Training in progress, epoch 1

logs/events.out.tfevents.1709717864.adc675a344d5.67573.4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0e926384d601e7580afd02031ac153c8d8b5e9c81a37747812c32f83eae15b76
-size 5314
+oid sha256:8205def1ffaee3a56054ff41eff371a0a74c352f2174322f8e6f4de32a4a7fb0
+size 6351

logs/events.out.tfevents.1709718096.adc675a344d5.67573.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76993f5e85ffe1b3d2a8c7e4936e214f89f887a4ffd429aea3ee179d7fbb691a
+size 5314

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:de0cafea6a0e6a4c517cdb42c18145404c25fcf88044931866f1604e66d2c782
+oid sha256:03de7b47dbf2c7a75f6e9ae666c226fd153cbe25b9dde09fbdac73c4e190baef
 size 17549312

run-3/checkpoint-192/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb4d5e48b7b7dcbc4c6a78ba6f64447d41c31642b436b85f9712ebc3f1eb4e10
+oid sha256:9e0c46f77816f5d575f127322066eefcebb06da630356471ed06200afcc53f0c
 size 17549312

run-3/checkpoint-192/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ed70a7588f0522641062b0d7a3e3b4990f0e2467607917b4e82af82ed85fe05c
+oid sha256:015954ebf3c057deae6f1b47044ff63ce64834a114a40e74092b9bb613604fba
 size 35122746

run-3/checkpoint-192/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bbb03ae8de430d092285c0e0abd694736b22b9adda00f4f93056103d08f6a645
+oid sha256:67bee4695b943faaa8a9e14b36a73be86edf96c1c23da94257c10f9bdd2e556c
 size 1064

run-3/checkpoint-192/trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.5,
-  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-96",
+  "best_metric": 0.7984344422700587,
+  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-192",
   "epoch": 2.0,
   "eval_steps": 500,
   "global_step": 192,
@@ -10,55 +10,55 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 2.101290702819824,
-      "learning_rate": 1.1962059637335337e-06,
-      "loss": 0.5173,
+      "grad_norm": 1.4839651584625244,
+      "learning_rate": 0.00013176351778592356,
+      "loss": 0.4981,
       "step": 96
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5,
-      "eval_f1": 0.6666666666666666,
-      "eval_loss": 0.5046124458312988,
-      "eval_precision": 0.5,
-      "eval_recall": 1.0,
-      "eval_runtime": 28.251,
-      "eval_samples_per_second": 36.176,
-      "eval_steps_per_second": 1.133,
+      "eval_accuracy": 0.7651663405088063,
+      "eval_f1": 0.781021897810219,
+      "eval_loss": 0.4340898096561432,
+      "eval_precision": 0.7316239316239316,
+      "eval_recall": 0.837573385518591,
+      "eval_runtime": 28.2114,
+      "eval_samples_per_second": 36.227,
+      "eval_steps_per_second": 1.134,
       "step": 96
     },
     {
       "epoch": 2.0,
-      "grad_norm": 0.6350829005241394,
-      "learning_rate": 1.0466802182668422e-06,
-      "loss": 0.4989,
+      "grad_norm": 1.3078362941741943,
+      "learning_rate": 0.0,
+      "loss": 0.436,
       "step": 192
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.4843444227005871,
-      "eval_f1": 0.6216798277099785,
-      "eval_loss": 0.48798972368240356,
-      "eval_precision": 0.4909297052154195,
-      "eval_recall": 0.8473581213307241,
-      "eval_runtime": 28.492,
-      "eval_samples_per_second": 35.87,
-      "eval_steps_per_second": 1.123,
+      "eval_accuracy": 0.7984344422700587,
+      "eval_f1": 0.8134057971014492,
+      "eval_loss": 0.4166872203350067,
+      "eval_precision": 0.7571669477234402,
+      "eval_recall": 0.8786692759295499,
+      "eval_runtime": 28.4558,
+      "eval_samples_per_second": 35.915,
+      "eval_steps_per_second": 1.125,
       "step": 192
     }
   ],
   "logging_steps": 500,
-  "max_steps": 864,
+  "max_steps": 192,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 9,
+  "num_train_epochs": 2,
   "save_steps": 500,
   "total_flos": 471390394560.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.586704262013931,
-    "learning_rate": 1.3457317092002255e-06,
-    "num_train_epochs": 9,
-    "temperature": 21
+    "alpha": 0.7300485455769454,
+    "learning_rate": 0.0002635270355718471,
+    "num_train_epochs": 2,
+    "temperature": 7
   }
 }

