Training in progress, step 3800, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:107c9414271dd8048634a9c774c400f0c6d03e9654793160ef3667b375df0445
+oid sha256:4e87bfed0468bb6b480b57505509387737e688bdf7ada10b2e31019104c8cf6c
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90c054c1cd1763a1f903411d046aa39df6be5b920e59a57933858da1fe965242
+oid sha256:728b359a0642d40385600e7570bec754fcf812fef54041885406721c33c21a44
 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": 1.9962486028671265,
-  "best_model_checkpoint": "./outputs/checkpoint-3700",
-  "epoch": 2.6958105646630237,
+  "best_metric": 1.9892688989639282,
+  "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": 11.636,
       "eval_steps_per_second": 1.456,
       "step": 3700
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 0.0002,
+      "loss": 1.9502,
+      "step": 3800
+    },
+    {
+      "epoch": 2.77,
+      "eval_loss": 1.9892688989639282,
+      "eval_runtime": 540.3139,
+      "eval_samples_per_second": 11.612,
+      "eval_steps_per_second": 1.453,
+      "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
 }