Training in progress, epoch 2

logs/events.out.tfevents.1709707514.a2333cf40ebd.21010.41

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bfa8c1498e201bd3a30de466fb607872f802977cdaa65610c8ff3d9ec4a3316c
-size 5314
+oid sha256:c2f711a5ec506a9a696f18c8f63c43fb1a2c80a718a4da2aed3d9eb383d758a1
+size 5997

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:09ba047a380994384e599e41981d871ceaba0d611aa17668901a037a269e6885
+oid sha256:0f16e0475c91a898b7f27878fd3f4741d35bb2812143905032487cdc6b500731
 size 17549312

run-41/checkpoint-192/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3fdd89e21494babf3f8e1e915b0fe6f3e7841fdd2b40b51b83f6a7aaa8e37ec6
+oid sha256:0f16e0475c91a898b7f27878fd3f4741d35bb2812143905032487cdc6b500731
 size 17549312

run-41/checkpoint-192/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:506981a863366ff3e12985d91082d331edfe322e16142b05fdb15c4e238e2b05
+oid sha256:80d46c5cddf908ecbbe3ba73e2792a404a588d20acfcab030f4360ef8b897ea7
 size 35122746

run-41/checkpoint-192/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cac39eb81d302d89d0d93731a31fb32c12d0c9638749ff4694c2b43672d79f90
+oid sha256:3cc89d7435dad97d7cfce432c467706cf7b123fe9c606ad4b73de741b6ff575b
 size 1064

run-41/checkpoint-192/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.8199608610567515,
+  "best_metric": 0.8082191780821918,
   "best_model_checkpoint": "tiny-bert-sst2-distilled/run-41/checkpoint-192",
   "epoch": 2.0,
   "eval_steps": 500,
@@ -10,55 +10,55 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 2.904797315597534,
-      "learning_rate": 0.0001800620290766639,
-      "loss": 0.5678,
+      "grad_norm": 1.4870496988296509,
+      "learning_rate": 0.00011097537892965708,
+      "loss": 0.5464,
       "step": 96
     },
     {
       "epoch": 1.0,
       "eval_accuracy": 0.7749510763209393,
-      "eval_f1": 0.8060708263069141,
-      "eval_loss": 0.47195959091186523,
-      "eval_precision": 0.7081481481481482,
-      "eval_recall": 0.9354207436399217,
-      "eval_runtime": 32.7213,
-      "eval_samples_per_second": 31.233,
-      "eval_steps_per_second": 0.978,
+      "eval_f1": 0.8017241379310346,
+      "eval_loss": 0.4673503041267395,
+      "eval_precision": 0.7164869029275809,
+      "eval_recall": 0.9099804305283757,
+      "eval_runtime": 28.2108,
+      "eval_samples_per_second": 36.227,
+      "eval_steps_per_second": 1.134,
       "step": 96
     },
     {
       "epoch": 2.0,
-      "grad_norm": 4.956325054168701,
-      "learning_rate": 9.003101453833196e-05,
-      "loss": 0.4548,
+      "grad_norm": 1.9109100103378296,
+      "learning_rate": 9.710345656344995e-05,
+      "loss": 0.4617,
       "step": 192
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8199608610567515,
-      "eval_f1": 0.8327272727272728,
-      "eval_loss": 0.42627328634262085,
-      "eval_precision": 0.7775891341256367,
-      "eval_recall": 0.8962818003913894,
-      "eval_runtime": 30.8838,
-      "eval_samples_per_second": 33.092,
-      "eval_steps_per_second": 1.036,
+      "eval_accuracy": 0.8082191780821918,
+      "eval_f1": 0.8262411347517731,
+      "eval_loss": 0.43051373958587646,
+      "eval_precision": 0.7552674230145867,
+      "eval_recall": 0.9119373776908023,
+      "eval_runtime": 28.0456,
+      "eval_samples_per_second": 36.441,
+      "eval_steps_per_second": 1.141,
       "step": 192
     }
   ],
   "logging_steps": 500,
-  "max_steps": 288,
+  "max_steps": 864,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 3,
+  "num_train_epochs": 9,
   "save_steps": 500,
   "total_flos": 471390394560.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.9482965338753422,
-    "learning_rate": 0.0002700930436149959,
-    "num_train_epochs": 3,
-    "temperature": 9
+    "alpha": 0.8078288065121237,
+    "learning_rate": 0.0001248473012958642,
+    "num_train_epochs": 9,
+    "temperature": 4
   }
 }

run-41/checkpoint-192/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a79ffddd3dbb36d7991e9db486bb85c65a0c011c05a4c2d55c9d8aaeca1b28f2
+oid sha256:9599a9eb05fb1abdd888ed4c4bfde1c3333a817ee2a411e437b9a1cf607a92ec
 size 4920