End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/distilbert_km_100_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: distilbert_km_100_v1_mnli
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE MNLI
+      type: glue
+      args: mnli
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.6706672091131001
 ---
 
 <!-- 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. -->
 
 # distilbert_km_100_v1_mnli
 
-This model is a fine-tuned version of [Hartunka/distilbert_km_100_v1](https://huggingface.co/Hartunka/distilbert_km_100_v1) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/distilbert_km_100_v1](https://huggingface.co/Hartunka/distilbert_km_100_v1) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0757
-- Accuracy: 0.6649
+- Loss: 0.7617
+- Accuracy: 0.6707
 
 ## Model description
 

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 9.0,
+    "epoch_mm": 9.0,
+    "eval_accuracy": 0.6660213958227204,
+    "eval_accuracy_mm": 0.6706672091131001,
+    "eval_loss": 0.7720940709114075,
+    "eval_loss_mm": 0.7616729736328125,
+    "eval_runtime": 3.9529,
+    "eval_runtime_mm": 4.0535,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 2482.982,
+    "eval_samples_per_second_mm": 2425.574,
+    "eval_steps_per_second": 9.866,
+    "eval_steps_per_second_mm": 9.621,
+    "total_flos": 2.340951304214108e+17,
+    "train_loss": 0.6590785060121618,
+    "train_runtime": 3235.5798,
+    "train_samples": 392702,
+    "train_samples_per_second": 6068.495,
+    "train_steps_per_second": 23.705
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 9.0,
+    "epoch_mm": 9.0,
+    "eval_accuracy": 0.6660213958227204,
+    "eval_accuracy_mm": 0.6706672091131001,
+    "eval_loss": 0.7720940709114075,
+    "eval_loss_mm": 0.7616729736328125,
+    "eval_runtime": 3.9529,
+    "eval_runtime_mm": 4.0535,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 2482.982,
+    "eval_samples_per_second_mm": 2425.574,
+    "eval_steps_per_second": 9.866,
+    "eval_steps_per_second_mm": 9.621
+}

logs/events.out.tfevents.1744912267.s_005_m.2758941.17

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4bc73ade35877c8072ac3791d907e7926c4e4a2db86b09ea1d86ad871264602
+size 686

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.0,
+    "total_flos": 2.340951304214108e+17,
+    "train_loss": 0.6590785060121618,
+    "train_runtime": 3235.5798,
+    "train_samples": 392702,
+    "train_samples_per_second": 6068.495,
+    "train_steps_per_second": 23.705
+}

trainer_state.json

@@ -0,0 +1,196 @@
+{
+  "best_global_step": 6136,
+  "best_metric": 0.7720940709114075,
+  "best_model_checkpoint": "distilbert_km_100_v1_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": 1.0559039115905762,
+      "learning_rate": 4.9e-05,
+      "loss": 0.9885,
+      "step": 1534
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5736118186449313,
+      "eval_loss": 0.9072005748748779,
+      "eval_runtime": 4.1042,
+      "eval_samples_per_second": 2391.449,
+      "eval_steps_per_second": 9.502,
+      "step": 1534
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.8218348026275635,
+      "learning_rate": 4.8e-05,
+      "loss": 0.8707,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6334182373917473,
+      "eval_loss": 0.8225810527801514,
+      "eval_runtime": 3.9519,
+      "eval_samples_per_second": 2483.624,
+      "eval_steps_per_second": 9.869,
+      "step": 3068
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.6593018770217896,
+      "learning_rate": 4.7e-05,
+      "loss": 0.7789,
+      "step": 4602
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6577687213448803,
+      "eval_loss": 0.7826440930366516,
+      "eval_runtime": 3.945,
+      "eval_samples_per_second": 2487.979,
+      "eval_steps_per_second": 9.886,
+      "step": 4602
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.9753878116607666,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.7095,
+      "step": 6136
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6660213958227204,
+      "eval_loss": 0.7720940709114075,
+      "eval_runtime": 3.9402,
+      "eval_samples_per_second": 2490.961,
+      "eval_steps_per_second": 9.898,
+      "step": 6136
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 2.337071657180786,
+      "learning_rate": 4.5e-05,
+      "loss": 0.6453,
+      "step": 7670
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6733571064696893,
+      "eval_loss": 0.7771408557891846,
+      "eval_runtime": 3.9647,
+      "eval_samples_per_second": 2475.607,
+      "eval_steps_per_second": 9.837,
+      "step": 7670
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.6085305213928223,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.5808,
+      "step": 9204
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6682628629648497,
+      "eval_loss": 0.8248651027679443,
+      "eval_runtime": 3.9573,
+      "eval_samples_per_second": 2480.236,
+      "eval_steps_per_second": 9.855,
+      "step": 9204
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.5376288890838623,
+      "learning_rate": 4.3e-05,
+      "loss": 0.5166,
+      "step": 10738
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6683647478349465,
+      "eval_loss": 0.8570982217788696,
+      "eval_runtime": 3.972,
+      "eval_samples_per_second": 2471.072,
+      "eval_steps_per_second": 9.819,
+      "step": 10738
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 2.901345729827881,
+      "learning_rate": 4.2e-05,
+      "loss": 0.4507,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6682628629648497,
+      "eval_loss": 0.9123534560203552,
+      "eval_runtime": 3.9523,
+      "eval_samples_per_second": 2483.358,
+      "eval_steps_per_second": 9.868,
+      "step": 12272
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 3.4146969318389893,
+      "learning_rate": 4.1e-05,
+      "loss": 0.3906,
+      "step": 13806
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6649006622516557,
+      "eval_loss": 1.0756607055664062,
+      "eval_runtime": 3.9483,
+      "eval_samples_per_second": 2485.892,
+      "eval_steps_per_second": 9.878,
+      "step": 13806
+    },
+    {
+      "epoch": 9.0,
+      "step": 13806,
+      "total_flos": 2.340951304214108e+17,
+      "train_loss": 0.6590785060121618,
+      "train_runtime": 3235.5798,
+      "train_samples_per_second": 6068.495,
+      "train_steps_per_second": 23.705
+    }
+  ],
+  "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": 2.340951304214108e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}