Training in progress, step 4100, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1bf6e29139b0f24b399e976d09d4884b6235ffdaaa52ee8a0b7938baae37b428
+oid sha256:b4492da231f945853f6c83ca380b96f25222aa65080b3b146679cd0a751f8627
 size 31466288

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a7a42b3115d3199d3743de0e8e6b98305c990ec30c405bfdddcdad84bc32c6e1
+oid sha256:59596cd46764af0061bceb411c43f6515caf05658f5e0c0563785422ea1da3b2
 size 62950917

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a6367fbbdff331ec434b7f226419eafab1e9622cbf8e5630817798ff44dfdb8a
+oid sha256:cab967de45c663aa30264993bcda98120463fb94906abc05f58530bcb7e737be
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:593f43b0c83adb27a2db37a6418c2ef12a213bbc2a02f2dc881de6846a69a931
+oid sha256:f55ebe991e4838deda9ff3044672335e03697b65b3fe0753cca61998e974ee51
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.1375267505645752,
-  "best_model_checkpoint": "./outputs/checkpoint-4000",
-  "epoch": 2.9143897996357016,
+  "best_metric": 1.1210918426513672,
+  "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": 11.275,
       "eval_steps_per_second": 1.411,
       "step": 4000
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 0.0002,
+      "loss": 1.0693,
+      "step": 4100
+    },
+    {
+      "epoch": 2.99,
+      "eval_loss": 1.1210918426513672,
+      "eval_runtime": 556.5587,
+      "eval_samples_per_second": 11.273,
+      "eval_steps_per_second": 1.41,
+      "step": 4100
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.0412060628345815e+18,
+  "total_flos": 1.0672136421373379e+18,
   "trial_name": null,
   "trial_params": null
 }