Training in progress, epoch 2

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:73e4d6a6e947f36beb3343b91199b7400ab1fad9f13965ee5f2f874b0f4d0ea0
+oid sha256:c2ced358dc3bc4605895c8027d988748fb8764af9b57b5edaad008d467ac9df5
 size 116795404

run-2/checkpoint-1100/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c0363f67352cbd2526e198d0c20d3193cdea64d07435cfaa2ea803f2af4cc753
+oid sha256:c2ced358dc3bc4605895c8027d988748fb8764af9b57b5edaad008d467ac9df5
 size 116795404

run-2/checkpoint-1100/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6f615d3633d28e1402491ae73af70416cc428deec610ad3f6c9513f9e00f23ec
+oid sha256:171427067aaa2194d8a81db089c70b669189c9c98fde6c739061a0a3a2068e40
 size 233621882

run-2/checkpoint-1100/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3180da2883c56c8e0a6a790414652de70473fb9817815bef9e488338d650da67
+oid sha256:4ada2bb04b88d0294fe44121ff921a07d99208db29ca33d0b0f391e55a9684bc
 size 13990

run-2/checkpoint-1100/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:85e5ab558f76098b7874b0df51d8d53a08350d7818f4d448db67ff258e0c2c24
+oid sha256:cee9d68d7ffd6e02e89ecadac88c09363aa50fa3ffbe10ea51c490d81f296210
 size 1064

run-2/checkpoint-1100/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.8571428571428571,
+  "best_metric": 0.7755102040816326,
   "best_model_checkpoint": "rubert-tiny2-finetuned-classification\\run-2\\checkpoint-1100",
   "epoch": 2.0,
   "eval_steps": 500,
@@ -10,47 +10,47 @@
   "log_history": [
     {
       "epoch": 0.91,
-      "learning_rate": 1.5039435145104783e-05,
-      "loss": 1.8268,
+      "learning_rate": 7.123150895957912e-06,
+      "loss": 2.1151,
       "step": 500
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7224489795918367,
-      "eval_loss": 1.1639864444732666,
-      "eval_runtime": 1.3165,
-      "eval_samples_per_second": 186.096,
-      "eval_steps_per_second": 12.153,
+      "eval_accuracy": 0.6326530612244898,
+      "eval_loss": 1.7264677286148071,
+      "eval_runtime": 1.125,
+      "eval_samples_per_second": 217.778,
+      "eval_steps_per_second": 14.222,
       "step": 550
     },
     {
       "epoch": 1.82,
-      "learning_rate": 1.0616071867132787e-05,
-      "loss": 1.0445,
+      "learning_rate": 5.540228474633931e-06,
+      "loss": 1.5823,
       "step": 1000
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8571428571428571,
-      "eval_loss": 0.730604887008667,
-      "eval_runtime": 1.103,
-      "eval_samples_per_second": 222.131,
-      "eval_steps_per_second": 14.507,
+      "eval_accuracy": 0.7755102040816326,
+      "eval_loss": 1.2948896884918213,
+      "eval_runtime": 1.575,
+      "eval_samples_per_second": 155.555,
+      "eval_steps_per_second": 10.159,
       "step": 1100
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2200,
+  "max_steps": 2750,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 4,
+  "num_train_epochs": 5,
   "save_steps": 500,
-  "total_flos": 1918563271800.0,
+  "total_flos": 1945715472750.0,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": {
-    "learning_rate": 1.946279842307678e-05,
-    "num_train_epochs": 4,
+    "learning_rate": 8.706073317281892e-06,
+    "num_train_epochs": 5,
     "per_device_train_batch_size": 4,
-    "seed": 23
+    "seed": 1
   }
 }

run-2/checkpoint-1100/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24844b54c4e4c2ebe3c69bc8bd3a5158a5ee05c4aebe057006cf6ebf084d9bab
+oid sha256:6bf6249fcaffd3bf1a6e41efb7044edf4d8222cdd0729549fa8f87e1c5b20556
 size 4728

run-2/checkpoint-1650/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.8734693877551021,
+  "best_metric": 0.8,
   "best_model_checkpoint": "rubert-tiny2-finetuned-classification\\run-2\\checkpoint-1650",
   "epoch": 3.0,
   "eval_steps": 500,
@@ -10,62 +10,62 @@
   "log_history": [
     {
       "epoch": 0.91,
-      "learning_rate": 1.5039435145104783e-05,
-      "loss": 1.8268,
+      "learning_rate": 7.123150895957912e-06,
+      "loss": 2.1151,
       "step": 500
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7224489795918367,
-      "eval_loss": 1.1639864444732666,
-      "eval_runtime": 1.3165,
-      "eval_samples_per_second": 186.096,
-      "eval_steps_per_second": 12.153,
+      "eval_accuracy": 0.6326530612244898,
+      "eval_loss": 1.7264677286148071,
+      "eval_runtime": 1.125,
+      "eval_samples_per_second": 217.778,
+      "eval_steps_per_second": 14.222,
       "step": 550
     },
     {
       "epoch": 1.82,
-      "learning_rate": 1.0616071867132787e-05,
-      "loss": 1.0445,
+      "learning_rate": 5.540228474633931e-06,
+      "loss": 1.5823,
       "step": 1000
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8571428571428571,
-      "eval_loss": 0.730604887008667,
-      "eval_runtime": 1.103,
-      "eval_samples_per_second": 222.131,
-      "eval_steps_per_second": 14.507,
+      "eval_accuracy": 0.7755102040816326,
+      "eval_loss": 1.2948896884918213,
+      "eval_runtime": 1.575,
+      "eval_samples_per_second": 155.555,
+      "eval_steps_per_second": 10.159,
       "step": 1100
     },
     {
       "epoch": 2.73,
-      "learning_rate": 6.192708589160794e-06,
-      "loss": 0.7493,
+      "learning_rate": 3.9573060533099506e-06,
+      "loss": 1.2756,
       "step": 1500
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8734693877551021,
-      "eval_loss": 0.5868724584579468,
-      "eval_runtime": 1.1252,
-      "eval_samples_per_second": 217.733,
-      "eval_steps_per_second": 14.219,
+      "eval_accuracy": 0.8,
+      "eval_loss": 1.0647388696670532,
+      "eval_runtime": 2.6287,
+      "eval_samples_per_second": 93.202,
+      "eval_steps_per_second": 6.087,
       "step": 1650
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2200,
+  "max_steps": 2750,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 4,
+  "num_train_epochs": 5,
   "save_steps": 500,
-  "total_flos": 2922545822700.0,
+  "total_flos": 2905718091150.0,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": {
-    "learning_rate": 1.946279842307678e-05,
-    "num_train_epochs": 4,
+    "learning_rate": 8.706073317281892e-06,
+    "num_train_epochs": 5,
     "per_device_train_batch_size": 4,
-    "seed": 23
+    "seed": 1
   }
 }

runs/Jan15_00-51-52_DESKTOP-RFEED41/events.out.tfevents.1705269786.DESKTOP-RFEED41.20976.18

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4578bfa83675a3cce33be98262b633d695f981f70048612e5dda786f2becb1cc
-size 5430
+oid sha256:335d341f4f46113a5629ed21e06e5ea529e9c101dd7cacbc2d2cd91116247770
+size 6390