Training in progress, epoch 2

logs/events.out.tfevents.1709709278.a2333cf40ebd.21010.47

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b5b548f74c25d94d70b6e0bb74510f76eedc69edaeffea8f010be4fda4d356e3
-size 5315
+oid sha256:521f4cc4877fd9bc55e65945cad678d82b968bcffc2534d2f8b914114094b7fc
+size 5998

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f82b95e45bf81d582afa4e71bc9d7a293f1df5f4cc4679ddfe0cd23d18205cf
+oid sha256:815fff10bdd7a02e0211babfb38b7a79fbd0f003f1b0c15d54aa5ddc6ce151a0
 size 17549312

run-47/checkpoint-192/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b650efa919f6a4973bb634fc29f097f7b8f22afbce00a3634222130b8799584e
+oid sha256:815fff10bdd7a02e0211babfb38b7a79fbd0f003f1b0c15d54aa5ddc6ce151a0
 size 17549312

run-47/checkpoint-192/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1162fc127b4cd021624e70033605878611ce1ce20b393ac289567fc571860a54
+oid sha256:8b33aa1bb9141c5e568b3bd78dee74d4de254d21a72451ebf2268ee3517c4e8c
 size 35122746

run-47/checkpoint-192/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:46ff4487f6324742fb6cf7e7658cb071ca8124651eeb533389b314bc24c7eb76
+oid sha256:3c29bcd8ab24ec1eb0bf59ae4d78d625d5cd1845d60be1cfcdcf9efc60719e55
 size 1064

run-47/checkpoint-192/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.8297455968688845,
+  "best_metric": 0.8043052837573386,
   "best_model_checkpoint": "tiny-bert-sst2-distilled/run-47/checkpoint-192",
   "epoch": 2.0,
   "eval_steps": 500,
@@ -10,55 +10,55 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 10.090946197509766,
-      "learning_rate": 0.0002726931435690813,
-      "loss": 0.5743,
+      "grad_norm": 1.9338581562042236,
+      "learning_rate": 0.0001802406049279757,
+      "loss": 0.5354,
       "step": 96
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7964774951076321,
-      "eval_f1": 0.8188153310104529,
-      "eval_loss": 0.4561440944671631,
-      "eval_precision": 0.7378335949764521,
-      "eval_recall": 0.9197651663405088,
-      "eval_runtime": 30.8585,
-      "eval_samples_per_second": 33.119,
-      "eval_steps_per_second": 1.037,
+      "eval_accuracy": 0.7671232876712328,
+      "eval_f1": 0.8023255813953489,
+      "eval_loss": 0.4617577791213989,
+      "eval_precision": 0.696969696969697,
+      "eval_recall": 0.9452054794520548,
+      "eval_runtime": 29.4455,
+      "eval_samples_per_second": 34.708,
+      "eval_steps_per_second": 1.087,
       "step": 96
     },
     {
       "epoch": 2.0,
-      "grad_norm": 7.258547306060791,
-      "learning_rate": 0.00018179542904605417,
-      "loss": 0.4449,
+      "grad_norm": 2.99806547164917,
+      "learning_rate": 0.00015771052931197875,
+      "loss": 0.4503,
       "step": 192
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8297455968688845,
-      "eval_f1": 0.8406593406593407,
-      "eval_loss": 0.4051818251609802,
-      "eval_precision": 0.7900172117039587,
-      "eval_recall": 0.898238747553816,
-      "eval_runtime": 31.0429,
-      "eval_samples_per_second": 32.922,
-      "eval_steps_per_second": 1.031,
+      "eval_accuracy": 0.8043052837573386,
+      "eval_f1": 0.8242530755711774,
+      "eval_loss": 0.42502960562705994,
+      "eval_precision": 0.748006379585327,
+      "eval_recall": 0.9178082191780822,
+      "eval_runtime": 28.5574,
+      "eval_samples_per_second": 35.788,
+      "eval_steps_per_second": 1.121,
       "step": 192
     }
   ],
   "logging_steps": 500,
-  "max_steps": 384,
+  "max_steps": 864,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 4,
+  "num_train_epochs": 9,
   "save_steps": 500,
   "total_flos": 471390394560.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.999132297568699,
-    "learning_rate": 0.00036359085809210834,
-    "num_train_epochs": 4,
-    "temperature": 14
+    "alpha": 0.8114321139277666,
+    "learning_rate": 0.00020277068054397267,
+    "num_train_epochs": 9,
+    "temperature": 7
   }
 }

run-47/checkpoint-192/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6e52d4ef52e4ae3e7583492431957732ea2f0822ace3819ef8d07c56bc154648
+oid sha256:0cca8f0e3496a02e33b0031c0a941841ea6a66348bacabeaa9cc0ea789d82d49
 size 4920