "train_runtime": 8924.8215,
+    "eval_samples_per_second": 1520.914,
+    "eval_samples_per_second_mm": 1529.421,
+    "eval_steps_per_second": 6.043,
+    "eval_steps_per_second_mm": 6.067,
+    "total_flos": 4.6496324415879475e+17,
+    "train_loss": 0.5496212416207741,
+    "train_runtime": 5787.733,
     "train_samples": 392702,
-    "train_samples_per_second": 2200.055,
-    "train_steps_per_second": 8.594
+    "train_samples_per_second": 3392.537,
+    "train_steps_per_second": 13.252
 }

eval_results.json

@@ -1,16 +1,16 @@
 {
-    "epoch": 14.0,
-    "epoch_mm": 14.0,
-    "eval_accuracy": 0.3273560876209883,
-    "eval_accuracy_mm": 0.3295362082994304,
-    "eval_loss": 1.0962358713150024,
-    "eval_loss_mm": 1.0962363481521606,
-    "eval_runtime": 6.2458,
-    "eval_runtime_mm": 6.2414,
+    "epoch": 9.0,
+    "epoch_mm": 9.0,
+    "eval_accuracy": 0.7112582781456953,
+    "eval_accuracy_mm": 0.7162327095199349,
+    "eval_loss": 0.7055869698524475,
+    "eval_loss_mm": 0.6798676252365112,
+    "eval_runtime": 6.4534,
+    "eval_runtime_mm": 6.4286,
     "eval_samples": 9815,
     "eval_samples_mm": 9832,
-    "eval_samples_per_second": 1571.45,
-    "eval_samples_per_second_mm": 1575.297,
-    "eval_steps_per_second": 6.244,
-    "eval_steps_per_second_mm": 6.249
+    "eval_samples_per_second": 1520.914,
+    "eval_samples_per_second_mm": 1529.421,
+    "eval_steps_per_second": 6.043,
+    "eval_steps_per_second_mm": 6.067
 }

logs/events.out.tfevents.1733323891.ki-g0008.1208741.17

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9316fa285d50c3deae54ed87953db1cca47045239d01fd282433d1f06ccd6497
+size 734

train_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 14.0,
-    "total_flos": 7.232761575803474e+17,
-    "train_loss": 1.098970256541823,
-    "train_runtime": 8924.8215,
+    "epoch": 9.0,
+    "total_flos": 4.6496324415879475e+17,
+    "train_loss": 0.5496212416207741,
+    "train_runtime": 5787.733,
     "train_samples": 392702,
-    "train_samples_per_second": 2200.055,
-    "train_steps_per_second": 8.594
+    "train_samples_per_second": 3392.537,
+    "train_steps_per_second": 13.252
 }

