Training in progress, step 4, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3652acfa2f26613920bdfc4ed56f1d05eae59cfc8ebd69dfe4a6606196a2c36c
+oid sha256:2e2f25c5e8547502483de03a82d8261e785b99fb9c046b6e4be80d46f535fad1
 size 147781268

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:43cf72c01ad6d74e6ee1f250d40288ec8ea6e21d72b7a73107d00b2b9e152a5a
+oid sha256:64348d23d61878278ae4c95949a84a77e8b4d69045ab6dab8b3fed6a24d4fe6d
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:60f8f27a9d76e31d3721bb37f2f4c76e866b8dedf8d4315f175dac906a4a966d
+oid sha256:c7adb031dc938779259a07e684ea318126e92e52b40c256b93ef474d2ac57a5b
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.005273566249176005,
+  "epoch": 0.01054713249835201,
   "eval_steps": 1,
-  "global_step": 2,
+  "global_step": 4,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -37,6 +37,36 @@
       "eval_samples_per_second": 7.683,
       "eval_steps_per_second": 3.841,
       "step": 2
+    },
+    {
+      "epoch": 0.007910349373764008,
+      "grad_norm": NaN,
+      "learning_rate": 3e-05,
+      "loss": 0.0,
+      "step": 3
+    },
+    {
+      "epoch": 0.007910349373764008,
+      "eval_loss": NaN,
+      "eval_runtime": 20.932,
+      "eval_samples_per_second": 7.644,
+      "eval_steps_per_second": 3.822,
+      "step": 3
+    },
+    {
+      "epoch": 0.01054713249835201,
+      "grad_norm": NaN,
+      "learning_rate": 4e-05,
+      "loss": 0.0,
+      "step": 4
+    },
+    {
+      "epoch": 0.01054713249835201,
+      "eval_loss": NaN,
+      "eval_runtime": 20.9038,
+      "eval_samples_per_second": 7.654,
+      "eval_steps_per_second": 3.827,
+      "step": 4
     }
   ],
   "logging_steps": 1,
@@ -56,7 +86,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1299056311664640.0,
+  "total_flos": 2598112623329280.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null