Training in progress, step 3800, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:09d4b91e1e9659278f7bba665d6e7114c775bcc0952d8abdb4b02198254f6920
+oid sha256:c509f8375bf1720fd86acb95642efa1052933ea7d6e340c9165e3784e9183a42
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f5140c747c623d47fd8a47dd5498f26aec184782d67d4562e667b045ba4f2c19
+oid sha256:27ce730a655936b0fc464f41c7491bb6aabfef7e584d6d2a0e944550fd1a6a1d
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d24dae49552cd627e5eaaed4e33ff7d215a6f396899031132e935859a4b208cd
+oid sha256:6b4ad38cbb21b65a6807ca7a900c58c750db7e99adcc03db559553fba9a08dd0
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c157d4508adfa804f177e78fb4986d5f3ffb3ec20ae255fb757aa13c0595fdb4
+oid sha256:3961d2f201ce99426ee113655da78d8214f6ef72fc082855daa3c565989888c2
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.994431972503662,
-  "best_model_checkpoint": "./outputs/checkpoint-3700",
-  "epoch": 2.6958105646630237,
+  "best_metric": 1.9890899658203125,
+  "best_model_checkpoint": "./outputs/checkpoint-3800",
+  "epoch": 2.768670309653916,
   "eval_steps": 100,
-  "global_step": 3700,
+  "global_step": 3800,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -525,13 +525,27 @@
       "eval_samples_per_second": 29.913,
       "eval_steps_per_second": 3.743,
       "step": 3700
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 0.0002,
+      "loss": 1.9505,
+      "step": 3800
+    },
+    {
+      "epoch": 2.77,
+      "eval_loss": 1.9890899658203125,
+      "eval_runtime": 225.3887,
+      "eval_samples_per_second": 27.836,
+      "eval_steps_per_second": 3.483,
+      "step": 3800
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.0863248925892608e+17,
+  "total_flos": 1.1157769634586624e+17,
   "trial_name": null,
   "trial_params": null
 }