Training in progress, step 3600, checkpoint

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4e102f17b2794bbd14232bd02dbc262c6478ca331b767e8dd8c9a76617a24c4a
+oid sha256:644b0c1c88c33e2a022f98955644ff34267945c75a81b42c4d95bcc38ae8f8b1
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:db361cd39bb994cf19ea6520272539731dcbd9bb0ca941e54ec9edb35ce1012a
+oid sha256:87c0fcfeab37012143c7c79178aba2786e8c4f37ed51418d423c2577c67add5f
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31f628eccc9ba6d64b75ba39cccade5f78d5b2cc5ecff830614f67b411a1b5b3
+oid sha256:1bf557781e688f003515f4d67f7f143f7669c3d7270f049833847c92d4d21f18
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31f466e63f2d702a9646f61d3cb0499d7a443ca833cfea51694a53eaa24cfd01
+oid sha256:43e3cf8d56a3f083d00cc85544d76ada2f884a1018c8752332d96f2799911117
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.2743042707443237,
-  "best_model_checkpoint": "./outputs/checkpoint-3500",
-  "epoch": 2.550455373406193,
+  "best_metric": 1.2626945972442627,
+  "best_model_checkpoint": "./outputs/checkpoint-3600",
+  "epoch": 2.6233151183970858,
   "eval_steps": 100,
-  "global_step": 3500,
+  "global_step": 3600,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -497,13 +497,27 @@
       "eval_samples_per_second": 47.033,
       "eval_steps_per_second": 5.885,
       "step": 3500
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 0.0002,
+      "loss": 1.2466,
+      "step": 3600
+    },
+    {
+      "epoch": 2.62,
+      "eval_loss": 1.2626945972442627,
+      "eval_runtime": 133.3673,
+      "eval_samples_per_second": 47.043,
+      "eval_steps_per_second": 5.886,
+      "step": 3600
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 2.086233778867323e+17,
+  "total_flos": 2.1457325759420006e+17,
   "trial_name": null,
   "trial_params": null
 }