Training in progress, step 3800, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1301197687934bec03b68698dfbf84acb41e5b76ecdc0fa4cc92eabf1f95293c
+oid sha256:695a22655c615960c3ee5a50450d9ec443fa7537ed70aa4634d9b53e3ae5a15b
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e15b194caf1392dda3b777c7d20059bb8ec9002cf85096b9bc1c33d92cee84ae
+oid sha256:8373c507c84ecce524596691310da68cf70b800fcd593ffedcb588b9f4b3f43b
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9625051d2f19a872551cef813d2b9903c199d8e6d249df84e50aeca1b8dc0b8d
+oid sha256:a84979b56dd7eaafef51eff439115d9fea01a7293cd2430bc52389be8cb5e3f0
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c157d4508adfa804f177e78fb4986d5f3ffb3ec20ae255fb757aa13c0595fdb4
+oid sha256:3961d2f201ce99426ee113655da78d8214f6ef72fc082855daa3c565989888c2
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 2.0208475589752197,
-  "best_model_checkpoint": "./outputs/checkpoint-3700",
-  "epoch": 2.6958105646630237,
+  "best_metric": 2.012598752975464,
+  "best_model_checkpoint": "./outputs/checkpoint-3800",
+  "epoch": 2.768670309653916,
   "eval_steps": 100,
-  "global_step": 3700,
+  "global_step": 3800,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -525,13 +525,27 @@
       "eval_samples_per_second": 30.202,
       "eval_steps_per_second": 3.779,
       "step": 3700
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 0.0002,
+      "loss": 1.9686,
+      "step": 3800
+    },
+    {
+      "epoch": 2.77,
+      "eval_loss": 2.012598752975464,
+      "eval_runtime": 207.8079,
+      "eval_samples_per_second": 30.191,
+      "eval_steps_per_second": 3.778,
+      "step": 3800
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.0792743237255168e+17,
+  "total_flos": 1.1085385650167808e+17,
   "trial_name": null,
   "trial_params": null
 }