Training in progress, step 20, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3a5c9c0a64ab8fdbae53dac1dd1f9e56b81ae97f263cf783b5a88524601126b2
+oid sha256:b5b001a2fc939ecf43fb0d86edcd5bf104abdef1bc3131e002054358dce29e25
 size 69844690

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b272a051d5e0f0cd1ed93bd213f58b61bcbf1db65bbdb62e56bf44bf5762a15c
+oid sha256:450126e07c0c8b1791f3f3aa2cf6e5120163749c0d302c76afbdee1feab14888
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bb578e75c11a81e85dda67a691f96ba4793a02960f1409fd3e1511aac873491a
+oid sha256:1d95f3fb1f9df4cdd5f470d1684a242bf1d1940d47b9622802f603a91ffa5bc0
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.004299226139294927,
+  "epoch": 0.008598452278589854,
   "eval_steps": 7,
-  "global_step": 10,
+  "global_step": 20,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -44,6 +44,35 @@
       "learning_rate": 0.00018,
       "loss": 0.0,
       "step": 9
+    },
+    {
+      "epoch": 0.005159071367153913,
+      "grad_norm": NaN,
+      "learning_rate": 0.0001913545457642601,
+      "loss": 0.0,
+      "step": 12
+    },
+    {
+      "epoch": 0.0060189165950128975,
+      "eval_loss": NaN,
+      "eval_runtime": 35.6499,
+      "eval_samples_per_second": 27.49,
+      "eval_steps_per_second": 13.745,
+      "step": 14
+    },
+    {
+      "epoch": 0.006448839208942391,
+      "grad_norm": NaN,
+      "learning_rate": 0.00015000000000000001,
+      "loss": 0.0,
+      "step": 15
+    },
+    {
+      "epoch": 0.007738607050730868,
+      "grad_norm": NaN,
+      "learning_rate": 8.954715367323468e-05,
+      "loss": 0.0,
+      "step": 18
     }
   ],
   "logging_steps": 3,
@@ -63,7 +92,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 170836438548480.0,
+  "total_flos": 345645817528320.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null