Training in progress, step 3000, checkpoint

checkpoint-3000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a66e7117cc7cced67377346b48be7fc8cab5d5a24f6a93e246b6a4182bcef639
+oid sha256:904c7d84748e09d61f3b730f98afad78f33d270c29b571007a6320cc7e607321
 size 932401285

checkpoint-3000/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:da871d1621b267fecfa6c214edcb594366de290865521c183853dbe07d7f5389
+oid sha256:77f1fd2675cc86a12e06ab663c3f9fb10c3dc1cdb1e4ef87fae7ef5af7873dc3
 size 466205145

checkpoint-3000/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:73a860bf7adbc7134c78671fd5c6e6588c8b56d0bda60913c03b8c144c3849d7
+oid sha256:7249c42f48d99a7463b035eeb788a527848411ab95b413e6f1032df9ad08c927
 size 14575

checkpoint-3000/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 1.1078879833221436,
+  "best_metric": 1.1077626943588257,
   "best_model_checkpoint": "/kaggle/output/checkpoint-1000",
-  "epoch": 48.0,
+  "epoch": 0.12222946544980444,
   "eval_steps": 1000,
   "global_step": 3000,
   "is_hyper_param_search": false,
@@ -9,60 +9,60 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.02,
+      "epoch": 0.0,
       "learning_rate": 2.7777777777777777e-11,
-      "loss": 1.5477,
+      "loss": 1.1383,
       "step": 1
     },
     {
-      "epoch": 16.0,
+      "epoch": 0.04,
       "learning_rate": 2.7750000000000004e-08,
-      "loss": 1.2028,
+      "loss": 1.1424,
       "step": 1000
     },
     {
-      "epoch": 16.0,
-      "eval_accuracy": 0.3387225548902196,
-      "eval_loss": 1.1078879833221436,
-      "eval_runtime": 54.1057,
-      "eval_samples_per_second": 92.596,
-      "eval_steps_per_second": 11.588,
+      "epoch": 0.04,
+      "eval_accuracy": 0.32375249500998005,
+      "eval_loss": 1.1077626943588257,
+      "eval_runtime": 54.8633,
+      "eval_samples_per_second": 91.318,
+      "eval_steps_per_second": 11.428,
       "step": 1000
     },
     {
-      "epoch": 32.0,
+      "epoch": 0.08,
       "learning_rate": 5.5527777777777784e-08,
-      "loss": 1.1152,
+      "loss": 1.1244,
       "step": 2000
     },
     {
-      "epoch": 32.0,
-      "eval_accuracy": 0.3349301397205589,
-      "eval_loss": 1.1080944538116455,
-      "eval_runtime": 54.2029,
-      "eval_samples_per_second": 92.43,
-      "eval_steps_per_second": 11.568,
+      "epoch": 0.08,
+      "eval_accuracy": 0.33652694610778444,
+      "eval_loss": 1.1080161333084106,
+      "eval_runtime": 54.7384,
+      "eval_samples_per_second": 91.526,
+      "eval_steps_per_second": 11.454,
       "step": 2000
     },
     {
-      "epoch": 48.0,
+      "epoch": 0.12,
       "learning_rate": 8.327777777777778e-08,
-      "loss": 1.1036,
+      "loss": 1.1228,
       "step": 3000
     },
     {
-      "epoch": 48.0,
-      "eval_accuracy": 0.3377245508982036,
-      "eval_loss": 1.1081210374832153,
-      "eval_runtime": 54.0467,
-      "eval_samples_per_second": 92.698,
-      "eval_steps_per_second": 11.601,
+      "epoch": 0.12,
+      "eval_accuracy": 0.34331337325349304,
+      "eval_loss": 1.1084064245224,
+      "eval_runtime": 54.7948,
+      "eval_samples_per_second": 91.432,
+      "eval_steps_per_second": 11.443,
       "step": 3000
     }
   ],
   "logging_steps": 1000,
   "max_steps": 10000000,
-  "num_train_epochs": 161291,
+  "num_train_epochs": 408,
   "save_steps": 1000,
   "total_flos": 6271065391104000.0,
   "trial_name": null,

checkpoint-3000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6c09d25f832c6c38e287ec7a7fe1b191faf603815bb3b91e32f16768a2d7d3ff
+oid sha256:a9a52921b15d7eeb53b158a9c0f943d88288ccd9232004774ffdf3a600a048d5
 size 4091