End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/tiny_bert_rand_20_v2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: tiny_bert_rand_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: 0.6246948738812043
 ---
 
 <!-- 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_20_v2_mnli
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_rand_20_v2](https://huggingface.co/Hartunka/tiny_bert_rand_20_v2) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_rand_20_v2](https://huggingface.co/Hartunka/tiny_bert_rand_20_v2) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9863
-- Accuracy: 0.6298
+- Loss: 0.8598
+- Accuracy: 0.6247
 
 ## Model description
 

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.6182373917473255,
+    "eval_accuracy_mm": 0.6246948738812043,
+    "eval_loss": 0.8496004343032837,
+    "eval_loss_mm": 0.859828531742096,
+    "eval_runtime": 2.821,
+    "eval_runtime_mm": 2.8543,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 3479.207,
+    "eval_samples_per_second_mm": 3444.656,
+    "eval_steps_per_second": 13.825,
+    "eval_steps_per_second_mm": 13.664,
+    "total_flos": 1.029836100834816e+17,
+    "train_loss": 0.7532363434036832,
+    "train_runtime": 2078.1752,
+    "train_samples": 392702,
+    "train_samples_per_second": 9448.241,
+    "train_steps_per_second": 36.907
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.6182373917473255,
+    "eval_accuracy_mm": 0.6246948738812043,
+    "eval_loss": 0.8496004343032837,
+    "eval_loss_mm": 0.859828531742096,
+    "eval_runtime": 2.821,
+    "eval_runtime_mm": 2.8543,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 3479.207,
+    "eval_samples_per_second_mm": 3444.656,
+    "eval_steps_per_second": 13.825,
+    "eval_steps_per_second_mm": 13.664
+}

logs/events.out.tfevents.1745316352.s_005_m.2850172.35

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:129bf6045888b5824f76ff9cc01531c54f0ea9cffbd56d63238f19f86c7ae24a
+size 686

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 1.029836100834816e+17,
+    "train_loss": 0.7532363434036832,
+    "train_runtime": 2078.1752,
+    "train_samples": 392702,
+    "train_samples_per_second": 9448.241,
+    "train_steps_per_second": 36.907
+}

trainer_state.json

@@ -0,0 +1,212 @@
+{
+  "best_global_step": 7670,
+  "best_metric": 0.8496004343032837,
+  "best_model_checkpoint": "tiny_bert_rand_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": 1.1339234113693237,
+      "learning_rate": 4.9e-05,
+      "loss": 0.9926,
+      "step": 1534
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5506877228731534,
+      "eval_loss": 0.9313808679580688,
+      "eval_runtime": 3.0,
+      "eval_samples_per_second": 3271.614,
+      "eval_steps_per_second": 13.0,
+      "step": 1534
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.3619678020477295,
+      "learning_rate": 4.8e-05,
+      "loss": 0.9032,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5823739174732552,
+      "eval_loss": 0.8910712599754333,
+      "eval_runtime": 2.8007,
+      "eval_samples_per_second": 3504.45,
+      "eval_steps_per_second": 13.925,
+      "step": 3068
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.7281922101974487,
+      "learning_rate": 4.7e-05,
+      "loss": 0.8523,
+      "step": 4602
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6031584309730005,
+      "eval_loss": 0.8651266098022461,
+      "eval_runtime": 2.8129,
+      "eval_samples_per_second": 3489.278,
+      "eval_steps_per_second": 13.865,
+      "step": 4602
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.9255449771881104,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.8089,
+      "step": 6136
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6095771777890984,
+      "eval_loss": 0.8570564985275269,
+      "eval_runtime": 2.9455,
+      "eval_samples_per_second": 3332.204,
+      "eval_steps_per_second": 13.241,
+      "step": 6136
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.6890971660614014,
+      "learning_rate": 4.5e-05,
+      "loss": 0.7665,
+      "step": 7670
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6182373917473255,
+      "eval_loss": 0.8496004343032837,
+      "eval_runtime": 2.7913,
+      "eval_samples_per_second": 3516.255,
+      "eval_steps_per_second": 13.972,
+      "step": 7670
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.9976463317871094,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.7234,
+      "step": 9204
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.624350483953133,
+      "eval_loss": 0.8671989440917969,
+      "eval_runtime": 2.8098,
+      "eval_samples_per_second": 3493.092,
+      "eval_steps_per_second": 13.88,
+      "step": 9204
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.1107654571533203,
+      "learning_rate": 4.3e-05,
+      "loss": 0.6827,
+      "step": 10738
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6287315333672949,
+      "eval_loss": 0.8687392473220825,
+      "eval_runtime": 2.9761,
+      "eval_samples_per_second": 3297.938,
+      "eval_steps_per_second": 13.104,
+      "step": 10738
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 2.3106932640075684,
+      "learning_rate": 4.2e-05,
+      "loss": 0.6402,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6142638818135507,
+      "eval_loss": 0.9203874468803406,
+      "eval_runtime": 2.8068,
+      "eval_samples_per_second": 3496.895,
+      "eval_steps_per_second": 13.895,
+      "step": 12272
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 3.1391849517822266,
+      "learning_rate": 4.1e-05,
+      "loss": 0.6004,
+      "step": 13806
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6229240957717779,
+      "eval_loss": 0.9506785869598389,
+      "eval_runtime": 2.9969,
+      "eval_samples_per_second": 3275.072,
+      "eval_steps_per_second": 13.014,
+      "step": 13806
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 3.063190221786499,
+      "learning_rate": 4e-05,
+      "loss": 0.5621,
+      "step": 15340
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.6297503820682628,
+      "eval_loss": 0.9863426685333252,
+      "eval_runtime": 2.7969,
+      "eval_samples_per_second": 3509.228,
+      "eval_steps_per_second": 13.944,
+      "step": 15340
+    },
+    {
+      "epoch": 10.0,
+      "step": 15340,
+      "total_flos": 1.029836100834816e+17,
+      "train_loss": 0.7532363434036832,
+      "train_runtime": 2078.1752,
+      "train_samples_per_second": 9448.241,
+      "train_steps_per_second": 36.907
+    }
+  ],
+  "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
+}