End of training

README.md

@@ -1,12 +1,27 @@
 ---
+language:
+- en
 base_model: Hartunka/tiny_bert_rand_100_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: tiny_bert_rand_100_v1_rte
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE RTE
+      type: glue
+      args: rte
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.5631768953068592
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -14,10 +29,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # tiny_bert_rand_100_v1_rte
 
-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 RTE dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.1948
-- Accuracy: 0.5054
+- Loss: 0.6844
+- Accuracy: 0.5632
 
 ## Model description
 

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.5631768953068592,
+    "eval_loss": 0.6843975782394409,
+    "eval_runtime": 0.1375,
+    "eval_samples": 277,
+    "eval_samples_per_second": 2014.59,
+    "eval_steps_per_second": 14.546,
+    "total_flos": 457076966492160.0,
+    "train_loss": 0.5635066100529262,
+    "train_runtime": 17.6204,
+    "train_samples": 2490,
+    "train_samples_per_second": 7065.673,
+    "train_steps_per_second": 28.376
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.5631768953068592,
+    "eval_loss": 0.6843975782394409,
+    "eval_runtime": 0.1375,
+    "eval_samples": 277,
+    "eval_samples_per_second": 2014.59,
+    "eval_steps_per_second": 14.546
+}

logs/events.out.tfevents.1744832859.s_004_m

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4bbafbe429a9d4e06089acc0114d35fa9685762f94df95d69fa3cf2967e32624
+size 40

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 7.0,
+    "total_flos": 457076966492160.0,
+    "train_loss": 0.5635066100529262,
+    "train_runtime": 17.6204,
+    "train_samples": 2490,
+    "train_samples_per_second": 7065.673,
+    "train_steps_per_second": 28.376
+}

trainer_state.json

@@ -0,0 +1,142 @@
+{
+  "best_metric": 0.6843975782394409,
+  "best_model_checkpoint": "tiny_bert_rand_100_v1_rte/checkpoint-20",
+  "epoch": 7.0,
+  "eval_steps": 500,
+  "global_step": 70,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.3619806468486786,
+      "learning_rate": 4.9e-05,
+      "loss": 0.7003,
+      "step": 10
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5306859205776173,
+      "eval_loss": 0.6884024739265442,
+      "eval_runtime": 0.1156,
+      "eval_samples_per_second": 2395.52,
+      "eval_steps_per_second": 17.296,
+      "step": 10
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.4103142321109772,
+      "learning_rate": 4.8e-05,
+      "loss": 0.6895,
+      "step": 20
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5631768953068592,
+      "eval_loss": 0.6843975782394409,
+      "eval_runtime": 0.111,
+      "eval_samples_per_second": 2495.505,
+      "eval_steps_per_second": 18.018,
+      "step": 20
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.47117242217063904,
+      "learning_rate": 4.7e-05,
+      "loss": 0.6719,
+      "step": 30
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5595667870036101,
+      "eval_loss": 0.6904226541519165,
+      "eval_runtime": 0.1129,
+      "eval_samples_per_second": 2452.783,
+      "eval_steps_per_second": 17.71,
+      "step": 30
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.0662132501602173,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.618,
+      "step": 40
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5667870036101083,
+      "eval_loss": 0.7529085278511047,
+      "eval_runtime": 0.1415,
+      "eval_samples_per_second": 1958.082,
+      "eval_steps_per_second": 14.138,
+      "step": 40
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 2.153291940689087,
+      "learning_rate": 4.5e-05,
+      "loss": 0.5321,
+      "step": 50
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5487364620938628,
+      "eval_loss": 0.8656222224235535,
+      "eval_runtime": 0.1082,
+      "eval_samples_per_second": 2560.377,
+      "eval_steps_per_second": 18.486,
+      "step": 50
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.455439567565918,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.4247,
+      "step": 60
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5126353790613718,
+      "eval_loss": 1.024060606956482,
+      "eval_runtime": 0.1095,
+      "eval_samples_per_second": 2530.575,
+      "eval_steps_per_second": 18.271,
+      "step": 60
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.5471134185791016,
+      "learning_rate": 4.3e-05,
+      "loss": 0.3081,
+      "step": 70
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5054151624548736,
+      "eval_loss": 1.194822907447815,
+      "eval_runtime": 0.109,
+      "eval_samples_per_second": 2540.829,
+      "eval_steps_per_second": 18.345,
+      "step": 70
+    },
+    {
+      "epoch": 7.0,
+      "step": 70,
+      "total_flos": 457076966492160.0,
+      "train_loss": 0.5635066100529262,
+      "train_runtime": 17.6204,
+      "train_samples_per_second": 7065.673,
+      "train_steps_per_second": 28.376
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 500,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 50,
+  "save_steps": 500,
+  "total_flos": 457076966492160.0,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}