Training in progress, step 18, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9c75c6ce5e5fdff082e26ba3abb6d5c434cdb94d9336f50d89b918b5ef6770d0
+oid sha256:e201cceff065035a61d4d55c4c23f42b47fedcd200ebf54f7fa2f6a796766443
 size 1001863522

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4767faf7c126974119bfcc0ce8d9ad1b33912977bc9b7533647b8330927aa657
+oid sha256:108a6cca1849ac1771f7859cf719208aacd1fd0e71073c8a0081a91cdbddc28f
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:621519582414cbcc961b30cff113e8b1e10923f26e46ef8e0ad7c29a35554318
+oid sha256:d403ac4d7711325e0c8bec676128002b741c5058ffa34f8bfc8e7d7e4547c436
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.011759421828426769,
+  "epoch": 0.023518843656853538,
   "eval_steps": 9,
-  "global_step": 9,
+  "global_step": 18,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -44,6 +44,35 @@
       "eval_samples_per_second": 5.867,
       "eval_steps_per_second": 2.934,
       "step": 9
+    },
+    {
+      "epoch": 0.015679229104569026,
+      "grad_norm": NaN,
+      "learning_rate": 0.00012,
+      "loss": 0.0,
+      "step": 12
+    },
+    {
+      "epoch": 0.019599036380711282,
+      "grad_norm": NaN,
+      "learning_rate": 0.00015000000000000001,
+      "loss": 0.0,
+      "step": 15
+    },
+    {
+      "epoch": 0.023518843656853538,
+      "grad_norm": NaN,
+      "learning_rate": 0.00018,
+      "loss": 0.0,
+      "step": 18
+    },
+    {
+      "epoch": 0.023518843656853538,
+      "eval_loss": NaN,
+      "eval_runtime": 439.3537,
+      "eval_samples_per_second": 5.868,
+      "eval_steps_per_second": 2.934,
+      "step": 18
     }
   ],
   "logging_steps": 3,
@@ -63,7 +92,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 9.205525140996096e+16,
+  "total_flos": 1.8411050281992192e+17,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null