End of training

README.md

@@ -1,13 +1,31 @@
 ---
+language:
+- en
 base_model: Hartunka/tiny_bert_rand_100_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - matthews_correlation
 - accuracy
 model-index:
 - name: tiny_bert_rand_100_v1_cola
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE COLA
+      type: glue
+      args: cola
+    metrics:
+    - name: Matthews Correlation
+      type: matthews_correlation
+      value: 0.025208083291660098
+    - name: Accuracy
+      type: accuracy
+      value: 0.6893576383590698
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -15,11 +33,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 # tiny_bert_rand_100_v1_cola
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_rand_100_v1](https://huggingface.co/Hartunka/tiny_bert_rand_100_v1) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_rand_100_v1](https://huggingface.co/Hartunka/tiny_bert_rand_100_v1) on the GLUE COLA dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.8240
-- Matthews Correlation: 0.0899
-- Accuracy: 0.6366
+- Loss: 0.6167
+- Matthews Correlation: 0.0252
+- Accuracy: 0.6894
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.6893576383590698,
+    "eval_loss": 0.6167479157447815,
+    "eval_matthews_correlation": 0.025208083291660098,
+    "eval_runtime": 0.4029,
+    "eval_samples": 1043,
+    "eval_samples_per_second": 2588.686,
+    "eval_steps_per_second": 12.41,
+    "total_flos": 1793902531485696.0,
+    "train_loss": 0.5089688897132874,
+    "train_runtime": 55.5554,
+    "train_samples": 8551,
+    "train_samples_per_second": 7695.926,
+    "train_steps_per_second": 30.6
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.6893576383590698,
+    "eval_loss": 0.6167479157447815,
+    "eval_matthews_correlation": 0.025208083291660098,
+    "eval_runtime": 0.4029,
+    "eval_samples": 1043,
+    "eval_samples_per_second": 2588.686,
+    "eval_steps_per_second": 12.41
+}

logs/events.out.tfevents.1744830135.s_004_m

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc7d703f9539e361968e6d101f772a2a0b1e2a3dcc4a48e500133dec06afeafc
+size 317

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "total_flos": 1793902531485696.0,
+    "train_loss": 0.5089688897132874,
+    "train_runtime": 55.5554,
+    "train_samples": 8551,
+    "train_samples_per_second": 7695.926,
+    "train_steps_per_second": 30.6
+}

trainer_state.json

@@ -0,0 +1,166 @@
+{
+  "best_metric": 0.6167479157447815,
+  "best_model_checkpoint": "tiny_bert_rand_100_v1_cola/checkpoint-102",
+  "epoch": 8.0,
+  "eval_steps": 500,
+  "global_step": 272,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.4837210476398468,
+      "learning_rate": 4.9e-05,
+      "loss": 0.6124,
+      "step": 34
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6912751793861389,
+      "eval_loss": 0.6174571514129639,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 0.3823,
+      "eval_samples_per_second": 2728.439,
+      "eval_steps_per_second": 13.08,
+      "step": 34
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.0265470743179321,
+      "learning_rate": 4.8e-05,
+      "loss": 0.6026,
+      "step": 68
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6912751793861389,
+      "eval_loss": 0.6177305579185486,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 0.4107,
+      "eval_samples_per_second": 2539.808,
+      "eval_steps_per_second": 12.175,
+      "step": 68
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.342545986175537,
+      "learning_rate": 4.7e-05,
+      "loss": 0.5763,
+      "step": 102
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6893576383590698,
+      "eval_loss": 0.6167479157447815,
+      "eval_matthews_correlation": 0.025208083291660098,
+      "eval_runtime": 0.3785,
+      "eval_samples_per_second": 2755.505,
+      "eval_steps_per_second": 13.21,
+      "step": 102
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 2.29557728767395,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.532,
+      "step": 136
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6759347915649414,
+      "eval_loss": 0.6644594669342041,
+      "eval_matthews_correlation": 0.11268021696428915,
+      "eval_runtime": 0.3793,
+      "eval_samples_per_second": 2750.016,
+      "eval_steps_per_second": 13.183,
+      "step": 136
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 2.1529641151428223,
+      "learning_rate": 4.5e-05,
+      "loss": 0.4889,
+      "step": 170
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6605944633483887,
+      "eval_loss": 0.6772807240486145,
+      "eval_matthews_correlation": 0.10937979173366325,
+      "eval_runtime": 0.3842,
+      "eval_samples_per_second": 2714.468,
+      "eval_steps_per_second": 13.013,
+      "step": 170
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.524132490158081,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.4476,
+      "step": 204
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6260786056518555,
+      "eval_loss": 0.716733992099762,
+      "eval_matthews_correlation": 0.06212067397571388,
+      "eval_runtime": 0.3939,
+      "eval_samples_per_second": 2647.9,
+      "eval_steps_per_second": 12.694,
+      "step": 204
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 3.1822924613952637,
+      "learning_rate": 4.3e-05,
+      "loss": 0.4192,
+      "step": 238
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.61649090051651,
+      "eval_loss": 0.7916640043258667,
+      "eval_matthews_correlation": 0.09216190103080887,
+      "eval_runtime": 0.4376,
+      "eval_samples_per_second": 2383.322,
+      "eval_steps_per_second": 11.425,
+      "step": 238
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 3.470726490020752,
+      "learning_rate": 4.2e-05,
+      "loss": 0.3928,
+      "step": 272
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6366251111030579,
+      "eval_loss": 0.8240132331848145,
+      "eval_matthews_correlation": 0.08989469784450668,
+      "eval_runtime": 0.3797,
+      "eval_samples_per_second": 2747.14,
+      "eval_steps_per_second": 13.169,
+      "step": 272
+    },
+    {
+      "epoch": 8.0,
+      "step": 272,
+      "total_flos": 1793902531485696.0,
+      "train_loss": 0.5089688897132874,
+      "train_runtime": 55.5554,
+      "train_samples_per_second": 7695.926,
+      "train_steps_per_second": 30.6
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 1700,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 50,
+  "save_steps": 500,
+  "total_flos": 1793902531485696.0,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}