Training in progress, step 3900, checkpoint

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:23002b0d8f963d797fd8e72a8a22e3e253aa7554e3d7989108719386372f0e3f
+oid sha256:dde893a5b8c9903703f5fe18b4db6a74d80d1e605a1654c710070dd9d9ede78f
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b489bcf3c7281812031eb03e2499fb2a763e3bd45d8a3f35a2b924ff5c4f983f
+oid sha256:df8a418b0d49534a4c1e36fe0283b8c460e6713524b33bd3cf5f0b2bae83bf5e
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3ce0143f6362c76d1e380d7be8616206851e20defa9a0906aa8f6f5faf57b03b
+oid sha256:15ffa04f9b4c47e23e16441fa376bd921a73b836cf43b7e6e2b1582f10fd2d81
 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": 1.2429386377334595,
-  "best_model_checkpoint": "./outputs/checkpoint-3800",
-  "epoch": 2.769034608378871,
+  "best_metric": 1.2287747859954834,
+  "best_model_checkpoint": "./outputs/checkpoint-3900",
+  "epoch": 2.841894353369763,
   "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": 47.016,
       "eval_steps_per_second": 5.883,
       "step": 3800
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 0.0002,
+      "loss": 1.2171,
+      "step": 3900
+    },
+    {
+      "epoch": 2.84,
+      "eval_loss": 1.2287747859954834,
+      "eval_runtime": 133.4716,
+      "eval_samples_per_second": 47.006,
+      "eval_steps_per_second": 5.881,
+      "step": 3900
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 2.264898693695324e+17,
+  "total_flos": 2.3246378200834867e+17,
   "trial_name": null,
   "trial_params": null
 }