Training in progress, step 4100, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dbe9093b63aa6935b81b1765d36cbd9b201f606f89ba43b0a5fe0a64b928c06f
+oid sha256:7e7e48010a6d93d7059da6d6d6e44fcca3ca3d8cb060e088ea2ba498a88e1dc5
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8d2c45ca84abccd5d307d5e44aeb622c4682ac420c14f31c81c28fbfd966e5b4
+oid sha256:75f36595d4c3ac574a932088dc44683425a691dcfe6a8a827150f4a96c791cd6
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a9ef6c3858e7d84aa5f4da6d612efb7619c9f1a991da1e501295c2a72877c544
+oid sha256:ae8ecb1d7e34eae942eaf84679636cb08f247f9d84a52bdede4840acdcb4d717
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:593f43b0c83adb27a2db37a6418c2ef12a213bbc2a02f2dc881de6846a69a931
+oid sha256:eae9ca0d3721bf9dba058e72ab04c07410d5fe1aa8caf36cc99551db465db3a9
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 2.0607223510742188,
-  "best_model_checkpoint": "./outputs/checkpoint-4000",
-  "epoch": 2.9143897996357016,
+  "best_metric": 2.0496439933776855,
+  "best_model_checkpoint": "./outputs/checkpoint-4100",
+  "epoch": 2.987249544626594,
   "eval_steps": 100,
-  "global_step": 4000,
+  "global_step": 4100,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -567,13 +567,27 @@
       "eval_samples_per_second": 30.565,
       "eval_steps_per_second": 3.824,
       "step": 4000
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 0.0002,
+      "loss": 2.0315,
+      "step": 4100
+    },
+    {
+      "epoch": 2.99,
+      "eval_loss": 2.0496439933776855,
+      "eval_runtime": 205.7242,
+      "eval_samples_per_second": 30.497,
+      "eval_steps_per_second": 3.816,
+      "step": 4100
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.1387458434336768e+17,
+  "total_flos": 1.1671768877819904e+17,
   "trial_name": null,
   "trial_params": null
 }