Training in progress, epoch 2

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:78a0904d05584058bc1c93a27fc67433b942291b1b39acfc2f2505e394fa31ac
+oid sha256:fc2934c19e3f145b2c8af3a46df740321a95d0b384d33395bbea1d2f8f05e99a
 size 409103316

run-0/checkpoint-1066/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7755e65b6162f8636a9241f6923420b193ca30b3a5dca7418b51a2bcc380f86d
+oid sha256:fc2934c19e3f145b2c8af3a46df740321a95d0b384d33395bbea1d2f8f05e99a
 size 409103316

run-0/checkpoint-1066/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:63d963cf3cbd7391c62eb322b08595cc0bd5b707c4533beeae4a064153966de0
+oid sha256:f6a4292fbcc3bbbd0b7905857012c6060a4fdc03a263336f02ef5ecd3b2c2742
 size 818327802

run-0/checkpoint-1066/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4fbf55a5bf069992fe6cfbfae4a696b1cd7d93685a9743ad6546c3030830f638
+oid sha256:b0531b49646ec50dbec094636163f9366944098e7fa07140b9e91e313783c4ea
 size 1064

run-0/checkpoint-1066/trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.6369415521621704,
-  "best_model_checkpoint": "BERT-WMM/run-0/checkpoint-533",
+  "best_metric": 0.791917085647583,
+  "best_model_checkpoint": "BERT-WMM/run-0/checkpoint-1066",
   "epoch": 2.0,
   "eval_steps": 500,
   "global_step": 1066,
@@ -10,49 +10,46 @@
   "log_history": [
     {
       "epoch": 0.94,
-      "grad_norm": 11.681629180908203,
-      "learning_rate": 2.5842410540171958e-05,
-      "loss": 0.7181,
+      "grad_norm": 10.492278099060059,
+      "learning_rate": 7.698774119953696e-07,
+      "loss": 0.9956,
       "step": 500
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7525821596244131,
-      "eval_loss": 0.6369415521621704,
-      "eval_runtime": 1.9836,
-      "eval_samples_per_second": 1073.792,
-      "eval_steps_per_second": 67.553,
+      "eval_accuracy": 0.6281690140845071,
+      "eval_loss": 0.8787437081336975,
+      "eval_runtime": 2.0952,
+      "eval_samples_per_second": 1016.614,
+      "eval_steps_per_second": 63.956,
       "step": 533
     },
     {
       "epoch": 1.88,
-      "grad_norm": 6.492458820343018,
-      "learning_rate": 2.1053231418568556e-05,
-      "loss": 0.4571,
+      "grad_norm": 16.712677001953125,
+      "learning_rate": 4.1961471318036977e-07,
+      "loss": 0.858,
       "step": 1000
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.7497652582159624,
-      "eval_loss": 0.6769182682037354,
-      "eval_runtime": 2.0259,
-      "eval_samples_per_second": 1051.385,
-      "eval_steps_per_second": 66.144,
+      "eval_accuracy": 0.6732394366197183,
+      "eval_loss": 0.791917085647583,
+      "eval_runtime": 2.1109,
+      "eval_samples_per_second": 1009.044,
+      "eval_steps_per_second": 63.48,
       "step": 1066
     }
   ],
   "logging_steps": 500,
-  "max_steps": 3198,
+  "max_steps": 1599,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 6,
+  "num_train_epochs": 3,
   "save_steps": 500,
   "total_flos": 338261076519408.0,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": {
-    "learning_rate": 3.063158966177536e-05,
-    "num_train_epochs": 6,
-    "per_device_train_batch_size": 16,
-    "weight_decay": 2.6426532465921284e-05
+    "learning_rate": 1.1201401108103694e-06
   }
 }

run-0/checkpoint-1066/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48c24530a27c9d1e90a961cc0645745786b44285899342b0769165200ee8a2b1
+oid sha256:f3b32ee6920eea73a8ff94d684bfa22384952a503144760a3a67dfa2dfe3f39f
 size 4856

runs/Apr18_16-31-07_544fc269209b/events.out.tfevents.1713457870.544fc269209b.792.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71e100b62b72129822e6624a876e5247b23379a95320e68e4e423bf875bc6df1
-size 5428
+oid sha256:8aea9294fd5b07cc0e83aa6ecc325232e8f25d9c6601fc8a5d6d9e88eee0356f
+size 5962