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:ea4a0be6179474b77c4f901f87ee098da86361f7badcaf455e50448c67538c7f
+oid sha256:c88d96aa2bcbe4fed0ade7d3e26f85ba08c04e3a700a6817eb795a7b38929543
 size 50338848

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cd7efbe1ad3f0abf54495b9a704a20e5c1675ebdb224841eaf7cf75fa2ceb0ab
+oid sha256:8a2a1f1198d8ab61503e0bb2e72a84b32f5c81a571dcd7023c2d864ec126ad8c
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a161e00ab25573dd553d2410ed15a310a4aaac2b7db73ad1b76c9c75404f8fda
+oid sha256:71128331774d3b754a22777700176c9e9d829274c21a3111675887d72431a159
 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": 1.2722229957580566,
-  "best_model_checkpoint": "./outputs/checkpoint-3500",
-  "epoch": 2.769398907103825,
+  "best_metric": 1.268664002418518,
+  "best_model_checkpoint": "./outputs/checkpoint-3900",
+  "epoch": 2.8422586520947175,
   "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": 47.127,
       "eval_steps_per_second": 5.897,
       "step": 3800
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 0.0002,
+      "loss": 1.2453,
+      "step": 3900
+    },
+    {
+      "epoch": 2.84,
+      "eval_loss": 1.268664002418518,
+      "eval_runtime": 133.1158,
+      "eval_samples_per_second": 47.132,
+      "eval_steps_per_second": 5.897,
+      "step": 3900
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 2.265239844086661e+17,
+  "total_flos": 2.3248377037320192e+17,
   "trial_name": null,
   "trial_params": null
 }