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 SST2
       type: glue
       config: sst2
       split: validation
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.8107798165137615
+      value: 0.7786697247706422
 ---
 
 <!-- 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. -->
 
 # hBERTv2_new_pretrain_48_KD_sst2
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_48_KD](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_48_KD) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_48_KD](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_48_KD) on the GLUE SST2 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6627
-- Accuracy: 0.8108
+- Loss: 0.4825
+- Accuracy: 0.7787
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 6.0,
+    "eval_accuracy": 0.7786697247706422,
+    "eval_loss": 0.48251545429229736,
+    "eval_runtime": 1.9353,
+    "eval_samples": 872,
+    "eval_samples_per_second": 450.586,
+    "eval_steps_per_second": 3.617,
+    "train_loss": 0.27495639373953024,
+    "train_runtime": 2846.6495,
+    "train_samples": 67349,
+    "train_samples_per_second": 1182.952,
+    "train_steps_per_second": 9.256
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 6.0,
+    "eval_accuracy": 0.7786697247706422,
+    "eval_loss": 0.48251545429229736,
+    "eval_runtime": 1.9353,
+    "eval_samples": 872,
+    "eval_samples_per_second": 450.586,
+    "eval_steps_per_second": 3.617
+}

logs/events.out.tfevents.1686762317.serv-9207.3373475.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97cf1f07738828254827a028e0dad61eb551d30702c7895ab6c97e6dc13364f6
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 6.0,
+    "train_loss": 0.27495639373953024,
+    "train_runtime": 2846.6495,
+    "train_samples": 67349,
+    "train_samples_per_second": 1182.952,
+    "train_steps_per_second": 9.256
+}

trainer_state.json

@@ -0,0 +1,115 @@
+{
+  "best_metric": 0.48251545429229736,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_48_KD_sst2/checkpoint-527",
+  "epoch": 6.0,
+  "global_step": 3162,
+  "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": 0.6251,
+      "step": 527
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7786697247706422,
+      "eval_loss": 0.48251545429229736,
+      "eval_runtime": 1.9025,
+      "eval_samples_per_second": 458.334,
+      "eval_steps_per_second": 3.679,
+      "step": 527
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.2866,
+      "step": 1054
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8073394495412844,
+      "eval_loss": 0.6289057731628418,
+      "eval_runtime": 1.9173,
+      "eval_samples_per_second": 454.804,
+      "eval_steps_per_second": 3.651,
+      "step": 1054
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.2223,
+      "step": 1581
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.805045871559633,
+      "eval_loss": 0.4859866201877594,
+      "eval_runtime": 1.9156,
+      "eval_samples_per_second": 455.205,
+      "eval_steps_per_second": 3.654,
+      "step": 1581
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.1929,
+      "step": 2108
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8107798165137615,
+      "eval_loss": 0.5174009799957275,
+      "eval_runtime": 1.8997,
+      "eval_samples_per_second": 459.01,
+      "eval_steps_per_second": 3.685,
+      "step": 2108
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.1698,
+      "step": 2635
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.805045871559633,
+      "eval_loss": 0.4868491291999817,
+      "eval_runtime": 1.9027,
+      "eval_samples_per_second": 458.286,
+      "eval_steps_per_second": 3.679,
+      "step": 2635
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.1531,
+      "step": 3162
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8107798165137615,
+      "eval_loss": 0.662692129611969,
+      "eval_runtime": 1.8967,
+      "eval_samples_per_second": 459.749,
+      "eval_steps_per_second": 3.691,
+      "step": 3162
+    },
+    {
+      "epoch": 6.0,
+      "step": 3162,
+      "total_flos": 5.913478047360614e+16,
+      "train_loss": 0.27495639373953024,
+      "train_runtime": 2846.6495,
+      "train_samples_per_second": 1182.952,
+      "train_steps_per_second": 9.256
+    }
+  ],
+  "max_steps": 26350,
+  "num_train_epochs": 50,
+  "total_flos": 5.913478047360614e+16,
+  "trial_name": null,
+  "trial_params": null
+}