Training in progress, epoch 1

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:01943b36fe2d099072ef77b2811221d172183457a8423ec8462545bbe2d496e3
+oid sha256:fa79b65d0ba7529aee0550dc910f9468bf4b0ad08164867f686e834e0e9fd408
 size 409103316

run-3/checkpoint-534/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fed076af4c274336a1bbda01e7b9e95519d0bf0a1fe5c7f712970d2aa901f093
+oid sha256:61096730720c51ac6648ccb652d6e8883705568c6be1662245a2d864a0f89450
 size 409103316

run-3/checkpoint-534/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b2699e5ce5a66cbca0b105d6408b39db28914becd331e717afd744ad8431c462
+oid sha256:01d1170bce624b8918138e8cc63259f0004659314d431d193f9b965250a6158b
 size 818327802

run-3/checkpoint-534/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9844fc449bc7d92d10fa8131195695656fc5bae595590761350f9f767121bb98
+oid sha256:8a56050495e1938ae70b239ae8ece3c43ac800b58f0a6e02bd7ed6cd1669797d
 size 1064

run-3/checkpoint-534/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.5784130692481995,
+  "best_metric": 0.6037538051605225,
   "best_model_checkpoint": "BERT-WMM/run-3/checkpoint-534",
   "epoch": 2.0,
   "eval_steps": 500,
@@ -10,39 +10,39 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7549295774647887,
-      "eval_loss": 0.603171169757843,
-      "eval_runtime": 2.3997,
-      "eval_samples_per_second": 887.602,
-      "eval_steps_per_second": 55.84,
+      "eval_accuracy": 0.7530516431924883,
+      "eval_loss": 0.6065873503684998,
+      "eval_runtime": 2.5921,
+      "eval_samples_per_second": 821.724,
+      "eval_steps_per_second": 51.695,
       "step": 267
     },
     {
       "epoch": 1.87,
-      "grad_norm": 9.50742244720459,
-      "learning_rate": 8.206819746059557e-06,
-      "loss": 0.6057,
+      "grad_norm": 9.447236061096191,
+      "learning_rate": 9.825065893169892e-07,
+      "loss": 0.6083,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.7676056338028169,
-      "eval_loss": 0.5784130692481995,
-      "eval_runtime": 2.4488,
-      "eval_samples_per_second": 869.828,
-      "eval_steps_per_second": 54.722,
+      "eval_accuracy": 0.7615023474178404,
+      "eval_loss": 0.6037538051605225,
+      "eval_runtime": 2.5455,
+      "eval_samples_per_second": 836.779,
+      "eval_steps_per_second": 52.642,
       "step": 534
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1068,
+  "max_steps": 534,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 4,
+  "num_train_epochs": 2,
   "save_steps": 500,
   "total_flos": 349221394826640.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "num_train_epochs": 4
+    "weight_decay": 0.021167311357718124
   }
 }

run-3/checkpoint-534/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8c637656beacaec83a46c1d786535828a5c615851f4d21b0069fcadd723fa83
+oid sha256:1238f3e9130ff13042dcb9e0ded88b87b25d59c0eb7bc889417ea846582c3c9b
 size 4856

run-4/checkpoint-267/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:709e0174af7e7bfc36b2ca42177cb07663d435c4830e3ef2fd95156cc7684138
+oid sha256:ca0dc6ea72694ff54a88f458db9aff463c496bd6d83166f6fb57a4e51c5bb172
 size 409103316

run-4/checkpoint-267/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:52ef4431d2c45c5021200b1e749d9a863141163026b276944e492c1db9ee3f59
+oid sha256:b5921b1879136a2de0f32be39d0aa1e2c1df44b7767609fbb8ab7984ca270865
 size 818327802

run-4/checkpoint-267/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44c8fc955f6ab66538b705aa40f6ebb08474ec445464a7a19ca29971b9fc42d4
+oid sha256:ef3be8be9037b2d6a9348611f95c19fb1d17ca7cb1da92486df22b7e0a3c655a
 size 1064

run-4/checkpoint-267/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.6081846952438354,
+  "best_metric": 0.6065725684165955,
   "best_model_checkpoint": "BERT-WMM/run-4/checkpoint-267",
   "epoch": 1.0,
   "eval_steps": 500,
@@ -10,23 +10,23 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7577464788732394,
-      "eval_loss": 0.6081846952438354,
-      "eval_runtime": 2.4392,
-      "eval_samples_per_second": 873.231,
-      "eval_steps_per_second": 54.936,
+      "eval_accuracy": 0.7530516431924883,
+      "eval_loss": 0.6065725684165955,
+      "eval_runtime": 2.5417,
+      "eval_samples_per_second": 838.036,
+      "eval_steps_per_second": 52.722,
       "step": 267
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1335,
+  "max_steps": 534,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 2,
   "save_steps": 500,
   "total_flos": 0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "num_train_epochs": 5
+    "weight_decay": 4.018191065683781e-05
   }
 }

run-4/checkpoint-267/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d42e5e5ec734e47895bff309ce0f98b97c457fc4dc71b42d53c94016f0c7a855
+oid sha256:61f2b8ff5812e9a8e681e14d8d4a502beb7efad66683c16ee39765361b35b768
 size 4856

run-4/checkpoint-534/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:976557a36bef1a4a492485f3f6cb8948205b7b3a9ea05ce8ce0eb8982fcc1d9e
+oid sha256:b5b3ada415323df4681659f8a3c19121e9379a2b0c7922cade957c3cc9526017
 size 409103316

run-4/checkpoint-534/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:890b6f5cc14af75c85002de6e2745609f1bf02f4a33a71cd0af847d31243fe8a
+oid sha256:492785e0b2d4bc2fddbf03fc7aa5070a9d5992be16a4b8b568d67b7ad8b49cd5
 size 818327802