run-3/checkpoint-192/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b765cab6920f955460cc4f812305ba7f085db0f4dbb906f08e22ff912fece657
+oid sha256:71c9cf86ae836f33dfc80cf04e65e588263c7e4400c5721df283c07a74df5e2e
 size 4920

run-4/checkpoint-96/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a544be95f6f04106deb7c2137fc736ef24e79997e59480896b72c927cd7c7aee
+oid sha256:03de7b47dbf2c7a75f6e9ae666c226fd153cbe25b9dde09fbdac73c4e190baef
 size 17549312

run-4/checkpoint-96/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:92e3700eb6e67af73a6d17192d6328f2e3438132782e383d04facbbbc511695e
+oid sha256:134ad570012451e34f610fb7aebf0a50a95f255efd0f06143cf454f2e8414625
 size 35122746

run-4/checkpoint-96/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:98307cb8361e610ce6078aa8e9ec07ca2d0f6a51244074e95f21bfc44ecff6e0
+oid sha256:15638891015f984827fd9d5a759b43a6ab5e05047f16704e6ec5463662b0acc7
 size 1064

run-4/checkpoint-96/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.5,
+  "best_metric": 0.541095890410959,
   "best_model_checkpoint": "tiny-bert-sst2-distilled/run-4/checkpoint-96",
   "epoch": 1.0,
   "eval_steps": 500,
@@ -10,36 +10,36 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 0.6504594683647156,
-      "learning_rate": 7.787665972957169e-06,
-      "loss": 0.59,
+      "grad_norm": 0.4191925525665283,
+      "learning_rate": 0.00022738674133918383,
+      "loss": 0.098,
       "step": 96
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5,
-      "eval_f1": 0.0,
-      "eval_loss": 0.5820541977882385,
-      "eval_precision": 0.0,
-      "eval_recall": 0.0,
-      "eval_runtime": 28.7531,
-      "eval_samples_per_second": 35.544,
-      "eval_steps_per_second": 1.113,
+      "eval_accuracy": 0.541095890410959,
+      "eval_f1": 0.20642978003384094,
+      "eval_loss": 0.08223184198141098,
+      "eval_precision": 0.7625,
+      "eval_recall": 0.11937377690802348,
+      "eval_runtime": 28.0826,
+      "eval_samples_per_second": 36.393,
+      "eval_steps_per_second": 1.139,
       "step": 96
     }
   ],
   "logging_steps": 500,
-  "max_steps": 192,
+  "max_steps": 480,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 2,
+  "num_train_epochs": 5,
   "save_steps": 500,
   "total_flos": 235695197280.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.7979164056999755,
-    "learning_rate": 1.5575331945914338e-05,
-    "num_train_epochs": 2,
-    "temperature": 5
+    "alpha": 0.09292461434001364,
+    "learning_rate": 0.0002842334266739798,
+    "num_train_epochs": 5,
+    "temperature": 2
   }
 }

run-4/checkpoint-96/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d75bccfe6c5373ca9b1b18dfaff7dd2ce79806a7acbd4f6cd52c93fdf2f1e00c
+oid sha256:f441eefbf68b7466d3f5396e1b086e4cb1ae1be01fffc857e92bd1ad437f6f6d
 size 4920

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71c9cf86ae836f33dfc80cf04e65e588263c7e4400c5721df283c07a74df5e2e
+oid sha256:f441eefbf68b7466d3f5396e1b086e4cb1ae1be01fffc857e92bd1ad437f6f6d
 size 4920