Training in progress, step 3400, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:72a9b3238f128bf1dbe68749e2830687ee771e8e6997bd21936fbd8c9e9d69b8
+oid sha256:93cbce12db1bcd458309268a7da42d4825128c957e0d1c699debbc51749fb41b
 size 19669752

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f80c684412745d5b933bb172431b91e4aa8432bebc668f5e5bb93fb386fb459
+oid sha256:9b10d35bb1284042f0db66531404020fd05cfb121db46ee069e129f290e14e34
 size 39356997

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c63d6d6384d612982216f57b7930ead02ff385875d3d4426f2a6519af4c21380
+oid sha256:77af6e432eee0998a8a3f6243344f174cbdce2efbbae495079d6a78a1eab9661
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ad85ddedbe158ea4b6f60faba10617b6fb17a69c778e6d7c0a8e4338dd95e31f
+oid sha256:cb3309524185001782da002e1efb7406e18966db2aee795c24a16a9ee18dd8d0
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.9004936218261719,
-  "best_model_checkpoint": "./outputs/checkpoint-3300",
-  "epoch": 4.4,
+  "best_metric": 0.8973463177680969,
+  "best_model_checkpoint": "./outputs/checkpoint-3400",
+  "epoch": 4.533333333333333,
   "eval_steps": 100,
-  "global_step": 3300,
+  "global_step": 3400,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -469,6 +469,20 @@
       "eval_samples_per_second": 2.483,
       "eval_steps_per_second": 0.311,
       "step": 3300
+    },
+    {
+      "epoch": 4.53,
+      "learning_rate": 0.0002,
+      "loss": 0.6887,
+      "step": 3400
+    },
+    {
+      "epoch": 4.53,
+      "eval_loss": 0.8973463177680969,
+      "eval_runtime": 777.6425,
+      "eval_samples_per_second": 2.482,
+      "eval_steps_per_second": 0.311,
+      "step": 3400
     }
   ],
   "logging_steps": 100,
@@ -476,7 +490,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 5,
   "save_steps": 100,
-  "total_flos": 8.547856629710438e+17,
+  "total_flos": 8.805891139634381e+17,
   "trial_name": null,
   "trial_params": null
 }