End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/bert_base_rand_50_v2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: bert_base_rand_50_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: 0.6518510984540277
 ---
 
 <!-- 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_rand_50_v2_mnli
 
-This model is a fine-tuned version of [Hartunka/bert_base_rand_50_v2](https://huggingface.co/Hartunka/bert_base_rand_50_v2) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/bert_base_rand_50_v2](https://huggingface.co/Hartunka/bert_base_rand_50_v2) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.1619
-- Accuracy: 0.6554
+- Loss: 0.8036
+- Accuracy: 0.6519
 
 ## Model description
 

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 9.0,
+    "epoch_mm": 9.0,
+    "eval_accuracy": 0.6495160468670402,
+    "eval_accuracy_mm": 0.6518510984540277,
+    "eval_loss": 0.8063306212425232,
+    "eval_loss_mm": 0.803560733795166,
+    "eval_runtime": 6.1973,
+    "eval_runtime_mm": 6.3058,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 1583.75,
+    "eval_samples_per_second_mm": 1559.212,
+    "eval_steps_per_second": 6.293,
+    "eval_steps_per_second_mm": 6.185,
+    "total_flos": 4.6496324415879475e+17,
+    "train_loss": 0.6600460896332437,
+    "train_runtime": 5801.0591,
+    "train_samples": 392702,
+    "train_samples_per_second": 3384.744,
+    "train_steps_per_second": 13.222
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 9.0,
+    "epoch_mm": 9.0,
+    "eval_accuracy": 0.6495160468670402,
+    "eval_accuracy_mm": 0.6518510984540277,
+    "eval_loss": 0.8063306212425232,
+    "eval_loss_mm": 0.803560733795166,
+    "eval_runtime": 6.1973,
+    "eval_runtime_mm": 6.3058,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 1583.75,
+    "eval_samples_per_second_mm": 1559.212,
+    "eval_steps_per_second": 6.293,
+    "eval_steps_per_second_mm": 6.185
+}

logs/events.out.tfevents.1745510884.s_005_m.2892304.89

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78a2213a3c6eaa646922a73ff71647b47aace5969756c96871d60ad7dd923a4e
+size 686

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.0,
+    "total_flos": 4.6496324415879475e+17,
+    "train_loss": 0.6600460896332437,
+    "train_runtime": 5801.0591,
+    "train_samples": 392702,
+    "train_samples_per_second": 3384.744,
+    "train_steps_per_second": 13.222
+}

trainer_state.json

@@ -0,0 +1,196 @@
+{
+  "best_global_step": 6136,
+  "best_metric": 0.8063306212425232,
+  "best_model_checkpoint": "bert_base_rand_50_v2_mnli/checkpoint-6136",
+  "epoch": 9.0,
+  "eval_steps": 500,
+  "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": 1.1981600522994995,
+      "learning_rate": 4.9e-05,
+      "loss": 0.98,
+      "step": 1534
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5750382068262863,
+      "eval_loss": 0.905325174331665,
+      "eval_runtime": 6.3505,
+      "eval_samples_per_second": 1545.554,
+      "eval_steps_per_second": 6.141,
+      "step": 1534
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.4905537366867065,
+      "learning_rate": 4.8e-05,
+      "loss": 0.8735,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6041772796739684,
+      "eval_loss": 0.860352635383606,
+      "eval_runtime": 6.2421,
+      "eval_samples_per_second": 1572.4,
+      "eval_steps_per_second": 6.248,
+      "step": 3068
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.687028169631958,
+      "learning_rate": 4.7e-05,
+      "loss": 0.7969,
+      "step": 4602
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6297503820682628,
+      "eval_loss": 0.8222507238388062,
+      "eval_runtime": 6.2368,
+      "eval_samples_per_second": 1573.713,
+      "eval_steps_per_second": 6.253,
+      "step": 4602
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.9236862659454346,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.7229,
+      "step": 6136
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6495160468670402,
+      "eval_loss": 0.8063306212425232,
+      "eval_runtime": 6.3118,
+      "eval_samples_per_second": 1555.032,
+      "eval_steps_per_second": 6.179,
+      "step": 6136
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 2.2754101753234863,
+      "learning_rate": 4.5e-05,
+      "loss": 0.6538,
+      "step": 7670
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6612328069281712,
+      "eval_loss": 0.8120583295822144,
+      "eval_runtime": 6.2567,
+      "eval_samples_per_second": 1568.728,
+      "eval_steps_per_second": 6.233,
+      "step": 7670
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.9715635776519775,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.5838,
+      "step": 9204
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6595007641365257,
+      "eval_loss": 0.8697757124900818,
+      "eval_runtime": 6.2422,
+      "eval_samples_per_second": 1572.365,
+      "eval_steps_per_second": 6.248,
+      "step": 9204
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.3772523403167725,
+      "learning_rate": 4.3e-05,
+      "loss": 0.5103,
+      "step": 10738
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6576668364747835,
+      "eval_loss": 0.9112951755523682,
+      "eval_runtime": 6.2387,
+      "eval_samples_per_second": 1573.242,
+      "eval_steps_per_second": 6.251,
+      "step": 10738
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 3.106327772140503,
+      "learning_rate": 4.2e-05,
+      "loss": 0.4426,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6551197147223637,
+      "eval_loss": 1.000422716140747,
+      "eval_runtime": 6.3494,
+      "eval_samples_per_second": 1545.813,
+      "eval_steps_per_second": 6.142,
+      "step": 12272
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 4.314876556396484,
+      "learning_rate": 4.1e-05,
+      "loss": 0.3765,
+      "step": 13806
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.655425369332654,
+      "eval_loss": 1.1618634462356567,
+      "eval_runtime": 6.244,
+      "eval_samples_per_second": 1571.911,
+      "eval_steps_per_second": 6.246,
+      "step": 13806
+    },
+    {
+      "epoch": 9.0,
+      "step": 13806,
+      "total_flos": 4.6496324415879475e+17,
+      "train_loss": 0.6600460896332437,
+      "train_runtime": 5801.0591,
+      "train_samples_per_second": 3384.744,
+      "train_steps_per_second": 13.222
+    }
+  ],
+  "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": 4.6496324415879475e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}