Training in progress, step 20, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c4f3c993ae353778076f3aaef8e2bed7acfb869ee9591c62384726d972428592
+oid sha256:4da16b0e874343aca4c11d370c45917e6f162c63be579655f2d98600ad813541
 size 320194002

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:59c80006eccb20fcf4f3e5cb97069909cf631490e6e9d44183af55dc02be1f42
+oid sha256:dbadab122c5c0f7ffa8352df0bb2cf6ae56a1e473fbaee09e1a2facb6aab91e4
 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": 0.06543075245365322,
+  "epoch": 0.13086150490730644,
   "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": 0.07851690294438386,
+      "grad_norm": NaN,
+      "learning_rate": 0.00029265847744427303,
+      "loss": 0.0,
+      "step": 12
+    },
+    {
+      "epoch": 0.09814612868047982,
+      "grad_norm": NaN,
+      "learning_rate": 0.00025606601717798207,
+      "loss": 0.0,
+      "step": 15
+    },
+    {
+      "epoch": 0.10468920392584515,
+      "eval_loss": NaN,
+      "eval_runtime": 21.3125,
+      "eval_samples_per_second": 13.607,
+      "eval_steps_per_second": 4.551,
+      "step": 16
+    },
+    {
+      "epoch": 0.11777535441657579,
+      "grad_norm": NaN,
+      "learning_rate": 0.0001963525491562421,
+      "loss": 0.0,
+      "step": 18
     }
   ],
   "logging_steps": 3,
@@ -63,7 +92,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.4825479118782464e+16,
+  "total_flos": 2.977348285838131e+16,
   "train_batch_size": 3,
   "trial_name": null,
   "trial_params": null