End of training

all_results.json

@@ -1,14 +1,14 @@
 {
-    "epoch": 2.0,
-    "eval_accuracy": 0.7986,
-    "eval_loss": 0.33286044001579285,
-    "eval_runtime": 151.7682,
+    "epoch": 3.0,
+    "eval_accuracy": 0.8023,
+    "eval_loss": 0.438219279050827,
+    "eval_runtime": 146.5968,
     "eval_samples": 4887,
-    "eval_samples_per_second": 32.2,
-    "eval_steps_per_second": 2.016,
-    "train_loss": 0.0,
-    "train_runtime": 0.057,
+    "eval_samples_per_second": 33.336,
+    "eval_steps_per_second": 2.087,
+    "train_loss": 0.0359802827722491,
+    "train_runtime": 7164.0026,
     "train_samples": 87866,
-    "train_samples_per_second": 1541330.369,
-    "train_steps_per_second": 192661.911
+    "train_samples_per_second": 36.795,
+    "train_steps_per_second": 4.599
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 2.0,
-    "eval_accuracy": 0.7986,
-    "eval_loss": 0.33286044001579285,
-    "eval_runtime": 151.7682,
+    "epoch": 3.0,
+    "eval_accuracy": 0.8023,
+    "eval_loss": 0.438219279050827,
+    "eval_runtime": 146.5968,
     "eval_samples": 4887,
-    "eval_samples_per_second": 32.2,
-    "eval_steps_per_second": 2.016
+    "eval_samples_per_second": 33.336,
+    "eval_steps_per_second": 2.087
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5c129de941d14268599c41de3b73b6af275457445a117b4fb84eae464de58f04
+oid sha256:35721e0aa79f878718ecc57475a889391a1d6b6404c863fc4fc0d17a8d9ca7c3
 size 2950790023

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.0,
-    "train_loss": 0.0,
-    "train_runtime": 0.057,
+    "epoch": 3.0,
+    "train_loss": 0.0359802827722491,
+    "train_runtime": 7164.0026,
     "train_samples": 87866,
-    "train_samples_per_second": 1541330.369,
-    "train_steps_per_second": 192661.911
+    "train_samples_per_second": 36.795,
+    "train_steps_per_second": 4.599
 }

trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.9999544771703008,
-  "global_step": 21966,
+  "epoch": 2.999954477170301,
+  "global_step": 32949,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -285,17 +285,158 @@
     },
     {
       "epoch": 2.0,
-      "step": 21966,
-      "total_flos": 3.80468091420672e+17,
-      "train_loss": 0.0,
-      "train_runtime": 0.057,
-      "train_samples_per_second": 1541330.369,
-      "train_steps_per_second": 192661.911
+      "learning_rate": 3.3331303844865904e-05,
+      "loss": 0.138,
+      "step": 22000
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 3.1809187494292065e-05,
+      "loss": 0.104,
+      "step": 22500
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 3.0287071143718226e-05,
+      "loss": 0.112,
+      "step": 23000
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 2.876495479314439e-05,
+      "loss": 0.1109,
+      "step": 23500
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 2.724283844257055e-05,
+      "loss": 0.1059,
+      "step": 24000
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 2.5720722091996712e-05,
+      "loss": 0.1048,
+      "step": 24500
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 2.4198605741422876e-05,
+      "loss": 0.1067,
+      "step": 25000
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 2.2676489390849037e-05,
+      "loss": 0.1168,
+      "step": 25500
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 2.11543730402752e-05,
+      "loss": 0.1084,
+      "step": 26000
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 1.9632256689701363e-05,
+      "loss": 0.1146,
+      "step": 26500
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 1.8110140339127524e-05,
+      "loss": 0.107,
+      "step": 27000
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 1.6588023988553685e-05,
+      "loss": 0.1107,
+      "step": 27500
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 1.5065907637979847e-05,
+      "loss": 0.1087,
+      "step": 28000
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 1.354379128740601e-05,
+      "loss": 0.1109,
+      "step": 28500
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 1.2021674936832173e-05,
+      "loss": 0.101,
+      "step": 29000
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 1.0499558586258334e-05,
+      "loss": 0.1066,
+      "step": 29500
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 8.977442235684496e-06,
+      "loss": 0.1092,
+      "step": 30000
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 7.455325885110659e-06,
+      "loss": 0.1006,
+      "step": 30500
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 5.93320953453682e-06,
+      "loss": 0.1155,
+      "step": 31000
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 4.411093183962983e-06,
+      "loss": 0.1204,
+      "step": 31500
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 2.8889768333891442e-06,
+      "loss": 0.0972,
+      "step": 32000
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 1.3668604828153065e-06,
+      "loss": 0.0929,
+      "step": 32500
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8023,
+      "eval_loss": 0.438219279050827,
+      "eval_runtime": 146.5435,
+      "eval_samples_per_second": 33.348,
+      "eval_steps_per_second": 2.088,
+      "step": 32949
+    },
+    {
+      "epoch": 3.0,
+      "step": 32949,
+      "total_flos": 5.70697807036416e+17,
+      "train_loss": 0.0359802827722491,
+      "train_runtime": 7164.0026,
+      "train_samples_per_second": 36.795,
+      "train_steps_per_second": 4.599
     }
   ],
-  "max_steps": 10983,
-  "num_train_epochs": 1,
-  "total_flos": 3.80468091420672e+17,
+  "max_steps": 32949,
+  "num_train_epochs": 3,
+  "total_flos": 5.70697807036416e+17,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a2565c39287ef9a560b024aded0fa87f7dd9a636e2ef6ff73357484ea0a21c7
+oid sha256:4d00b7d9305a03ec44724d87e1c60fe5cfecdc9a609742300dd766c2e7395c56
 size 2927