Training in progress, epoch 2

logs/events.out.tfevents.1709737723.1dcff7455174.7719.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cdbda6302dedd3c7a125c014539e9be2875b24d549ede70f091bc89bed984ea6
-size 5315
+oid sha256:09a20a957e990e72af5f328fa379bb1a62e309b4b5b228aee4985819638a2f9e
+size 5998

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f580c7021c521e50c4f09989078be6521aed1a3504233121cb4128acd7ffaf3f
+oid sha256:691fe6dae22dbcf425f6968004894acb3404da27e32c1c6aaa864221b6e71a6b
 size 17549312

run-1/checkpoint-192/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:564f1f46932544fd89608b22974dc27996ca6dfb380e01ea2b6299c6c592a2bf
+oid sha256:691fe6dae22dbcf425f6968004894acb3404da27e32c1c6aaa864221b6e71a6b
 size 17549312

run-1/checkpoint-192/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8ee512f92ffc9dd07acbff560ce3a2fdb296640d0c6d82249296365ed13153f8
+oid sha256:a934e87408b233a8ea983945295f058bd23167055b3f8990bfd1682daa969cc9
 size 35122746

run-1/checkpoint-192/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0dbb0fba5b183172f8d39b2f13bd929411213cf1ad515fc75a81d08268624f8d
+oid sha256:b3d1ec3eff6c7b0aaaccde3d1da3e3c2d398ec546ecd0a5fb3c9941dfe4127dd
 size 1064

run-1/checkpoint-192/trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.8082191780821918,
-  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-1/checkpoint-192",
+  "best_metric": 0.5,
+  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-1/checkpoint-96",
   "epoch": 2.0,
   "eval_steps": 500,
   "global_step": 192,
@@ -10,55 +10,55 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 2.6617980003356934,
-      "learning_rate": 0.0006635023570917325,
-      "loss": 0.5037,
+      "grad_norm": 0.40490686893463135,
+      "learning_rate": 1.4953262133881219e-05,
+      "loss": 0.1703,
       "step": 96
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7661448140900196,
-      "eval_f1": 0.792714657415438,
-      "eval_loss": 0.4430159628391266,
-      "eval_precision": 0.7118380062305296,
-      "eval_recall": 0.8943248532289628,
-      "eval_runtime": 28.6854,
-      "eval_samples_per_second": 35.628,
-      "eval_steps_per_second": 1.116,
+      "eval_accuracy": 0.5,
+      "eval_f1": 0.0,
+      "eval_loss": 0.11485283076763153,
+      "eval_precision": 0.0,
+      "eval_recall": 0.0,
+      "eval_runtime": 25.7126,
+      "eval_samples_per_second": 39.747,
+      "eval_steps_per_second": 1.245,
       "step": 96
     },
     {
       "epoch": 2.0,
-      "grad_norm": 1.12320077419281,
-      "learning_rate": 0.0004976267678187994,
-      "loss": 0.4391,
+      "grad_norm": 0.2503454089164734,
+      "learning_rate": 1.2817081829041045e-05,
+      "loss": 0.1095,
       "step": 192
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8082191780821918,
-      "eval_f1": 0.8227848101265822,
-      "eval_loss": 0.41584911942481995,
-      "eval_precision": 0.7647058823529411,
-      "eval_recall": 0.8904109589041096,
-      "eval_runtime": 28.5215,
-      "eval_samples_per_second": 35.833,
-      "eval_steps_per_second": 1.122,
+      "eval_accuracy": 0.5,
+      "eval_f1": 0.0038986354775828462,
+      "eval_loss": 0.0978715717792511,
+      "eval_precision": 0.5,
+      "eval_recall": 0.0019569471624266144,
+      "eval_runtime": 25.921,
+      "eval_samples_per_second": 39.427,
+      "eval_steps_per_second": 1.235,
       "step": 192
     }
   ],
   "logging_steps": 500,
-  "max_steps": 480,
+  "max_steps": 768,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 8,
   "save_steps": 500,
   "total_flos": 471390394560.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.7818460169984094,
-    "learning_rate": 0.0008293779463646656,
-    "num_train_epochs": 5,
-    "temperature": 3
+    "alpha": 0.09656764013507046,
+    "learning_rate": 1.7089442438721394e-05,
+    "num_train_epochs": 8,
+    "temperature": 20
   }
 }

run-1/checkpoint-192/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1ab2cfe4c2ab241510bc23e9ec5727876e620ab477a4e5ad72a937b25e09df23
+oid sha256:40c83b924ef70318ad972d6c481c684f04bf59974ceae562e8816ac4c289de55
 size 4920