Training in progress, step 3900, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7904de6f3a6c81434524363955086f270d6e1bc9602b5eb50481becd62e1c130
+oid sha256:b5bd53816dec1f54087c7a98a07ce3e939ca34b4f4b6d19c7d7deda012989e55
 size 50338848

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24015c4376f4fabd69ec52827faf5405a347f818b5fb5e0c4305578d2a137c59
+oid sha256:00177b501de8ddd7c362d6cf0908f8399182383eaec609ad095a19b45401d4a1
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c6ef7c5a570735e8b6e72739891f16a22a1d772533cbc85af1fbd69a57a8ec31
+oid sha256:34985e8b1e69a205583c6b9115293e324729ed0c191fbcce181660393717f45d
 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": 1.3516861200332642,
-  "best_model_checkpoint": "./outputs/checkpoint-3800",
-  "epoch": 2.769398907103825,
+  "best_metric": 1.340726613998413,
+  "best_model_checkpoint": "./outputs/checkpoint-3900",
+  "epoch": 2.8422586520947175,
   "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": 47.173,
       "eval_steps_per_second": 5.902,
       "step": 3800
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 0.0002,
+      "loss": 1.3085,
+      "step": 3900
+    },
+    {
+      "epoch": 2.84,
+      "eval_loss": 1.340726613998413,
+      "eval_runtime": 132.9904,
+      "eval_samples_per_second": 47.176,
+      "eval_steps_per_second": 5.903,
+      "step": 3900
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 2.2625907263034163e+17,
+  "total_flos": 2.3221217626762445e+17,
   "trial_name": null,
   "trial_params": null
 }