Training in progress, epoch 2

logs/events.out.tfevents.1709716821.adc675a344d5.67573.3

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4cfec70353e1792393f9035254bbeb24c0f5630ab9c360b83e0ba97f5cb09f47
-size 5314
+oid sha256:3abafa2afbf81d92c37577fddf8088e64f889d1326f127e309cb795a84261afc
+size 5997

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:12de485927495a54a42aa2b293b0adf412d00170117941f81fbe4835dce02946
+oid sha256:8c2bbc0e16334675397da6eae3f9cec97a7ba066c14dfe9bfaeecbe8ce680b43
 size 17549312

run-2/checkpoint-192/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:444ec7032343c5e48a0bebcbc727c0d6dcc53de822b7281218d78d3d52fc74a5
+oid sha256:8c2bbc0e16334675397da6eae3f9cec97a7ba066c14dfe9bfaeecbe8ce680b43
 size 17549312

run-2/checkpoint-192/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a7d65e468a5ea6ece563e4a260f8a916b8919ff2c935298cbc0e77b353da1ae9
+oid sha256:f360aeb57e55cec8e68847ef51bb3de32b78e5a1d86bbea3082e69c5a0126e5b
 size 35122746

run-2/checkpoint-192/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76b1ca96b43548704a5f4d11597321c5afbd7e065f2cc83810db95ee4c1c77f7
+oid sha256:1bff0ab8e040a10e09a99435f8ce9d9312e47d8d81f03c526f96c8267c11c97d
 size 1064

run-2/checkpoint-192/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.8140900195694716,
+  "best_metric": 0.5675146771037182,
   "best_model_checkpoint": "tiny-bert-sst2-distilled/run-2/checkpoint-192",
   "epoch": 2.0,
   "eval_steps": 500,
@@ -10,55 +10,55 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 3.4323604106903076,
-      "learning_rate": 0.0004667999543114754,
-      "loss": 0.533,
+      "grad_norm": 0.663398027420044,
+      "learning_rate": 0.00027867701883546326,
+      "loss": 0.2787,
       "step": 96
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7984344422700587,
-      "eval_f1": 0.8049242424242424,
-      "eval_loss": 0.4442267119884491,
-      "eval_precision": 0.7798165137614679,
-      "eval_recall": 0.8317025440313112,
-      "eval_runtime": 28.6457,
-      "eval_samples_per_second": 35.677,
-      "eval_steps_per_second": 1.117,
+      "eval_accuracy": 0.5499021526418787,
+      "eval_f1": 0.24092409240924093,
+      "eval_loss": 0.25340619683265686,
+      "eval_precision": 0.7684210526315789,
+      "eval_recall": 0.14285714285714285,
+      "eval_runtime": 28.5207,
+      "eval_samples_per_second": 35.834,
+      "eval_steps_per_second": 1.122,
       "step": 96
     },
     {
       "epoch": 2.0,
-      "grad_norm": 2.4163033962249756,
-      "learning_rate": 0.0,
-      "loss": 0.4374,
+      "grad_norm": 0.7603411078453064,
+      "learning_rate": 0.00024384239148103037,
+      "loss": 0.2582,
       "step": 192
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8140900195694716,
-      "eval_f1": 0.8243992606284659,
-      "eval_loss": 0.4299594759941101,
-      "eval_precision": 0.7810858143607706,
-      "eval_recall": 0.87279843444227,
-      "eval_runtime": 28.93,
-      "eval_samples_per_second": 35.327,
-      "eval_steps_per_second": 1.106,
+      "eval_accuracy": 0.5675146771037182,
+      "eval_f1": 0.300632911392405,
+      "eval_loss": 0.2482573688030243,
+      "eval_precision": 0.7851239669421488,
+      "eval_recall": 0.18590998043052837,
+      "eval_runtime": 28.5782,
+      "eval_samples_per_second": 35.762,
+      "eval_steps_per_second": 1.12,
       "step": 192
     }
   ],
   "logging_steps": 500,
-  "max_steps": 192,
+  "max_steps": 864,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 2,
+  "num_train_epochs": 9,
   "save_steps": 500,
   "total_flos": 471390394560.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.88253398201078,
-    "learning_rate": 0.0009335999086229508,
-    "num_train_epochs": 2,
-    "temperature": 19
+    "alpha": 0.3483609408663828,
+    "learning_rate": 0.0003135116461898962,
+    "num_train_epochs": 9,
+    "temperature": 7
   }
 }

run-2/checkpoint-192/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2659042d49fa36ce1404b1f7057dd4643dc22f653d1fc44d521700862149e4b7
+oid sha256:f7adcc08e8d37cad0335688bbf483a90600b348a62bf08e2b72452e95967f2a4
 size 4920