End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: gokulsrinivasagan/bert_tiny_lda_100_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: bert_tiny_lda_100_v1_qnli
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE QNLI
+      type: glue
+      args: qnli
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.727622185612301
 ---
 
 <!-- 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_tiny_lda_100_v1_qnli
 
-This model is a fine-tuned version of [gokulsrinivasagan/bert_tiny_lda_100_v1](https://huggingface.co/gokulsrinivasagan/bert_tiny_lda_100_v1) on an unknown dataset.
+This model is a fine-tuned version of [gokulsrinivasagan/bert_tiny_lda_100_v1](https://huggingface.co/gokulsrinivasagan/bert_tiny_lda_100_v1) on the GLUE QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7845
-- Accuracy: 0.7293
+- Loss: 0.5783
+- Accuracy: 0.7276
 
 ## Model description
 

all_results.json

@@ -1,15 +1,15 @@
 {
-    "epoch": 6.0,
-    "eval_accuracy": 0.5053999633900788,
-    "eval_loss": 0.6933382749557495,
-    "eval_runtime": 1.6463,
+    "epoch": 9.0,
+    "eval_accuracy": 0.727622185612301,
+    "eval_loss": 0.5783476233482361,
+    "eval_runtime": 1.722,
     "eval_samples": 5463,
-    "eval_samples_per_second": 3318.433,
-    "eval_steps_per_second": 13.364,
-    "total_flos": 1.6480417453111296e+16,
-    "train_loss": 0.6940060127072218,
-    "train_runtime": 346.8815,
+    "eval_samples_per_second": 3172.415,
+    "eval_steps_per_second": 12.776,
+    "total_flos": 2.4720626179666944e+16,
+    "train_loss": 0.43138372529812946,
+    "train_runtime": 534.734,
     "train_samples": 104743,
-    "train_samples_per_second": 15097.809,
-    "train_steps_per_second": 59.098
+    "train_samples_per_second": 9793.934,
+    "train_steps_per_second": 38.337
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 6.0,
-    "eval_accuracy": 0.5053999633900788,
-    "eval_loss": 0.6933382749557495,
-    "eval_runtime": 1.6463,
+    "epoch": 9.0,
+    "eval_accuracy": 0.727622185612301,
+    "eval_loss": 0.5783476233482361,
+    "eval_runtime": 1.722,
     "eval_samples": 5463,
-    "eval_samples_per_second": 3318.433,
-    "eval_steps_per_second": 13.364
+    "eval_samples_per_second": 3172.415,
+    "eval_steps_per_second": 12.776
 }

logs/events.out.tfevents.1733324643.ki-g0008.1208741.23

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe9bf3868369c35a071a9f71e12c72a5cfb8b75e2d8d20fe632a449c8ee00737
+size 411

train_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 6.0,
-    "total_flos": 1.6480417453111296e+16,
-    "train_loss": 0.6940060127072218,
-    "train_runtime": 346.8815,
+    "epoch": 9.0,
+    "total_flos": 2.4720626179666944e+16,
+    "train_loss": 0.43138372529812946,
+    "train_runtime": 534.734,
     "train_samples": 104743,
-    "train_samples_per_second": 15097.809,
-    "train_steps_per_second": 59.098
+    "train_samples_per_second": 9793.934,
+    "train_steps_per_second": 38.337
 }

