End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/tiny_bert_rand_100_v2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: tiny_bert_rand_100_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: 0.6182872253864931
 ---
 
 <!-- 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_rand_100_v2_mnli
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_rand_100_v2](https://huggingface.co/Hartunka/tiny_bert_rand_100_v2) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_rand_100_v2](https://huggingface.co/Hartunka/tiny_bert_rand_100_v2) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9633
-- Accuracy: 0.6374
+- Loss: 0.8565
+- Accuracy: 0.6183
 
 ## Model description
 

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.6264900662251656,
+    "eval_accuracy_mm": 0.6182872253864931,
+    "eval_loss": 0.8414934277534485,
+    "eval_loss_mm": 0.8565420508384705,
+    "eval_runtime": 2.8366,
+    "eval_runtime_mm": 2.9049,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 3460.07,
+    "eval_samples_per_second_mm": 3384.644,
+    "eval_steps_per_second": 13.749,
+    "eval_steps_per_second_mm": 13.426,
+    "total_flos": 1.029836100834816e+17,
+    "train_loss": 0.7477874079459338,
+    "train_runtime": 2075.8939,
+    "train_samples": 392702,
+    "train_samples_per_second": 9458.624,
+    "train_steps_per_second": 36.948
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.6264900662251656,
+    "eval_accuracy_mm": 0.6182872253864931,
+    "eval_loss": 0.8414934277534485,
+    "eval_loss_mm": 0.8565420508384705,
+    "eval_runtime": 2.8366,
+    "eval_runtime_mm": 2.9049,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 3460.07,
+    "eval_samples_per_second_mm": 3384.644,
+    "eval_steps_per_second": 13.749,
+    "eval_steps_per_second_mm": 13.426
+}

logs/events.out.tfevents.1745325470.s_005_m.2850172.71

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f72a58dc0140621ee8cac5480604d1947853fc3e81f98c5aff3c003907af4836
+size 686

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 1.029836100834816e+17,
+    "train_loss": 0.7477874079459338,
+    "train_runtime": 2075.8939,
+    "train_samples": 392702,
+    "train_samples_per_second": 9458.624,
+    "train_steps_per_second": 36.948
+}

trainer_state.json

@@ -0,0 +1,212 @@
+{
+  "best_global_step": 7670,
+  "best_metric": 0.8414934277534485,
+  "best_model_checkpoint": "tiny_bert_rand_100_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": 1.0682040452957153,
+      "learning_rate": 4.9e-05,
+      "loss": 0.9887,
+      "step": 1534
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5558838512480897,
+      "eval_loss": 0.9264980554580688,
+      "eval_runtime": 2.8988,
+      "eval_samples_per_second": 3385.932,
+      "eval_steps_per_second": 13.454,
+      "step": 1534
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.6633317470550537,
+      "learning_rate": 4.8e-05,
+      "loss": 0.9027,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5872643912379012,
+      "eval_loss": 0.8873052000999451,
+      "eval_runtime": 2.7809,
+      "eval_samples_per_second": 3529.481,
+      "eval_steps_per_second": 14.024,
+      "step": 3068
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.4859426021575928,
+      "learning_rate": 4.7e-05,
+      "loss": 0.8512,
+      "step": 4602
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6098828323993887,
+      "eval_loss": 0.8563730716705322,
+      "eval_runtime": 2.8369,
+      "eval_samples_per_second": 3459.732,
+      "eval_steps_per_second": 13.747,
+      "step": 4602
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.5856523513793945,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.8058,
+      "step": 6136
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6228222109016811,
+      "eval_loss": 0.8466203808784485,
+      "eval_runtime": 2.7906,
+      "eval_samples_per_second": 3517.205,
+      "eval_steps_per_second": 13.976,
+      "step": 6136
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.5864737033843994,
+      "learning_rate": 4.5e-05,
+      "loss": 0.7615,
+      "step": 7670
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6264900662251656,
+      "eval_loss": 0.8414934277534485,
+      "eval_runtime": 2.8505,
+      "eval_samples_per_second": 3443.289,
+      "eval_steps_per_second": 13.682,
+      "step": 7670
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.2457175254821777,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.7179,
+      "step": 9204
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.634029546612328,
+      "eval_loss": 0.8559539914131165,
+      "eval_runtime": 2.8068,
+      "eval_samples_per_second": 3496.81,
+      "eval_steps_per_second": 13.895,
+      "step": 9204
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.9325635433197021,
+      "learning_rate": 4.3e-05,
+      "loss": 0.6747,
+      "step": 10738
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6418746816097809,
+      "eval_loss": 0.8655825853347778,
+      "eval_runtime": 2.8132,
+      "eval_samples_per_second": 3488.866,
+      "eval_steps_per_second": 13.863,
+      "step": 10738
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 2.247413396835327,
+      "learning_rate": 4.2e-05,
+      "loss": 0.6329,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6325012735608763,
+      "eval_loss": 0.9130884408950806,
+      "eval_runtime": 2.8091,
+      "eval_samples_per_second": 3493.943,
+      "eval_steps_per_second": 13.883,
+      "step": 12272
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 2.5739073753356934,
+      "learning_rate": 4.1e-05,
+      "loss": 0.5912,
+      "step": 13806
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6346408558329089,
+      "eval_loss": 0.9365619421005249,
+      "eval_runtime": 2.8119,
+      "eval_samples_per_second": 3490.513,
+      "eval_steps_per_second": 13.87,
+      "step": 13806
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 3.034921407699585,
+      "learning_rate": 4e-05,
+      "loss": 0.5514,
+      "step": 15340
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.6373917473255222,
+      "eval_loss": 0.9632840156555176,
+      "eval_runtime": 2.8001,
+      "eval_samples_per_second": 3505.223,
+      "eval_steps_per_second": 13.928,
+      "step": 15340
+    },
+    {
+      "epoch": 10.0,
+      "step": 15340,
+      "total_flos": 1.029836100834816e+17,
+      "train_loss": 0.7477874079459338,
+      "train_runtime": 2075.8939,
+      "train_samples_per_second": 9458.624,
+      "train_steps_per_second": 36.948
+    }
+  ],
+  "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
+}