Training in progress, step 3900, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c509f8375bf1720fd86acb95642efa1052933ea7d6e340c9165e3784e9183a42
+oid sha256:09af5650532bd8b482ed8a219ee62a8081101951ec735321e89017d31b176741
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:27ce730a655936b0fc464f41c7491bb6aabfef7e584d6d2a0e944550fd1a6a1d
+oid sha256:b09e364a0df2822ba617527aa0559e22bc3367696b395ef1b3e93fd9a7971bd2
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6b4ad38cbb21b65a6807ca7a900c58c750db7e99adcc03db559553fba9a08dd0
+oid sha256:b4d4fec55083981c7a8b41d24f9a0c06691f34476dffbd5f9f0bb89a650c7032
 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": 1.9890899658203125,
-  "best_model_checkpoint": "./outputs/checkpoint-3800",
-  "epoch": 2.768670309653916,
+  "best_metric": 1.9771522283554077,
+  "best_model_checkpoint": "./outputs/checkpoint-3900",
+  "epoch": 2.841530054644809,
   "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": 27.836,
       "eval_steps_per_second": 3.483,
       "step": 3800
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 0.0002,
+      "loss": 1.9371,
+      "step": 3900
+    },
+    {
+      "epoch": 2.84,
+      "eval_loss": 1.9771522283554077,
+      "eval_runtime": 216.1751,
+      "eval_samples_per_second": 29.023,
+      "eval_steps_per_second": 3.631,
+      "step": 3900
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.1157769634586624e+17,
+  "total_flos": 1.1452801265934336e+17,
   "trial_name": null,
   "trial_params": null
 }