Training in progress, step 500, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:98fd124bf42aea777816d09325e7ffdff5b493ffa0dd09d7c3a34559fbbc57a1
+oid sha256:e5f8208969c787c86b244b9e9a71626f98e35ffdfeaecbcc5ef207a5ddc9659d
 size 386379

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8db1924741cee0f05554aeca4688871e1e114cabeb2ea4420e919cde46a884c9
+oid sha256:965dc697c3ab28072f197959f194ef0efcc91bb79803752fd97f500aa7e8a5df
 size 1540661735

last-checkpoint/trainer_state.json

@@ -11,82 +11,82 @@
   "log_history": [
     {
       "epoch": 0.1,
-      "grad_norm": 2.3951845169067383,
+      "grad_norm": 2.191904306411743,
       "learning_rate": 0.0003973800426880847,
-      "loss": 9.800797119140626,
+      "loss": 9.861549072265625,
       "step": 100
     },
     {
       "epoch": 0.1,
-      "eval_accuracy": 0.09641291585127201,
-      "eval_loss": 9.091407775878906,
-      "eval_runtime": 39.6773,
-      "eval_samples_per_second": 25.203,
-      "eval_steps_per_second": 25.203,
+      "eval_accuracy": 0.09354794520547945,
+      "eval_loss": 9.128084182739258,
+      "eval_runtime": 32.2982,
+      "eval_samples_per_second": 30.962,
+      "eval_steps_per_second": 30.962,
       "step": 100
     },
     {
       "epoch": 0.2,
-      "grad_norm": 2.5533459186553955,
+      "grad_norm": 1.9173266887664795,
       "learning_rate": 0.0003762085737488283,
-      "loss": 8.910618286132813,
+      "loss": 8.851971435546876,
       "step": 200
     },
     {
       "epoch": 0.2,
-      "eval_accuracy": 0.09622896281800392,
-      "eval_loss": 8.752147674560547,
-      "eval_runtime": 33.1575,
-      "eval_samples_per_second": 30.159,
-      "eval_steps_per_second": 30.159,
+      "eval_accuracy": 0.10289236790606654,
+      "eval_loss": 8.73751163482666,
+      "eval_runtime": 26.6022,
+      "eval_samples_per_second": 37.591,
+      "eval_steps_per_second": 37.591,
       "step": 200
     },
     {
       "epoch": 0.3,
-      "grad_norm": 2.921639919281006,
+      "grad_norm": 2.392211437225342,
       "learning_rate": 0.00033594217168615465,
-      "loss": 8.53966796875,
+      "loss": 8.558801879882813,
       "step": 300
     },
     {
       "epoch": 0.3,
-      "eval_accuracy": 0.0970450097847358,
-      "eval_loss": 8.573400497436523,
-      "eval_runtime": 31.8348,
-      "eval_samples_per_second": 31.412,
-      "eval_steps_per_second": 31.412,
+      "eval_accuracy": 0.10008610567514677,
+      "eval_loss": 8.624308586120605,
+      "eval_runtime": 26.8044,
+      "eval_samples_per_second": 37.307,
+      "eval_steps_per_second": 37.307,
       "step": 300
     },
     {
       "epoch": 0.4,
-      "grad_norm": 2.1571640968322754,
+      "grad_norm": 3.655690908432007,
       "learning_rate": 0.00028094432596115747,
-      "loss": 8.399310302734374,
+      "loss": 8.405433349609375,
       "step": 400
     },
     {
       "epoch": 0.4,
-      "eval_accuracy": 0.10219373776908024,
-      "eval_loss": 8.501660346984863,
-      "eval_runtime": 32.4742,
-      "eval_samples_per_second": 30.794,
-      "eval_steps_per_second": 30.794,
+      "eval_accuracy": 0.10092172211350293,
+      "eval_loss": 8.508211135864258,
+      "eval_runtime": 26.8389,
+      "eval_samples_per_second": 37.259,
+      "eval_steps_per_second": 37.259,
       "step": 400
     },
     {
       "epoch": 0.5,
-      "grad_norm": 2.42037034034729,
+      "grad_norm": 8.362494468688965,
       "learning_rate": 0.00021717490653764342,
-      "loss": 8.401516723632813,
+      "loss": 8.498648681640624,
       "step": 500
     },
     {
       "epoch": 0.5,
-      "eval_accuracy": 0.10070058708414872,
-      "eval_loss": 8.452370643615723,
-      "eval_runtime": 32.4058,
-      "eval_samples_per_second": 30.859,
-      "eval_steps_per_second": 30.859,
+      "eval_accuracy": 0.07594911937377691,
+      "eval_loss": 9.306892395019531,
+      "eval_runtime": 26.9571,
+      "eval_samples_per_second": 37.096,
+      "eval_steps_per_second": 37.096,
       "step": 500
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f0098da0a3455c5880575320ae51291e35e8d0961f89a145f0520da235cee9b
+oid sha256:b47eb6e798b2dfe4e0fdea550782a8eb56b1b579932ae1b65db9c8ecf145627d
 size 5329