End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/distilbert_km_100_v2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: distilbert_km_100_v2_qnli
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE QNLI
+      type: glue
+      args: qnli
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.6370126304228446
 ---
 
 <!-- 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_v2_qnli
 
-This model is a fine-tuned version of [Hartunka/distilbert_km_100_v2](https://huggingface.co/Hartunka/distilbert_km_100_v2) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/distilbert_km_100_v2](https://huggingface.co/Hartunka/distilbert_km_100_v2) on the GLUE QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.3313
-- Accuracy: 0.6125
+- Loss: 0.6401
+- Accuracy: 0.6370
 
 ## Model description
 

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.6370126304228446,
+    "eval_loss": 0.6400607228279114,
+    "eval_runtime": 2.2258,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 2454.368,
+    "eval_steps_per_second": 9.884,
+    "total_flos": 4.856261458098893e+16,
+    "train_loss": 0.436550297222071,
+    "train_runtime": 691.8278,
+    "train_samples": 104743,
+    "train_samples_per_second": 7570.02,
+    "train_steps_per_second": 29.632
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.6370126304228446,
+    "eval_loss": 0.6400607228279114,
+    "eval_runtime": 2.2258,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 2454.368,
+    "eval_steps_per_second": 9.884
+}

logs/events.out.tfevents.1745339129.s_005_m.2850345.77

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6053bd9b4ea940d6353cf99394e364a3cf03c3dec375e151d41e8b0c577f846
+size 363

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 7.0,
+    "total_flos": 4.856261458098893e+16,
+    "train_loss": 0.436550297222071,
+    "train_runtime": 691.8278,
+    "train_samples": 104743,
+    "train_samples_per_second": 7570.02,
+    "train_steps_per_second": 29.632
+}

trainer_state.json

@@ -0,0 +1,164 @@
+{
+  "best_global_step": 820,
+  "best_metric": 0.6400607228279114,
+  "best_model_checkpoint": "distilbert_km_100_v2_qnli/checkpoint-820",
+  "epoch": 7.0,
+  "eval_steps": 500,
+  "global_step": 2870,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.5863474607467651,
+      "learning_rate": 4.9e-05,
+      "loss": 0.6675,
+      "step": 410
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6234669595460369,
+      "eval_loss": 0.6457479596138,
+      "eval_runtime": 2.2975,
+      "eval_samples_per_second": 2377.789,
+      "eval_steps_per_second": 9.576,
+      "step": 410
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.7952786684036255,
+      "learning_rate": 4.8e-05,
+      "loss": 0.6288,
+      "step": 820
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6370126304228446,
+      "eval_loss": 0.6400607228279114,
+      "eval_runtime": 2.2273,
+      "eval_samples_per_second": 2452.754,
+      "eval_steps_per_second": 9.877,
+      "step": 820
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 3.2203218936920166,
+      "learning_rate": 4.7e-05,
+      "loss": 0.559,
+      "step": 1230
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6344499359326378,
+      "eval_loss": 0.6570987701416016,
+      "eval_runtime": 2.206,
+      "eval_samples_per_second": 2476.481,
+      "eval_steps_per_second": 9.973,
+      "step": 1230
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 5.592960834503174,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.4497,
+      "step": 1640
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6267618524620172,
+      "eval_loss": 0.7618572115898132,
+      "eval_runtime": 2.221,
+      "eval_samples_per_second": 2459.649,
+      "eval_steps_per_second": 9.905,
+      "step": 1640
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 5.256136894226074,
+      "learning_rate": 4.5e-05,
+      "loss": 0.3337,
+      "step": 2050
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6254805052169138,
+      "eval_loss": 0.8714501857757568,
+      "eval_runtime": 2.2299,
+      "eval_samples_per_second": 2449.866,
+      "eval_steps_per_second": 9.866,
+      "step": 2050
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 6.902203559875488,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.2409,
+      "step": 2460
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6209042650558302,
+      "eval_loss": 1.0884989500045776,
+      "eval_runtime": 2.2353,
+      "eval_samples_per_second": 2443.966,
+      "eval_steps_per_second": 9.842,
+      "step": 2460
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 10.041999816894531,
+      "learning_rate": 4.3e-05,
+      "loss": 0.1762,
+      "step": 2870
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6124839831594362,
+      "eval_loss": 1.3313391208648682,
+      "eval_runtime": 2.2129,
+      "eval_samples_per_second": 2468.679,
+      "eval_steps_per_second": 9.942,
+      "step": 2870
+    },
+    {
+      "epoch": 7.0,
+      "step": 2870,
+      "total_flos": 4.856261458098893e+16,
+      "train_loss": 0.436550297222071,
+      "train_runtime": 691.8278,
+      "train_samples_per_second": 7570.02,
+      "train_steps_per_second": 29.632
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 20500,
+  "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": 4.856261458098893e+16,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}