Training in progress, epoch 2

logs/events.out.tfevents.1719316946.49cbd00d2005.382.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:32027f51491766380aa4b56b9780645239e4a1be2cee3a370fbedf1eeff9b8b9
-size 5428
+oid sha256:dc06a89bf4f3acb77b416b957edf4b8d0c7487ba1d7b08f8ca6b28909e0e8659
+size 5962

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:27cee0faa93c8ebfdb171e60acd1f01bcf52efd270df1bebe4847e474aab2212
+oid sha256:9b81af07e0ebeae51ba086b61ef32b5b3d7f690b9b86ffede2609f3651486fd3
 size 17549312

run-1/checkpoint-1054/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b4db5ce53872114bd04603e9f4cf8a2c9fff8b5eba2e6c35fa6175cd08190265
+oid sha256:9b81af07e0ebeae51ba086b61ef32b5b3d7f690b9b86ffede2609f3651486fd3
 size 17549312

run-1/checkpoint-1054/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e1dd136a0ee7ffbf29f6a85dc5e2e539fdd79d2195a573a0abab8048ddf9905f
+oid sha256:43ca175f0a3f0d3b6098f2ce36a4acbfd95d7d212be030eec4dc090dfb722108
 size 35123898

run-1/checkpoint-1054/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec5be829c4dd6c4d8b768ac32866ba5e4919b044a4f27c321cee266e36d17472
+oid sha256:c51af740af6cdafe257b8aaac73fb8a3ce20ed9c05fc5b7d8ec20c8584be4a5b
 size 1064

run-1/checkpoint-1054/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.8268348623853211,
+  "best_metric": 0.8130733944954128,
   "best_model_checkpoint": "tiny-bert-sst2-distilled/run-1/checkpoint-527",
   "epoch": 2.0,
   "eval_steps": 500,
@@ -10,34 +10,34 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 20.18393898010254,
-      "learning_rate": 0.0004400348679504635,
-      "loss": 1.3749,
+      "grad_norm": 6.467271327972412,
+      "learning_rate": 0.0005246601812780973,
+      "loss": 0.349,
       "step": 527
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.8268348623853211,
-      "eval_loss": 1.6724135875701904,
-      "eval_runtime": 2.5582,
-      "eval_samples_per_second": 340.862,
-      "eval_steps_per_second": 2.736,
+      "eval_accuracy": 0.8130733944954128,
+      "eval_loss": 0.5175108313560486,
+      "eval_runtime": 2.3764,
+      "eval_samples_per_second": 366.943,
+      "eval_steps_per_second": 2.946,
       "step": 527
     },
     {
       "epoch": 2.0,
-      "grad_norm": 4.33894681930542,
-      "learning_rate": 0.00022001743397523176,
-      "loss": 0.5714,
+      "grad_norm": 4.971531867980957,
+      "learning_rate": 0.00026233009063904864,
+      "loss": 0.1805,
       "step": 1054
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8222477064220184,
-      "eval_loss": 1.7203704118728638,
-      "eval_runtime": 2.5544,
-      "eval_samples_per_second": 341.369,
-      "eval_steps_per_second": 2.74,
+      "eval_accuracy": 0.8027522935779816,
+      "eval_loss": 0.6036062240600586,
+      "eval_runtime": 2.3772,
+      "eval_samples_per_second": 366.815,
+      "eval_steps_per_second": 2.945,
       "step": 1054
     }
   ],
@@ -62,9 +62,9 @@
   "train_batch_size": 128,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.22319309854701086,
-    "learning_rate": 0.0006600523019256953,
+    "alpha": 0.9438943885488221,
+    "learning_rate": 0.0007869902719171459,
     "num_train_epochs": 3,
-    "temperature": 16
+    "temperature": 2
   }
 }

run-1/checkpoint-1054/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ee9990afd96f204321e9e33fc4071f73d76355111015fd2ec723f4a4e4849ce7
+oid sha256:13747d4a4002463522d34a00fd1d294fed1696984634001cefecc3ef3eb702bf
 size 5176