trainer_state.json

@@ -1,245 +1,165 @@
 {
-  "best_metric": 1.0962358713150024,
-  "best_model_checkpoint": "bert_base_lda_100_v1_mnli/checkpoint-13806",
-  "epoch": 14.0,
+  "best_metric": 0.7055869698524475,
+  "best_model_checkpoint": "bert_base_lda_100_v1_mnli/checkpoint-6136",
+  "epoch": 9.0,
   "eval_steps": 500,
-  "global_step": 21476,
+  "global_step": 13806,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 0.02029704488813877,
-      "learning_rate": 0.00098,
-      "loss": 1.1038,
+      "grad_norm": 1.7309693098068237,
+      "learning_rate": 4.9e-05,
+      "loss": 0.9588,
       "step": 1534
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.3273560876209883,
-      "eval_loss": 1.0992883443832397,
-      "eval_runtime": 6.2882,
-      "eval_samples_per_second": 1560.868,
-      "eval_steps_per_second": 6.202,
+      "eval_accuracy": 0.6248599083036169,
+      "eval_loss": 0.8420299887657166,
+      "eval_runtime": 6.3265,
+      "eval_samples_per_second": 1551.4,
+      "eval_steps_per_second": 6.165,
       "step": 1534
     },
     {
       "epoch": 2.0,
-      "grad_norm": 0.11696068942546844,
-      "learning_rate": 0.00096,
-      "loss": 1.0989,
+      "grad_norm": 2.024912118911743,
+      "learning_rate": 4.8e-05,
+      "loss": 0.7857,
       "step": 3068
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.3273560876209883,
-      "eval_loss": 1.0990049839019775,
-      "eval_runtime": 6.2196,
-      "eval_samples_per_second": 1578.081,
-      "eval_steps_per_second": 6.271,
+      "eval_accuracy": 0.680794701986755,
+      "eval_loss": 0.7450552582740784,
+      "eval_runtime": 6.312,
+      "eval_samples_per_second": 1554.976,
+      "eval_steps_per_second": 6.179,
       "step": 3068
     },
     {
       "epoch": 3.0,
-      "grad_norm": 0.03827323019504547,
-      "learning_rate": 0.00094,
-      "loss": 1.0987,
+      "grad_norm": 1.8295753002166748,
+      "learning_rate": 4.7e-05,
+      "loss": 0.6825,
       "step": 4602
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.3273560876209883,
-      "eval_loss": 1.10149085521698,
-      "eval_runtime": 6.3142,
-      "eval_samples_per_second": 1554.441,
-      "eval_steps_per_second": 6.177,
+      "eval_accuracy": 0.6976057055527254,
+      "eval_loss": 0.7161675691604614,
+      "eval_runtime": 6.3639,
+      "eval_samples_per_second": 1542.29,
+      "eval_steps_per_second": 6.128,
       "step": 4602
     },
     {
       "epoch": 4.0,
-      "grad_norm": 0.062206000089645386,
-      "learning_rate": 0.00092,
-      "loss": 1.0985,
+      "grad_norm": 2.0847725868225098,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.5973,
       "step": 6136
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.31818644931227713,
-      "eval_loss": 1.10163414478302,
-      "eval_runtime": 6.3418,
-      "eval_samples_per_second": 1547.678,
-      "eval_steps_per_second": 6.15,
+      "eval_accuracy": 0.7112582781456953,
+      "eval_loss": 0.7055869698524475,
+      "eval_runtime": 6.3116,
+      "eval_samples_per_second": 1555.078,
+      "eval_steps_per_second": 6.179,
       "step": 6136
     },
     {
       "epoch": 5.0,
-      "grad_norm": 0.07343439757823944,
-      "learning_rate": 0.0009000000000000001,
-      "loss": 1.0986,
+      "grad_norm": 2.3416082859039307,
+      "learning_rate": 4.5e-05,
+      "loss": 0.5208,
       "step": 7670
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.3544574630667346,
-      "eval_loss": 1.0988649129867554,
-      "eval_runtime": 6.2795,
-      "eval_samples_per_second": 1563.023,
-      "eval_steps_per_second": 6.211,
+      "eval_accuracy": 0.7144167091186959,
+      "eval_loss": 0.7459524869918823,
+      "eval_runtime": 6.2457,
+      "eval_samples_per_second": 1571.471,
+      "eval_steps_per_second": 6.244,
       "step": 7670
     },
     {
       "epoch": 6.0,
-      "grad_norm": 0.047856856137514114,
-      "learning_rate": 0.00088,
-      "loss": 1.0987,
+      "grad_norm": 2.3869707584381104,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.4464,
       "step": 9204
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.3544574630667346,
-      "eval_loss": 1.0988649129867554,
-      "eval_runtime": 6.2109,
-      "eval_samples_per_second": 1580.289,
-      "eval_steps_per_second": 6.279,
+      "eval_accuracy": 0.7077941925624045,
+      "eval_loss": 0.790677011013031,
+      "eval_runtime": 6.3035,
+      "eval_samples_per_second": 1557.08,
+      "eval_steps_per_second": 6.187,
       "step": 9204
     },
     {
       "epoch": 7.0,
-      "grad_norm": 0.0791020318865776,
-      "learning_rate": 0.00086,
-      "loss": 1.0985,
+      "grad_norm": 2.6693198680877686,
