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:a78c9843fb911243642b3f5cbd5a60f30d0e28ad755a1f08b071cca2e2082d66
+oid sha256:cc2cb766a1c220e5c36ef30f09dada799887e7198a1373385752943aaecfa634
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:11317e8ede80c2379b0610ef6ef33da077503f5fd8d4940c0d81ea139b7e230c
+oid sha256:c63f850030f67885daf8a61b788ad50f93b35305e6823a1a9868b2d0bf6ceb82
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a84979b56dd7eaafef51eff439115d9fea01a7293cd2430bc52389be8cb5e3f0
+oid sha256:19f566442a994eb367f21419c663f9d4a1cd6caaa8e2461d98745e40c3e8edf1
 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": 2.0112404823303223,
-  "best_model_checkpoint": "./outputs/checkpoint-3800",
-  "epoch": 2.768670309653916,
+  "best_metric": 1.9985524415969849,
+  "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": 30.496,
       "eval_steps_per_second": 3.816,
       "step": 3800
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 0.0002,
+      "loss": 1.9544,
+      "step": 3900
+    },
+    {
+      "epoch": 2.84,
+      "eval_loss": 1.9985524415969849,
+      "eval_runtime": 205.5555,
+      "eval_samples_per_second": 30.522,
+      "eval_steps_per_second": 3.819,
+      "step": 3900
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.1085385650167808e+17,
+  "total_flos": 1.1378537348161536e+17,
   "trial_name": null,
   "trial_params": null
 }