Training in progress, step 3000, checkpoint

checkpoint-3000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9bd8b8d005c8498d0dc87f4cc00ffebdf7a4f1c4fbaf9c06374d9b4bb13b658f
+oid sha256:de7d6f6157cff9810cf8ea9f033c4fd681347531ec68cf04dea329e43b0c4db8
 size 932401285

checkpoint-3000/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c1404200e7060273f92a189004bc61a4f6d9ed1be64b6433f51dc931e1da4576
+oid sha256:e0e01d89e2c093d6586b64996b2d730467c6eec8ba6acb4fbf7c776362f563fe
 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": 1.1441665887832642,
+  "best_metric": 1.127414345741272,
   "best_model_checkpoint": "/kaggle/output/checkpoint-3000",
-  "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": 1.2395,
+      "loss": 1.029,
       "step": 1
     },
     {
-      "epoch": 16.0,
+      "epoch": 0.04,
       "learning_rate": 2.7638888888888893e-08,
-      "loss": 1.1422,
+      "loss": 1.181,
       "step": 1000
     },
     {
-      "epoch": 16.0,
-      "eval_accuracy": 0.33073852295409184,
-      "eval_loss": 1.1557353734970093,
-      "eval_runtime": 54.1805,
-      "eval_samples_per_second": 92.469,
-      "eval_steps_per_second": 11.572,
+      "epoch": 0.04,
+      "eval_accuracy": 0.3273453093812375,
+      "eval_loss": 1.1529844999313354,
+      "eval_runtime": 54.2837,
+      "eval_samples_per_second": 92.293,
+      "eval_steps_per_second": 11.55,
       "step": 1000
     },
     {
-      "epoch": 32.0,
+      "epoch": 0.08,
       "learning_rate": 5.541666666666667e-08,
-      "loss": 1.1401,
+      "loss": 1.1527,
       "step": 2000
     },
     {
-      "epoch": 32.0,
-      "eval_accuracy": 0.3311377245508982,
-      "eval_loss": 1.1507551670074463,
-      "eval_runtime": 53.98,
-      "eval_samples_per_second": 92.812,
-      "eval_steps_per_second": 11.615,
+      "epoch": 0.08,
+      "eval_accuracy": 0.33013972055888224,
+      "eval_loss": 1.1351025104522705,
+      "eval_runtime": 54.0518,
+      "eval_samples_per_second": 92.689,
+      "eval_steps_per_second": 11.6,
       "step": 2000
     },
     {
-      "epoch": 48.0,
+      "epoch": 0.12,
       "learning_rate": 8.316666666666666e-08,
-      "loss": 1.1287,
+      "loss": 1.142,
       "step": 3000
     },
     {
-      "epoch": 48.0,
-      "eval_accuracy": 0.3321357285429142,
-      "eval_loss": 1.1441665887832642,
-      "eval_runtime": 53.9873,
-      "eval_samples_per_second": 92.8,
-      "eval_steps_per_second": 11.614,
+      "epoch": 0.12,
+      "eval_accuracy": 0.3317365269461078,
+      "eval_loss": 1.127414345741272,
+      "eval_runtime": 54.0871,
+      "eval_samples_per_second": 92.628,
+      "eval_steps_per_second": 11.592,
       "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:cb2cd1b7e648a386ccb79e47a451842e11001981c4363e993b678e0818db5cab
+oid sha256:5520446b18da09f4a43173b75839457981a344c0e709fd4bbcb249e3f60c859e
 size 4091