End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/tiny_bert_km_50_v2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: tiny_bert_km_50_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: 0.6285923485264506
 ---
 
 <!-- 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_km_50_v2_qnli
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_km_50_v2](https://huggingface.co/Hartunka/tiny_bert_km_50_v2) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_km_50_v2](https://huggingface.co/Hartunka/tiny_bert_km_50_v2) on the GLUE QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0085
-- Accuracy: 0.6169
+- Loss: 0.6464
+- Accuracy: 0.6286
 
 ## Model description
 

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.6285923485264506,
+    "eval_loss": 0.646436333656311,
+    "eval_runtime": 1.5838,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 3449.225,
+    "eval_steps_per_second": 13.89,
+    "total_flos": 2.197388993748173e+16,
+    "train_loss": 0.48549206431319075,
+    "train_runtime": 454.2765,
+    "train_samples": 104743,
+    "train_samples_per_second": 11528.551,
+    "train_steps_per_second": 45.127
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.6285923485264506,
+    "eval_loss": 0.646436333656311,
+    "eval_runtime": 1.5838,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 3449.225,
+    "eval_steps_per_second": 13.89
+}

logs/events.out.tfevents.1745283852.s_005_m.2836625.59

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b912b0f232c64c47f1261bdc8172bf2cde60116f181f680677cc7a42c9fb158
+size 363

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "total_flos": 2.197388993748173e+16,
+    "train_loss": 0.48549206431319075,
+    "train_runtime": 454.2765,
+    "train_samples": 104743,
+    "train_samples_per_second": 11528.551,
+    "train_steps_per_second": 45.127
+}

trainer_state.json

@@ -0,0 +1,180 @@
+{
+  "best_global_step": 1230,
+  "best_metric": 0.646436333656311,
+  "best_model_checkpoint": "tiny_bert_km_50_v2_qnli/checkpoint-1230",
+  "epoch": 8.0,
+  "eval_steps": 500,
+  "global_step": 3280,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.2369039058685303,
+      "learning_rate": 4.9e-05,
+      "loss": 0.6669,
+      "step": 410
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6196229178107268,
+      "eval_loss": 0.648975670337677,
+      "eval_runtime": 1.6228,
+      "eval_samples_per_second": 3366.355,
+      "eval_steps_per_second": 13.557,
+      "step": 410
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.6113338470458984,
+      "learning_rate": 4.8e-05,
+      "loss": 0.6405,
+      "step": 820
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.633534687900421,
+      "eval_loss": 0.6481720805168152,
+      "eval_runtime": 1.571,
+      "eval_samples_per_second": 3477.482,
+      "eval_steps_per_second": 14.004,
+      "step": 820
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 2.5955770015716553,
+      "learning_rate": 4.7e-05,
+      "loss": 0.5995,
+      "step": 1230
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6285923485264506,
+      "eval_loss": 0.646436333656311,
+      "eval_runtime": 1.5557,
+      "eval_samples_per_second": 3511.528,
+      "eval_steps_per_second": 14.141,
+      "step": 1230
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 3.301100969314575,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.5338,
+      "step": 1640
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6223686619073769,
+      "eval_loss": 0.6942920684814453,
+      "eval_runtime": 1.597,
+      "eval_samples_per_second": 3420.686,
+      "eval_steps_per_second": 13.775,
+      "step": 1640
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 3.8189849853515625,
+      "learning_rate": 4.5e-05,
+      "loss": 0.4599,
+      "step": 2050
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6231008603331503,
+      "eval_loss": 0.7389899492263794,
+      "eval_runtime": 1.5849,
+      "eval_samples_per_second": 3446.901,
+      "eval_steps_per_second": 13.881,
+      "step": 2050
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 4.4279351234436035,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.3895,
+      "step": 2460
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6198059674171701,
+      "eval_loss": 0.8551831841468811,
+      "eval_runtime": 1.5541,
+      "eval_samples_per_second": 3515.177,
+      "eval_steps_per_second": 14.156,
+      "step": 2460
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 6.780888080596924,
+      "learning_rate": 4.3e-05,
+      "loss": 0.3245,
+      "step": 2870
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6088229910305692,
+      "eval_loss": 0.9642999768257141,
+      "eval_runtime": 1.6103,
+      "eval_samples_per_second": 3392.576,
+      "eval_steps_per_second": 13.662,
+      "step": 2870
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 4.599832057952881,
+      "learning_rate": 4.2e-05,
+      "loss": 0.2692,
+      "step": 3280
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6168771737140765,
+      "eval_loss": 1.008509874343872,
+      "eval_runtime": 1.5621,
+      "eval_samples_per_second": 3497.225,
+      "eval_steps_per_second": 14.084,
+      "step": 3280
+    },
+    {
+      "epoch": 8.0,
+      "step": 3280,
+      "total_flos": 2.197388993748173e+16,
+      "train_loss": 0.48549206431319075,
+      "train_runtime": 454.2765,
+      "train_samples_per_second": 11528.551,
+      "train_steps_per_second": 45.127
+    }
+  ],
+  "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": 2.197388993748173e+16,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}