Training in progress, step 3400, checkpoint

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:df6917260fe59b6b66718e1fe4f70bbdd83ed5e8ae6d9f2f85de0bfcfd13cbc2
+oid sha256:108000d0b4263f572d7b4519d3eda641f3f2ec4c43612d5727fb831b0c385b50
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:67792642355dde2606b315acacbd5edada1cd59b62a02a2c320e3e33fb08fbb8
+oid sha256:c20c45ed5c1bc45fb23558e1e4bf3c316496698ffd6dc20f23fae5698c7a07f2
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5e0da82eb6483d9c535630c26317d5a8b4f5e185e384513da7748e072891648c
+oid sha256:47de56a8404ec27b2bdffd90a886b12812553c28703c689742e9edd53f5d058d
 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": 1.2959859371185303,
-  "best_model_checkpoint": "./outputs/checkpoint-3300",
-  "epoch": 2.404735883424408,
+  "best_metric": 1.2855974435806274,
+  "best_model_checkpoint": "./outputs/checkpoint-3400",
+  "epoch": 2.4775956284153007,
   "eval_steps": 100,
-  "global_step": 3300,
+  "global_step": 3400,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -469,13 +469,27 @@
       "eval_samples_per_second": 46.988,
       "eval_steps_per_second": 5.879,
       "step": 3300
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 0.0002,
+      "loss": 1.2585,
+      "step": 3400
+    },
+    {
+      "epoch": 2.48,
+      "eval_loss": 1.2855974435806274,
+      "eval_runtime": 133.542,
+      "eval_samples_per_second": 46.981,
+      "eval_steps_per_second": 5.878,
+      "step": 3400
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.9670503691268096e+17,
+  "total_flos": 2.026608076191744e+17,
   "trial_name": null,
   "trial_params": null
 }