Training in progress, step 3500, checkpoint

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:108000d0b4263f572d7b4519d3eda641f3f2ec4c43612d5727fb831b0c385b50
+oid sha256:4e102f17b2794bbd14232bd02dbc262c6478ca331b767e8dd8c9a76617a24c4a
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c20c45ed5c1bc45fb23558e1e4bf3c316496698ffd6dc20f23fae5698c7a07f2
+oid sha256:db361cd39bb994cf19ea6520272539731dcbd9bb0ca941e54ec9edb35ce1012a
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47de56a8404ec27b2bdffd90a886b12812553c28703c689742e9edd53f5d058d
+oid sha256:31f628eccc9ba6d64b75ba39cccade5f78d5b2cc5ecff830614f67b411a1b5b3
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb3309524185001782da002e1efb7406e18966db2aee795c24a16a9ee18dd8d0
+oid sha256:31f466e63f2d702a9646f61d3cb0499d7a443ca833cfea51694a53eaa24cfd01
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.2855974435806274,
-  "best_model_checkpoint": "./outputs/checkpoint-3400",
-  "epoch": 2.4775956284153007,
+  "best_metric": 1.2743042707443237,
+  "best_model_checkpoint": "./outputs/checkpoint-3500",
+  "epoch": 2.550455373406193,
   "eval_steps": 100,
-  "global_step": 3400,
+  "global_step": 3500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -483,13 +483,27 @@
       "eval_samples_per_second": 46.981,
       "eval_steps_per_second": 5.878,
       "step": 3400
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 0.0002,
+      "loss": 1.2586,
+      "step": 3500
+    },
+    {
+      "epoch": 2.55,
+      "eval_loss": 1.2743042707443237,
+      "eval_runtime": 133.3944,
+      "eval_samples_per_second": 47.033,
+      "eval_steps_per_second": 5.885,
+      "step": 3500
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 2.026608076191744e+17,
+  "total_flos": 2.086233778867323e+17,
   "trial_name": null,
   "trial_params": null
 }