Training in progress, step 3900, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:695a22655c615960c3ee5a50450d9ec443fa7537ed70aa4634d9b53e3ae5a15b
+oid sha256:cdfba2964ce961719ae533a5b3885e6b255d0bf36c19035654780c1e1b5997c9
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8373c507c84ecce524596691310da68cf70b800fcd593ffedcb588b9f4b3f43b
+oid sha256:ead8728468007a0027bcdab04f1d09712fe6a49caa3d4e7641bc47ca53ca7e32
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a84979b56dd7eaafef51eff439115d9fea01a7293cd2430bc52389be8cb5e3f0
+oid sha256:19f566442a994eb367f21419c663f9d4a1cd6caaa8e2461d98745e40c3e8edf1
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3961d2f201ce99426ee113655da78d8214f6ef72fc082855daa3c565989888c2
+oid sha256:a42424f90eb68391066979b79f87ab18e2fce2a0abc640b46cf1b3f4c12c720a
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 2.012598752975464,
-  "best_model_checkpoint": "./outputs/checkpoint-3800",
-  "epoch": 2.768670309653916,
+  "best_metric": 2.0015170574188232,
+  "best_model_checkpoint": "./outputs/checkpoint-3900",
+  "epoch": 2.841530054644809,
   "eval_steps": 100,
-  "global_step": 3800,
+  "global_step": 3900,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -539,13 +539,27 @@
       "eval_samples_per_second": 30.191,
       "eval_steps_per_second": 3.778,
       "step": 3800
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 0.0002,
+      "loss": 1.9585,
+      "step": 3900
+    },
+    {
+      "epoch": 2.84,
+      "eval_loss": 2.0015170574188232,
+      "eval_runtime": 207.5957,
+      "eval_samples_per_second": 30.222,
+      "eval_steps_per_second": 3.781,
+      "step": 3900
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.1085385650167808e+17,
+  "total_flos": 1.1378537348161536e+17,
   "trial_name": null,
   "trial_params": null
 }