Training in progress, epoch 2

logs/events.out.tfevents.1719316225.49cbd00d2005.382.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:298724a4bb557672c166bc1e25254912866c59cdddf96831c2184896ca9bdd45
-size 5427
+oid sha256:6cd44486db105a3795a27bd7e66817354b0b0c46c44ae0bd71e8c1ac02b9c2e4
+size 5961

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e4101e13540b9ca54dbd8ddf761e2b1b8af8108fb9147967fcd9ee7dd0b7ff1
+oid sha256:9880a4d6b160b708dee866b97b67f4767461cdec03dbdef71638961456f15266
 size 17549312

run-0/checkpoint-1054/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dc0fee5c27b13a0a9c632d2aac5f70e2a09379e669796176adfed370c09bec74
+oid sha256:9880a4d6b160b708dee866b97b67f4767461cdec03dbdef71638961456f15266
 size 17549312

run-0/checkpoint-1054/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:180bafd9040bf7f35eca41800b8bc8fea98cfd11101725956e8708ac27fa2b2f
+oid sha256:f0d8a608e526473174b7c31c2121364dd63d55f262ae5e4c78252908e6e661c4
 size 35123898

run-0/checkpoint-1054/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:100464d33f6556fd1f8f558519b1030d76a9ca5f1f9e07ec8e190d748723f68d
+oid sha256:5a77675298fd65dc14bb0a83836346073a553ae69d629ef591046deeaac3870d
 size 1064

run-0/checkpoint-1054/trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.7809633027522935,
-  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-1054",
+  "best_metric": 0.8176605504587156,
+  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-527",
   "epoch": 2.0,
   "eval_steps": 500,
   "global_step": 1054,
@@ -10,41 +10,41 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 12.377176284790039,
-      "learning_rate": 1.5980854192157344e-05,
-      "loss": 3.0218,
+      "grad_norm": 15.690781593322754,
+      "learning_rate": 0.0004329892843734803,
+      "loss": 1.29,
       "step": 527
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7672018348623854,
-      "eval_loss": 2.191861391067505,
-      "eval_runtime": 2.5906,
-      "eval_samples_per_second": 336.603,
-      "eval_steps_per_second": 2.702,
+      "eval_accuracy": 0.8176605504587156,
+      "eval_loss": 1.5847134590148926,
+      "eval_runtime": 2.4131,
+      "eval_samples_per_second": 361.36,
+      "eval_steps_per_second": 2.901,
       "step": 527
     },
     {
       "epoch": 2.0,
-      "grad_norm": Infinity,
-      "learning_rate": 1.2796813034327135e-05,
-      "loss": 2.0797,
+      "grad_norm": 3.8686888217926025,
+      "learning_rate": 0.0002886595229156535,
+      "loss": 0.5557,
       "step": 1054
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.7809633027522935,
-      "eval_loss": 1.8230279684066772,
-      "eval_runtime": 2.5719,
-      "eval_samples_per_second": 339.043,
-      "eval_steps_per_second": 2.722,
+      "eval_accuracy": 0.8142201834862385,
+      "eval_loss": 1.6788822412490845,
+      "eval_runtime": 2.4261,
+      "eval_samples_per_second": 359.428,
+      "eval_steps_per_second": 2.885,
       "step": 1054
     }
   ],
   "logging_steps": 500,
-  "max_steps": 3162,
+  "max_steps": 2108,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 6,
+  "num_train_epochs": 4,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -62,9 +62,9 @@
   "train_batch_size": 128,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.275210688972374,
-    "learning_rate": 1.9177025030588814e-05,
-    "num_train_epochs": 6,
+    "alpha": 0.19981548442581198,
+    "learning_rate": 0.000577319045831307,
+    "num_train_epochs": 4,
     "temperature": 6
   }
 }

run-0/checkpoint-1054/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:15e7b4feae857373f91378a3d0efc15d0ec396bfba71e74c965086843aa6acf4
+oid sha256:9b553ab9101fed6c5a9e75fca4cb81df81514d2d3f8ff94edc0c546949c119f9
 size 5176