+      "learning_rate": 4.3e-05,
+      "loss": 0.3775,
       "step": 10738
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.31818644931227713,
-      "eval_loss": 1.0968025922775269,
-      "eval_runtime": 6.2652,
-      "eval_samples_per_second": 1566.588,
-      "eval_steps_per_second": 6.225,
+      "eval_accuracy": 0.7171676006113092,
+      "eval_loss": 0.8361655473709106,
+      "eval_runtime": 6.3703,
+      "eval_samples_per_second": 1540.741,
+      "eval_steps_per_second": 6.122,
       "step": 10738
     },
     {
       "epoch": 8.0,
-      "grad_norm": 0.016667626798152924,
-      "learning_rate": 0.00084,
-      "loss": 1.0986,
+      "grad_norm": 3.255511999130249,
+      "learning_rate": 4.2e-05,
+      "loss": 0.316,
       "step": 12272
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.31818644931227713,
-      "eval_loss": 1.0990766286849976,
-      "eval_runtime": 6.3215,
-      "eval_samples_per_second": 1552.637,
-      "eval_steps_per_second": 6.169,
+      "eval_accuracy": 0.7101375445746306,
+      "eval_loss": 0.9462762475013733,
+      "eval_runtime": 6.4,
+      "eval_samples_per_second": 1533.587,
+      "eval_steps_per_second": 6.094,
       "step": 12272
     },
     {
       "epoch": 9.0,
-      "grad_norm": 0.006579674780368805,
-      "learning_rate": 0.00082,
-      "loss": 1.0987,
+      "grad_norm": 3.7821061611175537,
+      "learning_rate": 4.1e-05,
+      "loss": 0.2617,
       "step": 13806
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.3273560876209883,
-      "eval_loss": 1.0962358713150024,
-      "eval_runtime": 6.3164,
-      "eval_samples_per_second": 1553.903,
-      "eval_steps_per_second": 6.174,
+      "eval_accuracy": 0.7110545084055018,
+      "eval_loss": 1.009425401687622,
+      "eval_runtime": 6.5216,
+      "eval_samples_per_second": 1505.005,
+      "eval_steps_per_second": 5.98,
       "step": 13806
     },
     {
-      "epoch": 10.0,
-      "grad_norm": 0.045499037951231,
-      "learning_rate": 0.0008,
-      "loss": 1.0986,
-      "step": 15340
-    },
-    {
-      "epoch": 10.0,
-      "eval_accuracy": 0.3273560876209883,
-      "eval_loss": 1.0992168188095093,
-      "eval_runtime": 6.299,
-      "eval_samples_per_second": 1558.187,
-      "eval_steps_per_second": 6.191,
-      "step": 15340
-    },
-    {
-      "epoch": 11.0,
-      "grad_norm": 0.009605828672647476,
-      "learning_rate": 0.0007800000000000001,
-      "loss": 1.0985,
-      "step": 16874
-    },
-    {
-      "epoch": 11.0,
-      "eval_accuracy": 0.3273560876209883,
-      "eval_loss": 1.0990049839019775,
-      "eval_runtime": 6.2315,
-      "eval_samples_per_second": 1575.066,
-      "eval_steps_per_second": 6.259,
-      "step": 16874
-    },
-    {
-      "epoch": 12.0,
-      "grad_norm": 0.03646823763847351,
-      "learning_rate": 0.00076,
-      "loss": 1.0984,
-      "step": 18408
-    },
-    {
-      "epoch": 12.0,
-      "eval_accuracy": 0.31818644931227713,
-      "eval_loss": 1.0990766286849976,
-      "eval_runtime": 6.1815,
-      "eval_samples_per_second": 1587.791,
-      "eval_steps_per_second": 6.309,
-      "step": 18408
-    },
-    {
-      "epoch": 13.0,
-      "grad_norm": 0.09090559184551239,
-      "learning_rate": 0.00074,
-      "loss": 1.0985,
-      "step": 19942
-    },
-    {
-      "epoch": 13.0,
-      "eval_accuracy": 0.31818644931227713,
-      "eval_loss": 1.0965192317962646,
-      "eval_runtime": 6.2476,
-      "eval_samples_per_second": 1571.012,
-      "eval_steps_per_second": 6.242,
-      "step": 19942
-    },
-    {
-      "epoch": 14.0,
-      "grad_norm": 0.03515646234154701,
-      "learning_rate": 0.0007199999999999999,
-      "loss": 1.0987,
-      "step": 21476
-    },
-    {
-      "epoch": 14.0,
-      "eval_accuracy": 0.3273560876209883,
-      "eval_loss": 1.0987216234207153,
-      "eval_runtime": 6.245,
-      "eval_samples_per_second": 1571.652,
-      "eval_steps_per_second": 6.245,
-      "step": 21476
-    },
-    {
-      "epoch": 14.0,
-      "step": 21476,
-      "total_flos": 7.232761575803474e+17,
-      "train_loss": 1.098970256541823,
-      "train_runtime": 8924.8215,
-      "train_samples_per_second": 2200.055,
-      "train_steps_per_second": 8.594
+      "epoch": 9.0,
+      "step": 13806,
+      "total_flos": 4.6496324415879475e+17,
+      "train_loss": 0.5496212416207741,
+      "train_runtime": 5787.733,
+      "train_samples_per_second": 3392.537,
+      "train_steps_per_second": 13.252
     }
   ],
   "logging_steps": 1,
@@ -268,7 +188,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 7.232761575803474e+17,
+  "total_flos": 4.6496324415879475e+17,
   "train_batch_size": 256,
   "trial_name": null,
   "trial_params": null