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:ee4d0e362935b12690a61f0d478a5b98a5a153ffc4f2cd0ed5ff8988dac962b7
+oid sha256:6fa71eca77e4cb3a287115209eb353cea96380f015f1fa4a42b4b1a0444d0319
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5b4662e5629f5194b357369d4bc3c5c52910288f2f3d4ea85714127256f5db25
+oid sha256:90ac9dd2da9c12a914155ed6d4094efe90b23009d894129ab048c74ce924d0e6
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d1e78fca4663bc24e80d7e8f779d7b0c03678f0aafed4295ea283e96374bf926
+oid sha256:cff68ccfd4b7f7441d9d6f3e2f7132d2890a46c56e8f2f0a975fd95b0cd6b66c
 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": 2.086164951324463,
-  "best_model_checkpoint": "./outputs/checkpoint-3700",
-  "epoch": 2.6958105646630237,
+  "best_metric": 2.0767734050750732,
+  "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": 30.431,
       "eval_steps_per_second": 3.808,
       "step": 3700
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 0.0002,
+      "loss": 2.0398,
+      "step": 3800
+    },
+    {
+      "epoch": 2.77,
+      "eval_loss": 2.0767734050750732,
+      "eval_runtime": 205.3418,
+      "eval_samples_per_second": 30.554,
+      "eval_steps_per_second": 3.823,
+      "step": 3800
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.0533333313456128e+17,
+  "total_flos": 1.0818983291332608e+17,
   "trial_name": null,
   "trial_params": null
 }