End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/tiny_bert_km_20_v2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: tiny_bert_km_20_v2_mnli
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE MNLI
+      type: glue
+      args: mnli
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.6295768917819365
 ---
 
 <!-- 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. -->
 
 # tiny_bert_km_20_v2_mnli
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_km_20_v2](https://huggingface.co/Hartunka/tiny_bert_km_20_v2) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_km_20_v2](https://huggingface.co/Hartunka/tiny_bert_km_20_v2) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9313
-- Accuracy: 0.6425
+- Loss: 0.8293
+- Accuracy: 0.6296
 
 ## Model description
 

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.6300560366785533,
+    "eval_accuracy_mm": 0.6295768917819365,
+    "eval_loss": 0.8323174118995667,
+    "eval_loss_mm": 0.8292797207832336,
+    "eval_runtime": 2.8453,
+    "eval_runtime_mm": 2.8402,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 3449.501,
+    "eval_samples_per_second_mm": 3461.682,
+    "eval_steps_per_second": 13.707,
+    "eval_steps_per_second_mm": 13.731,
+    "total_flos": 1.029836100834816e+17,
+    "train_loss": 0.7435631265851633,
+    "train_runtime": 2079.7547,
+    "train_samples": 392702,
+    "train_samples_per_second": 9441.065,
+    "train_steps_per_second": 36.879
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.6300560366785533,
+    "eval_accuracy_mm": 0.6295768917819365,
+    "eval_loss": 0.8323174118995667,
+    "eval_loss_mm": 0.8292797207832336,
+    "eval_runtime": 2.8453,
+    "eval_runtime_mm": 2.8402,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 3449.501,
+    "eval_samples_per_second_mm": 3461.682,
+    "eval_steps_per_second": 13.707,
+    "eval_steps_per_second_mm": 13.731
+}

logs/events.out.tfevents.1745283239.s_005_m.2836625.53

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55e65898ba8f84ba60df1c9ed21c39c064b31199af51e58f5bef7d10c925d99a
+size 686

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 1.029836100834816e+17,
+    "train_loss": 0.7435631265851633,
+    "train_runtime": 2079.7547,
+    "train_samples": 392702,
+    "train_samples_per_second": 9441.065,
+    "train_steps_per_second": 36.879
+}

trainer_state.json

@@ -0,0 +1,212 @@
+{
+  "best_global_step": 7670,
+  "best_metric": 0.8323174118995667,
+  "best_model_checkpoint": "tiny_bert_km_20_v2_mnli/checkpoint-7670",
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 15340,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.1402794122695923,
+      "learning_rate": 4.9e-05,
+      "loss": 0.9993,
+      "step": 1534
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5527254202750892,
+      "eval_loss": 0.9375848770141602,
+      "eval_runtime": 2.866,
+      "eval_samples_per_second": 3424.627,
+      "eval_steps_per_second": 13.608,
+      "step": 1534
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.3853436708450317,
+      "learning_rate": 4.8e-05,
+      "loss": 0.9067,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5892002037697401,
+      "eval_loss": 0.8823896646499634,
+      "eval_runtime": 2.7936,
+      "eval_samples_per_second": 3513.424,
+      "eval_steps_per_second": 13.961,
+      "step": 3068
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.4370675086975098,
+      "learning_rate": 4.7e-05,
+      "loss": 0.8527,
+      "step": 4602
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6076413652572593,
+      "eval_loss": 0.8542468547821045,
+      "eval_runtime": 2.8078,
+      "eval_samples_per_second": 3495.574,
+      "eval_steps_per_second": 13.89,
+      "step": 4602
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.9635460376739502,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.8046,
+      "step": 6136
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.621395822720326,
+      "eval_loss": 0.8475756645202637,
+      "eval_runtime": 2.7922,
+      "eval_samples_per_second": 3515.198,
+      "eval_steps_per_second": 13.968,
+      "step": 6136
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.8043975830078125,
+      "learning_rate": 4.5e-05,
+      "loss": 0.7574,
+      "step": 7670
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6300560366785533,
+      "eval_loss": 0.8323174118995667,
+      "eval_runtime": 2.7823,
+      "eval_samples_per_second": 3527.703,
+      "eval_steps_per_second": 14.017,
+      "step": 7670
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.153754711151123,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.7096,
+      "step": 9204
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6352521650534896,
+      "eval_loss": 0.8395307064056396,
+      "eval_runtime": 2.8595,
+      "eval_samples_per_second": 3432.38,
+      "eval_steps_per_second": 13.639,
+      "step": 9204
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.124469041824341,
+      "learning_rate": 4.3e-05,
+      "loss": 0.666,
+      "step": 10738
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6446255731023943,
+      "eval_loss": 0.8471359014511108,
+      "eval_runtime": 2.8009,
+      "eval_samples_per_second": 3504.256,
+      "eval_steps_per_second": 13.924,
+      "step": 10738
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 2.1384940147399902,
+      "learning_rate": 4.2e-05,
+      "loss": 0.6211,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6423841059602649,
+      "eval_loss": 0.876538097858429,
+      "eval_runtime": 2.7815,
+      "eval_samples_per_second": 3528.638,
+      "eval_steps_per_second": 14.021,
+      "step": 12272
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 2.8161027431488037,
+      "learning_rate": 4.1e-05,
+      "loss": 0.5794,
+      "step": 13806
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6421803362200713,
+      "eval_loss": 0.9186814427375793,
+      "eval_runtime": 2.8072,
+      "eval_samples_per_second": 3496.388,
+      "eval_steps_per_second": 13.893,
+      "step": 13806
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 3.022850751876831,
+      "learning_rate": 4e-05,
+      "loss": 0.539,
+      "step": 15340
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.6424859908303617,
+      "eval_loss": 0.9312591552734375,
+      "eval_runtime": 2.8099,
+      "eval_samples_per_second": 3493.058,
+      "eval_steps_per_second": 13.88,
+      "step": 15340
+    },
+    {
+      "epoch": 10.0,
+      "step": 15340,
+      "total_flos": 1.029836100834816e+17,
+      "train_loss": 0.7435631265851633,
+      "train_runtime": 2079.7547,
+      "train_samples_per_second": 9441.065,
+      "train_steps_per_second": 36.879
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 76700,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 50,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 5
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.029836100834816e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}