trainer_state.json

@@ -1,117 +1,165 @@
 {
-  "best_metric": 0.6933382749557495,
-  "best_model_checkpoint": "bert_tiny_lda_100_v1_qnli/checkpoint-410",
-  "epoch": 6.0,
+  "best_metric": 0.5783476233482361,
+  "best_model_checkpoint": "bert_tiny_lda_100_v1_qnli/checkpoint-1640",
+  "epoch": 9.0,
   "eval_steps": 500,
-  "global_step": 2460,
+  "global_step": 3690,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 0.14504414796829224,
-      "learning_rate": 0.00098,
-      "loss": 0.6982,
+      "grad_norm": 1.6202223300933838,
+      "learning_rate": 4.9e-05,
+      "loss": 0.6638,
       "step": 410
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6933382749557495,
-      "eval_runtime": 1.6586,
-      "eval_samples_per_second": 3293.755,
-      "eval_steps_per_second": 13.264,
+      "eval_accuracy": 0.6251144060040271,
+      "eval_loss": 0.6427455544471741,
+      "eval_runtime": 1.7242,
+      "eval_samples_per_second": 3168.366,
+      "eval_steps_per_second": 12.759,
       "step": 410
     },
     {
       "epoch": 2.0,
-      "grad_norm": 0.21841265261173248,
-      "learning_rate": 0.00096,
-      "loss": 0.6931,
+      "grad_norm": 1.5754749774932861,
+      "learning_rate": 4.8e-05,
+      "loss": 0.6373,
       "step": 820
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.6936336159706116,
-      "eval_runtime": 1.6485,
-      "eval_samples_per_second": 3313.97,
-      "eval_steps_per_second": 13.346,
+      "eval_accuracy": 0.6340838367197511,
+      "eval_loss": 0.6345319151878357,
+      "eval_runtime": 1.7183,
+      "eval_samples_per_second": 3179.304,
+      "eval_steps_per_second": 12.803,
       "step": 820
     },
     {
       "epoch": 3.0,
-      "grad_norm": 0.12709185481071472,
-      "learning_rate": 0.00094,
-      "loss": 0.6932,
+      "grad_norm": 3.2697596549987793,
+      "learning_rate": 4.7e-05,
+      "loss": 0.5861,
       "step": 1230
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6933382749557495,
-      "eval_runtime": 1.638,
-      "eval_samples_per_second": 3335.131,
-      "eval_steps_per_second": 13.431,
+      "eval_accuracy": 0.6809445359692476,
+      "eval_loss": 0.5894870162010193,
+      "eval_runtime": 1.746,
+      "eval_samples_per_second": 3128.855,
+      "eval_steps_per_second": 12.6,
       "step": 1230
     },
     {
       "epoch": 4.0,
-      "grad_norm": 0.024518601596355438,
-      "learning_rate": 0.00092,
-      "loss": 0.6932,
+      "grad_norm": 3.8899729251861572,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.4866,
       "step": 1640
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6933382749557495,
-      "eval_runtime": 1.6455,
-      "eval_samples_per_second": 3319.898,
-      "eval_steps_per_second": 13.37,
+      "eval_accuracy": 0.727622185612301,
+      "eval_loss": 0.5783476233482361,
+      "eval_runtime": 1.7484,
+      "eval_samples_per_second": 3124.518,
+      "eval_steps_per_second": 12.583,
       "step": 1640
     },
     {
       "epoch": 5.0,
-      "grad_norm": 0.047646842896938324,
-      "learning_rate": 0.0009000000000000001,
-      "loss": 0.6931,
+      "grad_norm": 4.109302997589111,
+      "learning_rate": 4.5e-05,
+      "loss": 0.4086,
       "step": 2050
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.6934648156166077,
-      "eval_runtime": 1.6442,
-      "eval_samples_per_second": 3322.653,
-      "eval_steps_per_second": 13.381,
+      "eval_accuracy": 0.7212154493867838,
+      "eval_loss": 0.5973373651504517,
+      "eval_runtime": 1.7308,
+      "eval_samples_per_second": 3156.341,
+      "eval_steps_per_second": 12.711,
       "step": 2050
     },
     {
       "epoch": 6.0,
-      "grad_norm": 0.05455218255519867,
-      "learning_rate": 0.00088,
-      "loss": 0.6932,
+      "grad_norm": 3.982114791870117,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.3472,
       "step": 2460
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6933382749557495,
-      "eval_runtime": 1.6522,
-      "eval_samples_per_second": 3306.479,
-      "eval_steps_per_second": 13.315,
+      "eval_accuracy": 0.7375068643602416,
+      "eval_loss": 0.5946151614189148,
+      "eval_runtime": 1.7169,
+      "eval_samples_per_second": 3181.859,
+      "eval_steps_per_second": 12.814,
       "step": 2460
     },
     {
-      "epoch": 6.0,
-      "step": 2460,
-      "total_flos": 1.6480417453111296e+16,
-      "train_loss": 0.6940060127072218,
-      "train_runtime": 346.8815,
-      "train_samples_per_second": 15097.809,
-      "train_steps_per_second": 59.098
+      "epoch": 7.0,
+      "grad_norm": 6.715934753417969,
+      "learning_rate": 4.3e-05,
+      "loss": 0.2941,
+      "step": 2870
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7173714076514736,
+      "eval_loss": 0.7428101301193237,
+      "eval_runtime": 1.7169,
+      "eval_samples_per_second": 3181.973,
+      "eval_steps_per_second": 12.814,
+      "step": 2870
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 2.238851547241211,
+      "learning_rate": 4.2e-05,
+      "loss": 0.2498,
+      "step": 3280
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.738239062786015,
+      "eval_loss": 0.6951647400856018,
+      "eval_runtime": 1.7348,
+      "eval_samples_per_second": 3149.124,
+      "eval_steps_per_second": 12.682,
+      "step": 3280
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 4.454581260681152,
+      "learning_rate": 4.1e-05,
+      "loss": 0.209,
+      "step": 3690
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.729269632070291,
+      "eval_loss": 0.7844876646995544,
+      "eval_runtime": 1.7105,
+      "eval_samples_per_second": 3193.794,
+      "eval_steps_per_second": 12.862,
+      "step": 3690
+    },
+    {
+      "epoch": 9.0,
+      "step": 3690,
+      "total_flos": 2.4720626179666944e+16,
+      "train_loss": 0.43138372529812946,
+      "train_runtime": 534.734,
+      "train_samples_per_second": 9793.934,
+      "train_steps_per_second": 38.337
     }
   ],
   "logging_steps": 1,
@@ -140,7 +188,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.6480417453111296e+16,
+  "total_flos": 2.4720626179666944e+16,
   "train_batch_size": 256,
   "trial_name": null,
   "trial_params": null