End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 license: apache-2.0
 tags:
 - generated_from_trainer
@@ -13,7 +15,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE MNLI
       type: glue
       config: mnli
       split: validation_matched
@@ -21,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.5453897096281203
+      value: 0.5431244914564687
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -29,10 +31,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # distilbert_sa_GLUE_Experiment_logit_kd_mnli_96
 
-This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the glue dataset.
+This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5495
-- Accuracy: 0.5454
+- Loss: 0.5438
+- Accuracy: 0.5431
 
 ## Model description
 

all_results.json

@@ -0,0 +1,21 @@
+{
+    "epoch": 13.0,
+    "epoch_mm": 13.0,
+    "eval_accuracy": 0.5410086602139582,
+    "eval_accuracy_mm": 0.5431244914564687,
+    "eval_loss": 0.5428069829940796,
+    "eval_loss_mm": 0.5437836647033691,
+    "eval_runtime": 12.8939,
+    "eval_runtime_mm": 12.7878,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 761.211,
+    "eval_samples_per_second_mm": 768.855,
+    "eval_steps_per_second": 3.025,
+    "eval_steps_per_second_mm": 3.05,
+    "train_loss": 0.5403076514665969,
+    "train_runtime": 8354.8792,
+    "train_samples": 392702,
+    "train_samples_per_second": 2350.136,
+    "train_steps_per_second": 9.18
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 13.0,
+    "epoch_mm": 13.0,
+    "eval_accuracy": 0.5410086602139582,
+    "eval_accuracy_mm": 0.5431244914564687,
+    "eval_loss": 0.5428069829940796,
+    "eval_loss_mm": 0.5437836647033691,
+    "eval_runtime": 12.8939,
+    "eval_runtime_mm": 12.7878,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 761.211,
+    "eval_samples_per_second_mm": 768.855,
+    "eval_steps_per_second": 3.025,
+    "eval_steps_per_second_mm": 3.05
+}

logs/events.out.tfevents.1675056314.serv-3328.2086075.26

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:431689069113dc9ddec93faa360619da9e467a4e3b9c8e63304cff64bc6b0ce9
+size 698

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 13.0,
+    "train_loss": 0.5403076514665969,
+    "train_runtime": 8354.8792,
+    "train_samples": 392702,
+    "train_samples_per_second": 2350.136,
+    "train_steps_per_second": 9.18
+}

trainer_state.json

@@ -0,0 +1,220 @@
+{
+  "best_metric": 0.5428069829940796,
+  "best_model_checkpoint": "distilbert_sa_GLUE_Experiment_logit_kd_mnli_96/checkpoint-12272",
+  "epoch": 13.0,
+  "global_step": 19942,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.9e-05,
+      "loss": 0.6023,
+      "step": 1534
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.4959755476311768,
+      "eval_loss": 0.5717693567276001,
+      "eval_runtime": 12.8396,
+      "eval_samples_per_second": 764.43,
+      "eval_steps_per_second": 3.037,
+      "step": 1534
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.5673,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5183902190524707,
+      "eval_loss": 0.5547171235084534,
+      "eval_runtime": 12.9668,
+      "eval_samples_per_second": 756.932,
+      "eval_steps_per_second": 3.008,
+      "step": 3068
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.5555,
+      "step": 4602
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5277636271013755,
+      "eval_loss": 0.5505178570747375,
+      "eval_runtime": 12.8585,
+      "eval_samples_per_second": 763.311,
+      "eval_steps_per_second": 3.033,
+      "step": 4602
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.5481,
+      "step": 6136
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5380539989811512,
+      "eval_loss": 0.5466012954711914,
+      "eval_runtime": 12.9073,
+      "eval_samples_per_second": 760.421,
+      "eval_steps_per_second": 3.022,
+      "step": 6136
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.5426,
+      "step": 7670
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5402954661232807,
+      "eval_loss": 0.5454197525978088,
+      "eval_runtime": 12.9677,
+      "eval_samples_per_second": 756.879,
+      "eval_steps_per_second": 3.007,
+      "step": 7670
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.5382,
+      "step": 9204
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5354049923586347,
+      "eval_loss": 0.5453567504882812,
+      "eval_runtime": 12.9712,
+      "eval_samples_per_second": 756.677,
+      "eval_steps_per_second": 3.007,
+      "step": 9204
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.30006518904824e-05,
+      "loss": 0.5341,
+      "step": 10738
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5343861436576668,
+      "eval_loss": 0.5452288389205933,
+      "eval_runtime": 13.0961,
+      "eval_samples_per_second": 749.461,
+      "eval_steps_per_second": 2.978,
+      "step": 10738
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.20013037809648e-05,
+      "loss": 0.5308,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5410086602139582,
+      "eval_loss": 0.5428069829940796,
+      "eval_runtime": 13.0005,
+      "eval_samples_per_second": 754.971,
+      "eval_steps_per_second": 3.0,
+      "step": 12272
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.10013037809648e-05,
+      "loss": 0.5271,
+      "step": 13806
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.5450840550178299,
+      "eval_loss": 0.5460476875305176,
+      "eval_runtime": 13.0324,
+      "eval_samples_per_second": 753.123,
+      "eval_steps_per_second": 2.993,
+      "step": 13806
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4.00019556714472e-05,
+      "loss": 0.5239,
+      "step": 15340
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.5462047885888945,
+      "eval_loss": 0.5449538230895996,
+      "eval_runtime": 13.0191,
+      "eval_samples_per_second": 753.893,
+      "eval_steps_per_second": 2.996,
+      "step": 15340
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.9002607561929596e-05,
+      "loss": 0.5209,
+      "step": 16874
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.5448802852776363,
+      "eval_loss": 0.5447258353233337,
+      "eval_runtime": 12.8387,
+      "eval_samples_per_second": 764.485,
+      "eval_steps_per_second": 3.038,
+      "step": 16874
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.80026075619296e-05,
+      "loss": 0.5179,
+      "step": 18408
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.5475292919001529,
+      "eval_loss": 0.5451586246490479,
+      "eval_runtime": 12.8829,
+      "eval_samples_per_second": 761.865,
+      "eval_steps_per_second": 3.027,
+      "step": 18408
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 3.7003259452412e-05,
+      "loss": 0.5152,
+      "step": 19942
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.5453897096281203,
+      "eval_loss": 0.5494815707206726,
+      "eval_runtime": 12.8734,
+      "eval_samples_per_second": 762.427,
+      "eval_steps_per_second": 3.03,
+      "step": 19942
+    },
+    {
+      "epoch": 13.0,
+      "step": 19942,
+      "total_flos": 2.974693409685504e+16,
+      "train_loss": 0.5403076514665969,
+      "train_runtime": 8354.8792,
+      "train_samples_per_second": 2350.136,
+      "train_steps_per_second": 9.18
+    }
+  ],
+  "max_steps": 76700,
+  "num_train_epochs": 50,
+  "total_flos": 2.974693409685504e+16,
+  "trial_name": null,
+  "trial_params": null
+}