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 MNLI
       type: glue
       config: mnli
       split: validation_matched
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.3273560876209883
+      value: 0.3522172497965826
 ---
 
 <!-- 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. -->
 
 # hBERTv1_new_pretrain_48_mnli
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_48) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.1024
-- Accuracy: 0.3274
+- Loss: 1.0962
+- Accuracy: 0.3522
 
 ## Model description
 

all_results.json

@@ -0,0 +1,21 @@
+{
+    "epoch": 13.0,
+    "epoch_mm": 13.0,
+    "eval_accuracy": 0.3544574630667346,
+    "eval_accuracy_mm": 0.3522172497965826,
+    "eval_loss": 1.0961673259735107,
+    "eval_loss_mm": 1.0962355136871338,
+    "eval_runtime": 17.3855,
+    "eval_runtime_mm": 17.3353,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 564.55,
+    "eval_samples_per_second_mm": 567.165,
+    "eval_steps_per_second": 4.429,
+    "eval_steps_per_second_mm": 4.442,
+    "train_loss": 1.0993973343007222,
+    "train_runtime": 25989.955,
+    "train_samples": 392702,
+    "train_samples_per_second": 755.488,
+    "train_steps_per_second": 5.902
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 13.0,
+    "epoch_mm": 13.0,
+    "eval_accuracy": 0.3544574630667346,
+    "eval_accuracy_mm": 0.3522172497965826,
+    "eval_loss": 1.0961673259735107,
+    "eval_loss_mm": 1.0962355136871338,
+    "eval_runtime": 17.3855,
+    "eval_runtime_mm": 17.3353,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 564.55,
+    "eval_samples_per_second_mm": 567.165,
+    "eval_steps_per_second": 4.429,
+    "eval_steps_per_second_mm": 4.442
+}

logs/events.out.tfevents.1686079518.serv-9219.1456809.26

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18c58b78aefc58737423682d53c698ac4adf8e98bc82fefee405fdcf85cc60c4
+size 698

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 13.0,
+    "train_loss": 1.0993973343007222,
+    "train_runtime": 25989.955,
+    "train_samples": 392702,
+    "train_samples_per_second": 755.488,
+    "train_steps_per_second": 5.902
+}

trainer_state.json

@@ -0,0 +1,220 @@
+{
+  "best_metric": 1.0961673259735107,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_48_mnli/checkpoint-24544",
+  "epoch": 13.0,
+  "global_step": 39884,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 1.1024,
+      "step": 3068
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.0986531972885132,
+      "eval_runtime": 17.3821,
+      "eval_samples_per_second": 564.66,
+      "eval_steps_per_second": 4.43,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 1.0989,
+      "step": 6136
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.1008286476135254,
+      "eval_runtime": 17.391,
+      "eval_samples_per_second": 564.371,
+      "eval_steps_per_second": 4.428,
+      "step": 6136
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 1.0991,
+      "step": 9204
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.0994316339492798,
+      "eval_runtime": 17.4071,
+      "eval_samples_per_second": 563.851,
+      "eval_steps_per_second": 4.423,
+      "step": 9204
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 1.099,
+      "step": 12272
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.0994316339492798,
+      "eval_runtime": 17.417,
+      "eval_samples_per_second": 563.529,
+      "eval_steps_per_second": 4.421,
+      "step": 12272
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 1.0991,
+      "step": 15340
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.0974847078323364,
+      "eval_runtime": 17.4178,
+      "eval_samples_per_second": 563.503,
+      "eval_steps_per_second": 4.421,
+      "step": 15340
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 1.0995,
+      "step": 18408
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.1003860235214233,
+      "eval_runtime": 17.4611,
+      "eval_samples_per_second": 562.106,
+      "eval_steps_per_second": 4.41,
+      "step": 18408
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 1.0992,
+      "step": 21476
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.0997039079666138,
+      "eval_runtime": 17.3853,
+      "eval_samples_per_second": 564.558,
+      "eval_steps_per_second": 4.429,
+      "step": 21476
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 1.0992,
+      "step": 24544
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.0961673259735107,
+      "eval_runtime": 17.4155,
+      "eval_samples_per_second": 563.579,
+      "eval_steps_per_second": 4.421,
+      "step": 24544
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 3.28e-05,
+      "loss": 1.099,
+      "step": 27612
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.10184907913208,
+      "eval_runtime": 17.4435,
+      "eval_samples_per_second": 562.675,
+      "eval_steps_per_second": 4.414,
+      "step": 27612
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 1.0991,
+      "step": 30680
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.0981477499008179,
+      "eval_runtime": 17.4189,
+      "eval_samples_per_second": 563.467,
+      "eval_steps_per_second": 4.42,
+      "step": 30680
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.1200000000000006e-05,
+      "loss": 1.0992,
+      "step": 33748
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.0980116128921509,
+      "eval_runtime": 17.4007,
+      "eval_samples_per_second": 564.057,
+      "eval_steps_per_second": 4.425,
+      "step": 33748
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.0400000000000004e-05,
+      "loss": 1.0993,
+      "step": 36816
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.096727728843689,
+      "eval_runtime": 17.4358,
+      "eval_samples_per_second": 562.924,
+      "eval_steps_per_second": 4.416,
+      "step": 36816
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 2.96e-05,
+      "loss": 1.0992,
+      "step": 39884
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.1024093627929688,
+      "eval_runtime": 17.4087,
+      "eval_samples_per_second": 563.798,
+      "eval_steps_per_second": 4.423,
+      "step": 39884
+    },
+    {
+      "epoch": 13.0,
+      "step": 39884,
+      "total_flos": 7.560587139109356e+17,
+      "train_loss": 1.0993973343007222,
+      "train_runtime": 25989.955,
+      "train_samples_per_second": 755.488,
+      "train_steps_per_second": 5.902
+    }
+  ],
+  "max_steps": 153400,
+  "num_train_epochs": 50,
+  "total_flos": 7.560587139109356e+17,
+  "trial_name": null,
+  "trial_params": null
+}