Training in progress, step 20, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bebec76956f7aaf741e594dc53a0b27608dea69de2c22e0d3b457654df9f17d2
+oid sha256:05b3a5dd23a8ba5febeaefbfdd7893eb2946c384ba9fc0ad0b8ca5b13934f9a5
 size 148047722

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dac7bede21f3322345b0b38171a42d331c8328d417c4f3993654026282712c53
+oid sha256:e489a61cf79a2a014b1066e7ba5a23873d72c453c37a5d97dc70ee7e933edb99
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8bb840dbc65c27ef8840855786e35bfea033749e5d7571cf601f6645aa29cc1e
+oid sha256:1d95f3fb1f9df4cdd5f470d1684a242bf1d1940d47b9622802f603a91ffa5bc0
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.13714285714285715,
+  "epoch": 0.18285714285714286,
   "eval_steps": 3,
-  "global_step": 15,
+  "global_step": 20,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -160,6 +160,49 @@
       "eval_samples_per_second": 23.227,
       "eval_steps_per_second": 11.738,
       "step": 15
+    },
+    {
+      "epoch": 0.1462857142857143,
+      "grad_norm": NaN,
+      "learning_rate": 0.00013090169943749476,
+      "loss": 0.0,
+      "step": 16
+    },
+    {
+      "epoch": 0.15542857142857142,
+      "grad_norm": NaN,
+      "learning_rate": 0.00011045284632676536,
+      "loss": 0.0,
+      "step": 17
+    },
+    {
+      "epoch": 0.16457142857142856,
+      "grad_norm": NaN,
+      "learning_rate": 8.954715367323468e-05,
+      "loss": 0.0,
+      "step": 18
+    },
+    {
+      "epoch": 0.16457142857142856,
+      "eval_loss": NaN,
+      "eval_runtime": 4.002,
+      "eval_samples_per_second": 23.239,
+      "eval_steps_per_second": 11.744,
+      "step": 18
+    },
+    {
+      "epoch": 0.1737142857142857,
+      "grad_norm": NaN,
+      "learning_rate": 6.909830056250527e-05,
+      "loss": 0.0,
+      "step": 19
+    },
+    {
+      "epoch": 0.18285714285714286,
+      "grad_norm": NaN,
+      "learning_rate": 5.000000000000002e-05,
+      "loss": 0.0,
+      "step": 20
     }
   ],
   "logging_steps": 1,
@@ -179,7 +222,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 3918806510469120.0,
+  "total_flos": 5225075347292160.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null