Training in progress, step 20, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:94e1b261f896b2732d77065096c5f36bec71f098215a806d86f6a53b24ec40d6
+oid sha256:e12ef96034ad2097f1f42387fd4576db0d8ced23dc7aba4ebd2536cf84b7a3af
 size 335922386

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a86ba0ad5079a830daa5d3a3f4b642aeb42993362465e2da1f3d7dd1419710ef
+oid sha256:5928b0834b3eb9beb3328868c0aba6d83caaeded442e7c47af1e1a696cbf497e
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6b28ee0addfe1fdb5c9554035d0aace280dc61b8aded9640b8ddc91a063558e8
+oid sha256:22e67cb47ff12ababb2db5752a0059510ef75c13ebd9ed242d32d1eb2ba8aa08
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.01254180602006689,
+  "epoch": 0.02508361204013378,
   "eval_steps": 8,
-  "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": 0.00027,
       "loss": 0.0,
       "step": 9
+    },
+    {
+      "epoch": 0.015050167224080268,
+      "grad_norm": NaN,
+      "learning_rate": 0.00029265847744427303,
+      "loss": 0.0,
+      "step": 12
+    },
+    {
+      "epoch": 0.018812709030100336,
+      "grad_norm": NaN,
+      "learning_rate": 0.00025606601717798207,
+      "loss": 0.0,
+      "step": 15
+    },
+    {
+      "epoch": 0.020066889632107024,
+      "eval_loss": NaN,
+      "eval_runtime": 37.3238,
+      "eval_samples_per_second": 6.752,
+      "eval_steps_per_second": 3.376,
+      "step": 16
+    },
+    {
+      "epoch": 0.0225752508361204,
+      "grad_norm": NaN,
+      "learning_rate": 0.0001963525491562421,
+      "loss": 0.0,
+      "step": 18
     }
   ],
   "logging_steps": 3,
@@ -63,7 +92,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 5285946895368192.0,
+  "total_flos": 1.0386421969846272e+16,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null