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:fe62918a4dffa4665f1d9717eddd64f5e60ba45038bb6ff732fdd8a3063355ac
+oid sha256:49f9868e4d1cd21d90596284b88521aacdf829dc439af2b6b38673ec9ca76abf
 size 50338848

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b63d8161ac2d59af0624fc5badc426f9233200135f1f2bea08b9fc3528678795
+oid sha256:8d638a7e1318513481edad79747bbd9d85854414852c1a384c0c1190d66d2abd
 size 100693001

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:966abaeea490518d524411c6c93b2cb01cabbd7b354c975c231fa3c86331e17e
+oid sha256:44f7a23b7a631b1fe28c32f80339cec68c940273b7a37ff68a8e64f1b9f09a97
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb592c268b6f8eab99358b5a12e18084bc178639195f5911973b566c7c32da81
+oid sha256:794a0a668693664bb45fb10418080981a1873b71841f3a55ed72c5aa6d455597
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.343408226966858,
-  "best_model_checkpoint": "./outputs/checkpoint-3800",
-  "epoch": 2.768670309653916,
+  "best_metric": 1.3290683031082153,
+  "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": 43.689,
       "eval_steps_per_second": 5.466,
       "step": 3800
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 0.0002,
+      "loss": 1.3154,
+      "step": 3900
+    },
+    {
+      "epoch": 2.84,
+      "eval_loss": 1.3290683031082153,
+      "eval_runtime": 143.5227,
+      "eval_samples_per_second": 43.714,
+      "eval_steps_per_second": 5.47,
+      "step": 3900
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 2.2624424255319245e+17,
+  "total_flos": 2.322136123818148e+17,
   "trial_name": null,
   "trial_params": null
 }