Training in progress, step 40000

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5afbb7db2024c06e7168cd07934240d81acfb417a962c3f9b27c2c5cac74f5f3
+oid sha256:b937992da12590ba2ee2ecfa5a6224eb75cd0a6dbf99f744bec6042034768fdb
 size 893438545

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:795fa208c4655907c54892ea46d5deef3057c257b0f29b327c60d57aebb752ef
+oid sha256:fe032267303d174768a4c71a5b4ab969ea1dd84441a01822ea3e2ec637af7091
 size 449471589

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:607b5c5487d5dccaf4d249f748d9be7bab5e418ed2dbdf5e21e48a8763bb2292
+oid sha256:6b9bf566fdf421530052a690839f4abb9b1083bf68461ba305086568aa88aeb4
 size 15587

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e833ce90acecedfbd8d212ba971de16be51f612497919359e2315f9ed22fdccf
+oid sha256:1bb55876e03a71e55e428a60dd59e185d61d2bbf026ccff833af7b992fa10a9a
 size 559

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:86e9753a509961ebb704900ef39dcda0523410f768cfe662fee4e7f583c2197f
+oid sha256:a7c3bc66ef595b060943f43f042dfbc9f45a9f2c4def064a14535b0336d42b5a
 size 623

last-checkpoint/trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 3.047187880898485,
-  "global_step": 35000,
+  "epoch": 3.4825004353125544,
+  "global_step": 40000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -482,11 +482,79 @@
       "eval_samples_per_second": 296.933,
       "eval_steps_per_second": 4.692,
       "step": 35000
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 7.537638471730726e-05,
+      "loss": 0.4281,
+      "step": 35500
+    },
+    {
+      "epoch": 3.13,
+      "learning_rate": 7.473378509304684e-05,
+      "loss": 0.426,
+      "step": 36000
+    },
+    {
+      "epoch": 3.18,
+      "learning_rate": 7.408641283347138e-05,
+      "loss": 0.4266,
+      "step": 36500
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 7.343442766799908e-05,
+      "loss": 0.4269,
+      "step": 37000
+    },
+    {
+      "epoch": 3.26,
+      "learning_rate": 7.27779904642137e-05,
+      "loss": 0.4266,
+      "step": 37500
+    },
+    {
+      "epoch": 3.31,
+      "learning_rate": 7.211726318817271e-05,
+      "loss": 0.427,
+      "step": 38000
+    },
+    {
+      "epoch": 3.35,
+      "learning_rate": 7.145240886444487e-05,
+      "loss": 0.4254,
+      "step": 38500
+    },
+    {
+      "epoch": 3.4,
+      "learning_rate": 7.078493301609684e-05,
+      "loss": 0.4262,
+      "step": 39000
+    },
+    {
+      "epoch": 3.44,
+      "learning_rate": 7.011232513402331e-05,
+      "loss": 0.4264,
+      "step": 39500
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 6.943744088754813e-05,
+      "loss": 0.4258,
+      "step": 40000
+    },
+    {
+      "epoch": 3.48,
+      "eval_loss": 0.4074101150035858,
+      "eval_runtime": 18.107,
+      "eval_samples_per_second": 276.137,
+      "eval_steps_per_second": 4.363,
+      "step": 40000
     }
   ],
   "max_steps": 100000,
   "num_train_epochs": 9,
-  "total_flos": 1.648395610407119e+21,
+  "total_flos": 1.8838886777718807e+21,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:795fa208c4655907c54892ea46d5deef3057c257b0f29b327c60d57aebb752ef
+oid sha256:fe032267303d174768a4c71a5b4ab969ea1dd84441a01822ea3e2ec637af7091
 size 449471589