Training in progress, epoch 2

logs/events.out.tfevents.1709701849.adc675a344d5.5455.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f17cea2e9947009500120825d917932ab211e3cf3fbb891f9661b2669b04247
-size 5315
+oid sha256:390f1a682c8c6754679f33a5b08baf722a2030d7fa0526e591a80b7ca2d7519a
+size 5998

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:436c1223e40cc411f216b7a6eb698829ee74318f91d2f9ee826e701f17df37dd
+oid sha256:a7c5e94d351fdbc0935e7dbe3174244feb7ddadc5f63dc2f16808c51b2af2a57
 size 17549312

run-0/checkpoint-192/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:97677f0927993afb82c2c7c208050b62610c5ab549a92c0419416c1a7cf7cc5d
+oid sha256:a7c5e94d351fdbc0935e7dbe3174244feb7ddadc5f63dc2f16808c51b2af2a57
 size 17549312

run-0/checkpoint-192/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:91dcd2b11b658feb6b086771ace577c2ffec323121432b2d402b37f23f12512e
+oid sha256:ac78cbe4519b949fd207235051b160ea7f7956b055fb088955c4a1ac6b0df268
 size 35122746

run-0/checkpoint-192/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8ffa140988cbc374b9f7bdd5029601ff518f02e9d39cb930987bfe22a8165d4e
+oid sha256:13915667501739529d666ffe624aa321331ccbffecf0ae7d04a789ecdbde4536
 size 1064

run-0/checkpoint-192/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.5958904109589042,
+  "best_metric": 0.7896281800391389,
   "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-192",
   "epoch": 2.0,
   "eval_steps": 500,
@@ -10,55 +10,55 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 0.4804088771343231,
-      "learning_rate": 2.9818131608660274e-05,
-      "loss": 0.4786,
+      "grad_norm": 1.8004393577575684,
+      "learning_rate": 0.0007697157586205755,
+      "loss": 0.4935,
       "step": 96
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5,
-      "eval_f1": 0.0,
-      "eval_loss": 0.4664871394634247,
-      "eval_precision": 0.0,
-      "eval_recall": 0.0,
-      "eval_runtime": 28.9797,
-      "eval_samples_per_second": 35.266,
-      "eval_steps_per_second": 1.104,
+      "eval_accuracy": 0.7436399217221135,
+      "eval_f1": 0.7689594356261024,
+      "eval_loss": 0.448881596326828,
+      "eval_precision": 0.6998394863563403,
+      "eval_recall": 0.8532289628180039,
+      "eval_runtime": 28.6476,
+      "eval_samples_per_second": 35.675,
+      "eval_steps_per_second": 1.117,
       "step": 96
     },
     {
       "epoch": 2.0,
-      "grad_norm": 1.77193021774292,
-      "learning_rate": 1.9878754405773516e-05,
-      "loss": 0.4481,
+      "grad_norm": 4.9879961013793945,
+      "learning_rate": 0.0006841917854405115,
+      "loss": 0.4571,
       "step": 192
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.5958904109589042,
-      "eval_f1": 0.37707390648567113,
-      "eval_loss": 0.4186549484729767,
-      "eval_precision": 0.8223684210526315,
-      "eval_recall": 0.2446183953033268,
-      "eval_runtime": 28.789,
-      "eval_samples_per_second": 35.5,
+      "eval_accuracy": 0.7896281800391389,
+      "eval_f1": 0.8057813911472448,
+      "eval_loss": 0.42557916045188904,
+      "eval_precision": 0.7483221476510067,
+      "eval_recall": 0.87279843444227,
+      "eval_runtime": 28.772,
+      "eval_samples_per_second": 35.521,
       "eval_steps_per_second": 1.112,
       "step": 192
     }
   ],
   "logging_steps": 500,
-  "max_steps": 384,
+  "max_steps": 960,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 4,
+  "num_train_epochs": 10,
   "save_steps": 500,
   "total_flos": 471390394560.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.6186723691968887,
-    "learning_rate": 3.975750881154703e-05,
-    "num_train_epochs": 4,
-    "temperature": 6
+    "alpha": 0.735211359802956,
+    "learning_rate": 0.0008552397318006393,
+    "num_train_epochs": 10,
+    "temperature": 25
   }
 }

run-0/checkpoint-192/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3aa29ea69dd18a01982daa5b87211d303a5144b609fb7e84569d6d408da8e995
+oid sha256:23804eaaa0feed30486319d2982557f65dce708f7d788864c811c20b2e8c33b4
 size 4920