run-4/checkpoint-534/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5e2aa06609570d504e0cfaf6c3bd801e1681706921a7512c6ccb9eded784901f
+oid sha256:8a56050495e1938ae70b239ae8ece3c43ac800b58f0a6e02bd7ed6cd1669797d
 size 1064

run-4/checkpoint-534/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.5776590704917908,
+  "best_metric": 0.6037774682044983,
   "best_model_checkpoint": "BERT-WMM/run-4/checkpoint-534",
   "epoch": 2.0,
   "eval_steps": 500,
@@ -10,39 +10,39 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7577464788732394,
-      "eval_loss": 0.6081846952438354,
-      "eval_runtime": 2.4392,
-      "eval_samples_per_second": 873.231,
-      "eval_steps_per_second": 54.936,
+      "eval_accuracy": 0.7530516431924883,
+      "eval_loss": 0.6065725684165955,
+      "eval_runtime": 2.5417,
+      "eval_samples_per_second": 838.036,
+      "eval_steps_per_second": 52.722,
       "step": 267
     },
     {
       "epoch": 1.87,
-      "grad_norm": 9.532360076904297,
-      "learning_rate": 9.65168237740807e-06,
-      "loss": 0.6057,
+      "grad_norm": 9.447751998901367,
+      "learning_rate": 9.825065893169892e-07,
+      "loss": 0.6083,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.7657276995305164,
-      "eval_loss": 0.5776590704917908,
-      "eval_runtime": 2.4594,
-      "eval_samples_per_second": 866.069,
-      "eval_steps_per_second": 54.485,
+      "eval_accuracy": 0.7615023474178404,
+      "eval_loss": 0.6037774682044983,
+      "eval_runtime": 2.5779,
+      "eval_samples_per_second": 826.263,
+      "eval_steps_per_second": 51.981,
       "step": 534
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1335,
+  "max_steps": 534,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 2,
   "save_steps": 500,
   "total_flos": 349221394826640.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "num_train_epochs": 5
+    "weight_decay": 4.018191065683781e-05
   }
 }

run-4/checkpoint-534/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d42e5e5ec734e47895bff309ce0f98b97c457fc4dc71b42d53c94016f0c7a855
+oid sha256:61f2b8ff5812e9a8e681e14d8d4a502beb7efad66683c16ee39765361b35b768
 size 4856

run-5/checkpoint-267/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:971a3ddf817b712dc07195c844772ef5d1ec4a8fd2faa3cfc511b714130228d0
+oid sha256:fa79b65d0ba7529aee0550dc910f9468bf4b0ad08164867f686e834e0e9fd408
 size 409103316

run-5/checkpoint-267/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cc8b54fc5e8c9423b8eb78032472c72dce9ff03db39846d1622f31a3d6911c46
+oid sha256:d47a133704ee5d893b30a7dc35c9b1d607c0d15dd4e32e8a90cf3cdd87699884
 size 818327802

run-5/checkpoint-267/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d605219d0cdfe1356d19dfc261c4f045664dbc44ddf04517d0b2ebd1dab264c7
+oid sha256:ef3be8be9037b2d6a9348611f95c19fb1d17ca7cb1da92486df22b7e0a3c655a
 size 1064

run-5/checkpoint-267/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.6019130349159241,
+  "best_metric": 0.6066888570785522,
   "best_model_checkpoint": "BERT-WMM/run-5/checkpoint-267",
   "epoch": 1.0,
   "eval_steps": 500,
@@ -10,23 +10,23 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7563380281690141,
-      "eval_loss": 0.6019130349159241,
-      "eval_runtime": 2.4929,
-      "eval_samples_per_second": 854.419,
-      "eval_steps_per_second": 53.752,
+      "eval_accuracy": 0.7539906103286385,
+      "eval_loss": 0.6066888570785522,
+      "eval_runtime": 2.5768,
+      "eval_samples_per_second": 826.614,
+      "eval_steps_per_second": 52.003,
       "step": 267
     }
   ],
   "logging_steps": 500,
-  "max_steps": 801,
+  "max_steps": 534,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 3,
+  "num_train_epochs": 2,
   "save_steps": 500,
   "total_flos": 0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "per_device_train_batch_size": 32
+    "weight_decay": 0.08935453780478574
   }
 }

run-5/checkpoint-267/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9cc12ec9008fe3ca8c11132361abe51e0525170497895a06dfb1609f916fd77f
+oid sha256:b57316b4d399893de9d059984f406888c25d2a6d5ffb20848fef7218735cf9bd
 size 4856

runs/Apr19_10-19-01_dcdaa3e6ec43/events.out.tfevents.1713522305.dcdaa3e6ec43.789.3

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5908acf12b09c9364f37fdcc1828450a70123a37e71027bfb4acc38d127db0d0
-size 5233
+oid sha256:5988cc2775814fe44975831f32c32caf7771d6edd2d5bc5052cdfabafbbe56d1
+size 6121

runs/Apr19_10-19-01_dcdaa3e6ec43/events.out.tfevents.1713522401.dcdaa3e6ec43.789.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fba934ae41c63dd7574183a022397f725e91e4705deaa7cfb417b01fd8f39be9
+size 6122

runs/Apr19_10-19-01_dcdaa3e6ec43/events.out.tfevents.1713522502.dcdaa3e6ec43.789.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb70d05343967559454757ce9e9c2dca96ac89f196cec4d89f868a35bc3f8c21
+size 5766

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1238f3e9130ff13042dcb9e0ded88b87b25d59c0eb7bc889417ea846582c3c9b
+oid sha256:b57316b4d399893de9d059984f406888c25d2a6d5ffb20848fef7218735cf9bd
 size 4856